Download: SAP Excellence

SAP Excellence

Series Editors: Professor Dr. Dr. h.c. mult. Peter Mertens Universität Erlangen-Nürnberg Dr. Peter Zencke SAP AG, Walldorf, Jörg Thomas Dickersbach

Characteristic Based Planning

with mySAP SCM™ Scenarios, Processes, and Functions With contributions by A. Forstreuter, C. Fuhlbrügge and T. John With 174 Figures and 3 Tables,

Dr. Jörg Thomas Dickersbach SAP AG Postfach 14 61

69185 Walldorf

Germany E-mail: email is hidden

SAP, SAP SEM, SAP SEM/BA, SAP BW, SAP Logo, R/2, R/3, BAPI, Management Cockpit, mySAP, mySAP.com as well as other products and services of SAP and the correspondent logos are trademarks or registered trademarks of SAP AG in Germany and in several other countries all over the world.All other products mentioned are trademarks or registered trademarks of their respective companies. EVA® is a registered trademark of Stern Stewart & Co. HTML and XML are trademarks or registered trademarks of W3C®, World Wide Web Consortium, Massachusetts Institute of Technology. Oros® and ABC Technologies® are registered trademarks of ABC Technologies, Inc. Microsoft®, WINDOWS®, NT®, Access®, Excel®, Word®, PowerPoint® and SQL Server® are registered trademarks of Microsoft Corporation. Powersim® is a registered trademark of Powersim Corporation. Cataloging-in-Publication Data Library of Congress Control Number: 2005926344

ISBN 3-540-25781-0 Springer Berlin Heidelberg New York

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is con- cerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, repro- duction on microfilm or in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the provisions of the German Copyright Law of September 9,1965, in its current version, and permission for use must always be obtained from Springer-Verlag. Violations are liable for prosecution under the German Copyright Law. Springer is a part of Springer Science+Business Media springeronline.com © Springer-Verlag Berlin Heidelberg 2005 Printed in Germany The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Cover design: Erich Kirchner Production: Helmut Petri Printing: Strauss Offsetdruck SPIN 11422655 Printed on acid-free paper – 42/3153 – 543210,

Preface

Characteristics are used in SAP as attributes, e.g. to specify the configura- tion of products or the properties of batches. In many industries – engi- neering, automotive, mill, pharmaceutical and foods to name the most typical – supply chain planning has to consider these characteristics. APO offers many different functionalities for planning with characteristics, where each of the functionalities has some prerequisites and incompatibili- ties. Within the design of an implementation there are multiple determi- nants for the system configuration, and it is very important to understand the interdependencies and limitations at an early stage of the project. This book offers help and advice for the basic design of the implementation by explaining • the processes and scenarios (process chains) for planning with characteristics, • the functionalities for planning with characteristics in APO includ- ing their prerequisites and incompatibilities and • the entities, dependencies and system configuration determinants for planning with characteristics in R/3 and APO in order to avoid the discovery towards the end of the implementation that some parts just do not work together – and this risk is much higher using characteristics because the interdependencies are much less obvious. We believe that especially with characteristic based planning (a newly introduced term to subsume the different functionalities for planning with characteristics) it is very important to understand the order flow in detail. Therefore we will focus whenever possible on the scenario and use a func- tionality oriented approach only for those functions which require an ex- tensive explanation or are used in multiple scenarios. The focus on the selected scenarios does not imply that these are the only possible ones. But with the understanding of these scenarios and the limitation of the functionalities it will be a lot easier to assess whether a specific design is somewhere near the trodden path or not and which in- compatibilities might arise. For the visualisation of the order flow we are using comparatively many screenshots because the appearance of the objects is different depending, VI Preface on the configuration, and for the practical implementation it is helpful to notice the difference whether the characteristics values of an order in APO are due to variant configuration, descriptive characteristics or batch selec- tion. This book is clearly not an introduction to R/3 and APO in general. Therefore we assume a fairly good understanding about the basic concepts of these systems – in particular SD on R/3-side and DP, PP/DS and ATP on APO-side. Even without a detailed understanding of all of these mod- ules it is possible to understand the basic messages of the book, but for im- plementation help other sources have to be found – e.g. Dickersbach 2004. References to the literature have been kept to a minimum, instead OSS notes are referenced. Since the focus of this book lies on the application of the APO and R/3 system and the processes and scenarios which can be modelled with these (and less on general advantages and disadvantages of certain processes), this seemed to me the more helpful way. This document is based on the releases SCM 4.1 and R/3 4.7. For earlier releases additional constraints apply. First of all I would like to thank Anton Forstreuter, Christian Fuhl- brügge and Thomas John for their extensive help – from the multiple hints about the correct configuration of the system up to the discussions about the ideas and purposes of the functions, processes and scenarios. Without their contribution this book would not have been possible. Many thanks as well to Stefan Elfner and Dr. Sven Eigemann for their help in the area of the batch selection and to Christoph Jerger, Veronika Schmid-Lutz and Dr. Frank Horlacher for their comments and corrections. Finally I would like to thank Tobias Götz for his generous support of this project. Jörg Thomas Dickersbach March 2005,

Contents

1 Motivation for Planning with Characteristics... 1 2 Characteristic Based Planning Overview ... 3 2.1 Process Overview... 3 2.2 Scenarios for Characteristic Based Planning... 5 2.3 System Configuration Determinants ... 10 2.4 Functions for Characteristic Based Planning in APO ... 13 2.5 Process Variants and their Usage in the Scenarios... 15 3 Characteristics and Classes... 19 3.1 Characteristics and Classes in R/3... 19 3.2 Transfer of Characteristics and Classes to APO ... 22 4 Configuration Scheme in APO... 25 5 Make-to-Order with Variant Configuration ... 27 5.1 Scenario Description ... 27 5.1.1 Process Chain ... 27 5.1.2 System Configuration Determinants ... 29 5.1.3 Configurable Material ... 30 5.1.4 Configuration Profile ... 30 5.2 Variant Configuration in the Sales Order... 31 5.3 Object Dependencies... 34 5.3.1 Structure and Usage of Object Dependencies... 34 5.3.2 Selection Conditions ... 38 5.3.3 Procedures ... 38 5.3.4 Procedures with Reference Characteristics... 39 5.3.5 Variant Functions ... 42 5.4 Multi-Level Configuration ... 44 5.5 Configuration via Material Class ... 46 5.6 Material Variants... 47 5.7 Configurable Material Variants... 50 5.8 Variant Tables ... 52, VIII Contents 6 Make-to-Order with VC and Demand Planning... 55 6.1 Make-to-Order with VC and Planning Overview ... 55 6.2 Characteristic Based Forecasting ... 56 6.3 Forecast Release and Forecast Consumption ... 60 6.4 Scenario Description for Planning in Inactive Version ... 65 6.4.1 Process Chain for Planning in Inactive Version ... 65 6.4.2 System Configuration Determinants ... 70 6.4.3 Transformation of the Dependent Demand ... 71 6.4.4 Forecast Consumption by the Dependent Demand ... 72 6.4.5 Component Logic ... 74 6.5 Scenario Description for Planning in Active Version ... 75 6.5.1 Process Chain for Planning in Active Version... 75 6.5.2 System Configuration Determinants ... 79 6.5.3 Forecast Check on Characteristic Level... 79 6.5.4 Allocation Check on Characteristic Level ... 80 6.5.5 Forecast Consumption on Characteristic Level ... 81 7 Sales from Stock with Characteristics... 83 7.1 Scenario Description ... 83 7.1.1 Process Chain ... 83 7.1.2 System Configuration Determinants ... 85 7.2 Batch Selection in the Sales Order ... 86 7.3 ATP with Characteristics ... 89 7.3.1 Characteristic View... 89 7.3.2 Rules-Based ATP with Characteristic Substitution ... 90 7.4 Delivery with Characteristics ... 93 8 Configure-to-Order with Propagation ... 95 8.1 Scenario Description ... 95 8.1.1 Process Chain ... 95 8.1.2 System Configuration Determinants ... 97 8.2 Characteristic Propagation within the PPM ... 99 8.3 Block Planning ... 102 8.3.1 Motivation for Block Definition ... 102 8.3.2 Block Definition in the Resource... 103 8.3.3 Activity Valuation... 106 8.3.4 Block Planning Functions ... 109 8.3.5 Assembly Planning Based on Block Definitions ... 109 8.4 Goods Receipt with Batch Characteristics ... 111, Contents IX 9 Planning with Shelf Life... 113 9.1 Scenario Description ... 113 9.1.1 Process Chain ... 113 9.1.2 System Configuration Determinants ... 114 9.1.3 Shelf Life Settings in the Material Master ... 115 9.2 Shelf Life Functionality... 116 9.3 Characteristics for Shelf Life ... 118 9.4 Customer Specific Shelf Life Requirements ... 120 9.5 Production Planning and Scheduling with Shelf Life ... 124 9.6 Goods Receipt with Shelf Life Characteristics ... 125 9.7 Shelf Life Alerts ... 126 10 Sales Order Oriented Planning ... 129 10.1 Scenario Description ... 129 10.1.1 Process Chain ... 129 10.1.2 System Configuration Determinants ... 131 10.2 Descriptive Characteristics... 131 10.3 Forecast Adjustment... 133 10.4 Planned Order Conversion with Conversion Rules... 135 11 Production Planning and Scheduling... 137 11.1 Production Planning with Characteristics ... 137 11.1 Production Planning with the VC-Configuration... 137 11.2 Production Planning with the CDP-Configuration ... 138 11.2 Detailed Scheduling with Characteristics... 140 11.2.1 Impact of Characteristics on Detailed Scheduling ... 140 11.2.2 Characteristics as Sorting Criteria for Heuristics ... 141 11.2.3 Characteristics for Set-Up Group Determination ... 144 11.3 Planned Order Integration ... 145 11.4 Display of Characteristics in the Product View ... 147 References ... 149 Abbreviations ... 151 Implementation Samples ... 153 Transactions... 157 Index ... 159, 1 Motivation for Planning with Characteristics Characteristics are used in SAP to specify and provide additional informa- tion to objects as materials, resources, batches or orders. From a planning point of view those characteristics are relevant which describe • the properties of a configurable product for an order (e.g. the en- gine and the colour of a car) and • the properties of a product batch (i.e. inventory that was produced under the same conditions and has the same properties). In R/3 the sales order-oriented configuration of a product lies in the area of the variant configuration, the batch specific product properties are covered with the batch management. In combination with APO both types of char- acteristic can be used for planning – and there are different ways how to use APO for demand planning, sales order fulfilment and production. To subsume the different possibilities of planning with characteristic in APO we introduce the term ‘characteristic based planning’ (CBP). • Variant Configuration In many industries there is a trend towards mass customisation (Knol- mayer/Mertens/Zeier 2002). The variant configuration helps to combine the oppositional requirements for large quantities in logistics and for in- creasing individualisation on the other hand. Due to increasing competition the number of material variants has been increasing recently (Eversheim 1996). The motivation to use characteristics to specify resp. to configure a product is to reduce the amount and effort for creating new master data for each combination of the characteristics. Though there are cases where it is possible to cover a few different configurations per product by using a dif- ferent material master for each configuration, in many cases the number of master data would explode and cause problems in performance, transpar- ency, interactive planning and master data management. The main advan- tage of using characteristics in this area is to reduce the complexity in or- der to keep the transparency within the supply chain and to avoid the other problems of huge master data quantities. For example it is usually easier to perform a forecast on attributes than for completely configured products., 2 1 Motivation for Planning with Characteristics Another aspect is the increase in the flexibility compared to the effort of creating a new set of master data (material master, BOM, routing, …) for each new characteristic combination. Typical industries for variant configuration are engineering and con- struction where often each machine is different in some aspects, automo- tive industry where most manufacturers support a configuration of the car, mill products (metal, paper & wood) where customers often have specific requirements regarding size and quality and even high tech and consumer products – e.g. for the configuration of PCs. There are cases where a few hundred characteristics are used per product. • Batch Management The batch characteristics on the other hand are used to describe the proper- ties of a concrete, existent product which are in most cases only known at the time of the goods receipt from the production. Examples for these properties are the shelf life resp. production date or the quality of the batch. The latter is of considerable significance in cases where the produc- tion process contains stochastic elements – i.e. the exact product quality of the individual batch is not predictable. Typical industries for these are once again the mill industries, chemical and pharma (which have often additional requirements regarding batch pureness and shelf life) and consumer products – most of all foods indus- tries. Usually either variant configuration or batches are used, but there are cases where both variant configuration and batches are required. An example for this are the mill industries as described in chapter 8., 2 Characteristic Based Planning Overview 2.1 Process Overview

Supply chain processes in general span the areas of demand planning,

(sales) order fulfilment, distribution, production and procurement. Col- laboration processes might be added both towards customers and towards suppliers. For characteristic based planning (CBP) however we focus on the demand planning, sales order fulfilment and production processes be- cause these are affected the most by planning with characteristics and have usually the biggest significance for the relevant industries (with the excep- tion of consumer goods). Another reason is that distribution and procure- ment support planning with characteristics only to a rather limited extent. Sales Demand Planning Sales Order Creation Sales Order Entry with Rqmt. Class Det. Forecasting Assignment of Characteristic Values Forecast Release Release ATP Check Forecast Forecast Production Consumption Consumption Production Planning Detailed Scheduling Production Execution Order Conversion Goods Receipt Delivery

Fig. 2.1. Processes and Process Steps for CBP

, 4 2 Characteristic Based Planning Overview Within demand planning, sales and production we define the processes and process steps that are shown in figure 2.1. This is a simplified view be- cause only those elements are shown which are relevant for characteristic based planning – e.g. pricing within the sales order creation process and production order confirmation within the production execution process are left out. Nevertheless it provides a framework for the processes which are considered in the following and their interdependencies. Naturally not all process steps are required for the different cases of supply chain planning with characteristics. Demand planning has the task to provide the production planning proc- ess with forecasts (also named planned independent requirements (PIRs)) to trigger production before the customer places the sales order. If the fin- ished product is configurable, production is only triggered for assembly groups (with the exception of material variants, see chapter 5). The de- mand for the assembly groups is either forecasted based on the history or calculated from the demand for the finished product. In the latter case de- mand planning has to be performed on characteristic level. With the fore- cast release the demand plan becomes relevant for production planning and the forecast consumption is triggered (the forecast is consumed by sales orders). Sales order creation starts with the entry of the sales item and the (back- ground) determination of the requirements class. The requirements class controls whether make-to-order (usual for variant configuration) or make- to-stock is used. As the next step the characteristic values are assigned to the sales order, either as a configuration (usually interactively) or via batch selection (usually without user interaction). The requested sales item is checked for availability (ATP, available-to-promise) after the characteris- tic values have been assigned to the sales order. Again a consumption of the forecast is performed. Based on the independent demand of the sales orders and the forecast production planning creates planned orders to cover the demand. To allow the production to start, the planned order must be converted to a production order, which is confirmed after execution and the goods receipt is posted. When the requested product is on stock, the de- livery is created and optionally transportation planning towards the cus- tomer is performed. The production starts with production planning. The result of the pro- duction planning process are planned orders to cover the requirements. The planned orders are created with the assumption of infinite capacity and component availability. In the subsequent step a feasible plan is created involving a finite scheduling and sequencing of the orders. Experience shows that even with the use of an APS system the two-step approach of an infinite production planning first and a subsequent finite scheduling in a, 2.2 Scenarios for Characteristic Based Planning 5 separate step is more adequate. Note 551124 provides additional informa- tion on this topic. Depending on the business, before or after the schedul- ing the planned orders are converted into production orders, production is executed, the production orders are confirmed and goods receipt is posted. Depending on the scenario the processes might differ significantly. By the scenario we understand the context and the process chain of the com- pany. Some common and typical scenarios are introduced in the following chapter. The quotation process is not in scope since it is not planning relevant (to provide the information regarding the price and the availability the same functions are used as for the sales order). The impact of characteristics on pricing is not described either since it is entirely within R/3, and APO does not consider prices. Neither in the scope of this book is the financial corpo- rate planning, since the focus is entirely on the SCM related processes. 2.2 Scenarios for Characteristic Based Planning SCM processes usually have a very wide variance, and the set of functions which are available in APO are combined in many different ways. Many functionalities support characteristics to a certain extent, some functional- ities require the use of characteristics and some functionalities do not sup- port characteristics at all. To make things more complicated, the character- istics are embedded in different technical configurations – the classes with their class type in R/3 and the configuration scheme in APO – which bear additional restrictions. Therefore the dependencies between the individual functions and their embedding into the context of the process chain – the scenario – is of crucial importance in order to avoid infeasible implementa- tions. • Scenarios Under a scenario we understand a typical case for supply chain planning. The scenario is described by a chain of processes which covers either the whole area of the supply chain processes – demand planning, sales and production or at least a significant and typical part of it. Not all of the se- lected scenarios contain demand planning or production. Each of the described scenarios is in some industries more common than in others. We have selected seven typical scenarios for characteristic based planning, but these are of course not all possible scenarios. Depending on the specific company requirements it might be appropriate to combine the functionality in a different way to create a new scenario. This combination, 6 2 Characteristic Based Planning Overview of functionalities is more complex in the area of characteristics planning than with normal SCM and bears the risk of running into a dead end be- cause of incompatibilities which are not obvious.

Based on the seven selected scenarios it is possible to explain the func-

tionality for characteristic based planning. Most parts of the book relate to one of the seven scenarios, except chapter 3 and 4 about the general prop- erties of characteristics, classes and the configuration scheme and chap- ter 11 about production planning with characteristics. The reason for the latter is that production planning with characteristics is used similarly throughout most of the scenarios. In the following we will provide an overview about these scenarios. • Scenario Overview

Figure 2.2 provides an overview about the seven selected scenarios regard-

ing their main features in the areas of demand planning, sales and produc- tion. Scenario Demand Planning Sales Production Production Planning with MTO with Variant Configuration N.A. Capable-to-Promise Characteristics MTO with Variant Configuration Characteristic Based Multi-Level ATP or Production Planning with and Planning in Inactive Version Forecasting Capable-to-Promise Characteristics MTO with Variant Configuration Characteristic Based Forecast Check and/or Production Planning with and Planning in Active Version Forecasting Allocation Check Characteristics Sales from Stock with No Impact ATP with No Impact Characteristics Characteristics PP with Characteristics, Configure-to-Order with Propagation N.A. Capable-to-Promise Block Planning (optional) No Impact No Impact Production Planning withPlanning with Shelf Life Shelf Life Sales Order Oriented Planning Demand Planning ATP Production Planning with Order Conversion Rules Fig. 2.2. Scenario Overview

For the scenario ‘configure-to-order with propagation’ the option exist to

use the block definitions to generate forecasts for the assembly groups (if the assembly groups do not require characteristics for planning). Block planning might be used in other scenarios as well. In the following we de- scribe these scenarios in short. • Make-to-Order with Variant Configuration

The make-to-order scenario with variant configuration is a pure make-to-

order scenario which implies that the finished product is not produced, 2.2 Scenarios for Characteristic Based Planning 7 unless there is a sales order for it. In the sales order the configuration of the product – i.e. the characteristic value assignment – is performed per sales order item. Based on the configuration of the sales order the depend- ent demand for the assembly groups and the capacity requirements for the production are determined via object dependencies. The configuration might be only for the finished product or for assem- bly groups as well (multi-level configuration). A special case is the use of material variants which refer to a fix configuration but are not configurable themselves. Since we are in a make-to-order (MTO) environment the ATP check in the sales order uses either • capable-to-promise (CTP) to create planned orders for the product considering the production capacity and optionally the component availability as well, • multi-level ATP to create planned orders for the product checking whether the components are available or the • checking horizon if neither the production capacity nor the com- ponent availability are real bottlenecks. Examples for this scenario are often found in the engineering industry. • Make-to-Order with Variant Configuration and Planning Make-to-order with variant configuration and demand planning has the ad- vantage that it is possible to plan the demand for the assembly groups based on the demand for the finished product: The required functionality to perform demand planning taking the characteristic values into account is characteristic based forecasting (CBF). This is performed in two different ways, and for each way we define an own scenario. Based on the demand for the finished product the dependent demand is calculated either in an inactive planning version or in the plan- ning segment of the active version. If the dependent demand is calculated in an inactive planning version, it is transformed into a forecast for the as- sembly group. For this scenario either the multi-level ATP or the CTP check are the best options. For the scenario with planning in the active ver- sion the forecast check is appropriate. • Sales from Stock with Characteristics Products like electrolytic capacitors, steel assembly groups or pharmaceu- tical products are produced with a make-to-stock strategy but have proper- ties which do not satisfy all customers. These properties might be con- nected with the expiry date or due to stochastic production conditions and are stored as batch characteristics. In this case the characteristics of the product have to be considered in the ATP check as well. In this scenario no, 8 2 Characteristic Based Planning Overview planning is performed on characteristic level. Another restriction for this scenario is that the ATP check is limited to batches – i.e. no planned re- ceipts are used for confirmation. • Configure-to-Order with Propagation Configure-to-order with propagation is a very common scenario for the mill industries. This scenario is again a make-to-order scenario, though sometimes without the use of a make-to-stock requirements class. The pe- culiarity in this process is that assembly groups and key components (steel coils, paper reels, ...) have batch characteristics which are important for planning. Since the mill industries have a divergent material flow, many different finished products will have dependent demands for the key com- ponents. These dependent demands might have different requirements for the characteristic values of the batches. Additionally these key components have usually a comparatively long lead time. One main objective in this scenario is therefore usually to check whether the right amount of the key component with the required properties is available. These properties are recorded as the batch valuation. Differing from the variant configuration not only the finished product is configurable but many assembly levels and the key component have batch characteristics as well. For the finished product both class types for variant configuration and for batches are required. The sales orders for the finished product has usually a make-to-order strategy, but the assembly groups and the key component are produced as make-to-stock. To create production orders with a valuation in a make-to- stock segment, the IS Mill solution is required on R/3-side. Typically the ATP check for the sales order uses the CTP functionality. If the production of the basic raw material is planned as well or significant set-up is required for the assembly group, block planning is often used. Another feature of this scenario is that the characteristics requirements for the batch selection are usually not discrete values but ranges of values. • Planning with Shelf Life In some industries as food or pharma the shelf life of the products needs to be considered already during planning. Typically these products are planned as make-to-stock, and the relevant products – often the finished product – have shelf life specific batch characteristics. For the planning in APO the shelf life uses separate functions, and the product is not relevant for characteristics planning. Characteristics are only required in combination with APO to transfer the shelf life of the batches., 2.2 Scenarios for Characteristic Based Planning 9 • Sales Order Oriented Planning In many cases neither a pure make-to-stock nor a pure make-to-order strat- egy is applied for the business. Sales order oriented planning allows to combine advantages of the make-to-stock production as the use of lot-sizes and the consideration of inventories with keeping a reference to the cus- tomer: Forecast is consumed on customer resp. customer group level and planned orders are only converted if they are pegged to sales orders. • Industry Focus of the Scenarios Each of the described scenarios has a main focus on one or more indus- tries. Figure 2.3 shows the industry relevance of these scenarios: High Tech/ Engineering Automotive Consumer Mill Chemicals & Foods Electronics Pharma Scenario MTO with Variant Configuration MTO with Variant Configuration and Planning in Inactive Version MTO with Variant Configuration and Planning in Active Version Sales from Stock with Characteristics Configure-to-Order with Propagation Planning with Shelf Life Sales Order Oriented Planning Fig. 2.3. Industry Relevance of the Scenarios In engineering – especially for special purpose machines – as well as in the automotive industry make-to-order with variant configuration is used to large extent. Sometimes make-to-order with variant configuration is used for the consumer goods and the high tech industries as well in more tech- nical consumer goods like PCs. The scenario ‘configure-to-order with characteristic propagation’ is mostly used in the mill industries, the make- to-order with variant configuration and planning in the active version of- fers for the mill industries additionally the possibility to plan for the as- sembly groups. The scenario ‘sales from stock with characteristics’ is also interesting for the mill industries, but less for processes with production than for the sales of unfinished goods. Other industries for this scenario are industries with perishable products as pharma or industries with stochastic production processes like the high tech industry. The consideration of, 10 2 Characteristic Based Planning Overview shelf life already at the planning stage is most important for foods indus- tries but often used for pharmaceutical companies as well. Sales order ori- ented planning finally might be used by any industry which is neither a pure make-to-stock nor a pure make-to-order production. 2.3 System Configuration Determinants There are some basic settings for the configuration of the scenarios which are rather fundamental in the sense that it is difficult to change them with- out major re-design or re-implementation. These four determinants are • the requirements strategy: Here only the account assignment is of importance, i.e. whether a make-to-stock or a make-to-order strategy is applied. • the class type: Characteristics are combined in classes. The classes have class types which limit their usage. • the configuration scheme: in APO two different configuration schemes exist, the VC (variant configuration)-configuration scheme and the CDP (characteristic dependent planning)- configuration scheme. • the master data object: In APO 4.1 there are two alternative mas- ter data objects for production planning, the production process model (PPM) and the production data structure (PDS). For these determinants a short explanation is given in the following. • Requirements Strategy One main differentiator from a business point of view is the production strategy (make-to-order vs. make-to-stock). The decision whether make-to- order or make-to-stock is used in combination has implications not only towards SCM. Other affected processes are billing, warehousing and de- livery. Usually make-to-order is used to keep the link between the customer and the configured or valuated product via customer order specific plan- ning segments and assignments. If configured or valuated products are u- sed within a make-to-stock scenario, it has to be ensured that the topics mentioned above are either no issues for the specific case or are taken care of in a different way. Technically the decision whether a scenario is make- to-stock or make-to-order is based on the account assignment in the re- quirements class as defined in the R/3-customising, figure 2.4., 2.3 System Configuration Determinants 11 Fig. 2.4. Requirements Class For a make-to-stock scenario the account assignment category is usually blank. The make-to-order strategy is common for the use of variant con- figuration and necessary in combination with the VC-configuration scheme. Make-to-stock is used for the sales from stock scenario and the shelf life scenario. • Class Type The class type is a technical setting in the class which controls its usage for certain functions and processes. The class type is described in more detail in chapter 3. • Configuration Scheme The configuration scheme is a technical setting in APO which controls the modelling of the characteristics for planning. In short the VC-configu- ration scheme allows a configuration only for sales orders, and the CDP- configuration scheme allows a configuration for other objects like forecast, planned orders, production orders and batches as well. The significance of the configuration scheme is described in detail in chapter 4. The choice of the configuration scheme is made on client level. • Master Data The PPM is the master data which was available from the very beginning of APO but will not be developed any further due to some structural limi- tations (mainly the lack of engineering change management). The alterna- tive to the PPM is the PDS. The functionality of the PDS was increased for APO 4.1. In the releases APO 3.1 and 4.0 the PDS was called run-time ob- ject (RTO). The SAP recommendation is to use PDS for new implementa- tions whenever possible., 12 2 Characteristic Based Planning Overview • System Configuration Determinants for the Selected Scenarios

Figure 2.5 provides an overview about the configuration determinants for

the selected scenarios: Strategy Class Type Config. Scheme Master Data MTO MTS 300 023 VC CDP PDS PPM MTO with Variant Configuration x - * x x (x) x - MTO with Variant Configuration * and Planning in Inactive Version x - x x (x) x - MTO with Variant Configuration and Planning in Active Version x (-x-) x - x x - Sales from Stock with Characteristics - x x - xxxConfigure-to-Order Requires with Propagation IS Mill! x (x) xxx(x) x Planning with Shelf Life - ** x x -x- xxxSales Order Oriented Planning (-x-) x x Supported / Required (x) Possible but Unusual Indifferent / Not Required - Not Compatible * Possible in Combination with the CDP-Configuration Scheme but Not Recommended ** Technically Possible but Not Compatible from a Business Point of View Fig. 2.5. Configuration Determinants for the Selected Scenarios

Though it is technically possible to perform the configuration of the sales

order in combination with a make-to-stock strategy and the VC- configuration scheme, the correct assignment of the receipts to the de- mands is not ensured any more. Variant configuration as a functionality might be used in the combination of CDP-configuration relevance and macros with the PPM as well, but our definition of the variant configura- tion scenario includes the transfer of object dependencies from R/3. This is not possible with the PPM.

Shelf life planning does not require any configuration on product level

and is therefore independent of the configuration scheme. It is however not possible to combine the shelf life planning functionality with a VC- configuration relevance (e.g. resulting from variant configuration).

As figure 2.5 shows there are quite some incompatibilities. If the

switches for the implementation are set wrongly – especially the configu- ration scheme in APO – this will result in significant additional work to correct this., 2.4 Functions for Characteristic Based Planning in APO 13 2.4 Functions for Characteristic Based Planning in APO

Within the five APO modules DP, SNP, PP/DS, ATP and TP/VS many

functions exist which support characteristics or do not support characteris- tics. Some of these are even incompatible with characteristics – i.e. if char- acteristics are used, the function does not work – others merely ignore the characteristics. Then there are some functions which require characteris- tics. Additionally there are some cross-module functions for which the same applies. Figure 2.6 provides an overview about the most common functions and how they relate to planning with characteristics. APO-DP APO-ATP APO-SNP APO-PP/DS APO-TP/VS Production Transport Forecasting Product Check SNP Heuristics Shelf Life Planning Planning ATP Check with Scheduling Object CBF SNP Optimiser Characteristics Heuristic Dependencies APO Forecast Characteristic CTM with SNP Interchange- Evaluation PP/DS Optimiser Continuous I/O Consumption Master Data ability (Pegging) Backorder Release to SNP Deployment Purchasing Subcontracting Alert Monitor Processing Statistical Transport Load Rapid Planning Sequence Supply Chain Allocation Check Forecast Builder Matrix Dependent Set-Up Engineer Promotion Vendor Managed CTM with PP/DS Characteristic Forecast Check Planning Inventory Master Data Propagation Incompatibility with Ignores Capable-to- Block Planning Supports Promise Requires Characteristics Multi-Level ATP Fig. 2.6. Overview about APO Functions and CBP

Most of the CBP relevant functionality is within the module PP/DS, but DP and ATP do offer some functions as well. SNP does not support CBP

with a small exception for the SNP optimiser and shelf life. TP/VS does not support characteristics at all nor does it have any relevance for its planning tasks.

CTM does neither support the VC- nor CDP-configuration scheme – if

the PDS or PPM contains object dependencies resp. characteristic propaga- tion, the CTM run terminates with an error. An exception is the use of ma- terial variants for CTM with PP/DS master data, where the object depend- encies are exploded for the PDS before their integration with APO., 14 2 Characteristic Based Planning Overview Descriptive characteristics on the other hand can be used for demand pri- oritisation. The ATP check with characteristics is only possible with the CDP- configuration scheme. Multi-level ATP and the rapid planning matrix are only compatible for the VC-configuration scheme. With ‘characteristic propagation’ we understand the function of the PPM to propagate a valua- tion of the output product to the input product (though the propagation is possible with the PDS as well, with this function we refer only to the PPM). Block planning does not necessarily require an integration of char- acteristics from R/3 to APO but can be used with a simple classification of the routing as well. Sequence dependent set-up is independent of any char- acteristics values, though reference characteristics can be applied to set the set-up group depending on characteristic values as explained in chapter 11. • System Configuration Determinants for the Planning Functions Figure 2.7 provides an overview about the prerequisites for the functional- ities that require characteristics. Strategy Class Type Config. Scheme Master Data MTO MTS 300 023 VC CDP PDS PPM Characteristic Based Forecasting x (-x-) x (-x-) xxx- Multi-Level ATP x -x- x x (-x-) x - ATP on Characteristic Level -x- -x- (x) x - x x - Block Planning x - (x) x x (x) x -x- Integration of Object Dependencies x (-x-) x x (x) x - Characteristic Propagation in the PPM x - x (x) -x- (x) - -x- Shelf Life (x) -x- x - (x) x - Rapid Planning Matrix (x) - (x) x -x- (-x-) x - x Supported / Required (x) Possible but Unusual Indifferent / Not Required - Not Compatible Fig. 2.7. System Configuration Determinants for Functions For the mill industries it might be desired to perform ATP on characteristic level including planned receipts (it is only with IS Mill possible to inte- grate valuated planned receipts in the make-to-stock segment with R/3). In this case the class type 300 is necessary to valuate the receipts. Note that rules based ATP with characteristic substitution is not released for variant configuration (i.e. it is only released if just batches are checked – as de- scribed in the scenario ‘sales from stock with characteristics’). Block plan- ning can be performed with a fix classification of the activities. In this case no integration of classes is required. In most cases configuration is used however. For the mill industries sometimes it is required to use variant, 2.5 Process Variants and their Usage in the Scenarios 15 configuration in the make-to-stock segment with the CDP-configuration scheme. This is rather a special case. The shelf life used to require the CDP-configuration scheme, but with APO 4.1 it is possible to use shelf life with the VC-configuration scheme as well. The product itself does not have any relevance for characteristic based planning anyhow – neither for the VC- nor for the CDP-configuration. 2.5 Process Variants and their Usage in the Scenarios Within these scenarios the processes resp. process steps are used with dif- ferent peculiarities. The following figures show the processes resp. the process step variants and in which scenario they are used. Some processes resp. process steps are used in more than one scenario, therefore it is indi- cated where the process resp. the process step is described. Some processes resp. process steps are ‘standard’ in the sense that they are not affected by the use of characteristics. These are not described in any detail. The proc- esses for production are described in a separate chapter because they are used in different scenarios without any scenario preference. • Demand Planning Process Step Variants The main processes or process steps for demand planning are the forecast- ing, the release of the forecast and the consumption of the forecast by the sales order. A pure make-to-stock production without forecast consump- tion is unusual if characteristics are used throughout the planning proc- esses, but possible in the scenarios ‘sales from stock with characteristics’ and ‘planning with shelf life’. Figure 2.8 shows the different process steps variants., 16 2 Characteristic Based Planning Overview Process / Process / Process Step Variant Process Step Forecasting Forecasting Characteristic Based Forecasting Release on Product and Location Level Release w/o Consumption Group to Inactive Version Release Release with Consumption Group to Planning Segment Release with Consumption Group to Stock Segment Transformation of Dependent Demand to Forecast Forecast Consumption Forecast Forecast Consumption with Descriptive Characteristics Consumption Forecast Consumption on Characteristic Level Forecast Consumption for Assembly Groups Process / Process Step is Used in and Described for the Scenario Process / Process Step is Used in the Scenario Fig. 2.8. Process Step Variants for Demand Planning

For forecasting the major change is the use of characteristic based fore-

casting (CBF) which allows to plan on the level of configuration- characteristic values. From the handling point of view there is no differ- ence to the standard demand planning but there are a lot of changes for the configuration. The release of the forecast can be different using CBP in the planning process if the forecast needs to contain additional information about the characteristic values. For this purpose the entity ‘consumption group’ is introduced. Depending on the use of the consumption group and the target of the release – the make-to-stock segment, the planning seg- ment or the inactive version – the process step varies. The forecast con- sumption finally depends whether the forecast is consumed on characteris- tic level and whether the consumption takes place for the finished product or for the assembly group. Note that the scenarios ‘make-to-order with variant configuration’ and ‘configure-to-order with propagation’ do not re- quire any demand planning. • Sales Process Step Variants

One main differentiator for the supply chain planning is the requirements

strategy that is used for the sales order creation – whether make-to-order is used or make-to-stock. The other process steps for the sales order entry are the assignment of the characteristic values and the ATP check. For the assignment of the characteristic values the sales order configuration or the batch selection functionality or both are used. MTO with VC MTO with VC & Planning in Inactive Version MTO with VC & Planning in Active Version Sales from Stock With Characteristics Configure-to-Order with Propagation Planning with Shelf Life Sales Order Oriented Planning, 2.5 Process Variants and their Usage in the Scenarios 17 Process / Process / Process Step Variant Process Step Sales Order for Make-to-Stock Sales Order Creation Sales Order for Make-to-Order Assignment of Sales Order Configuration Characteristic Values Batch Selection Product Check / Allocation Check Multi-Level ATP Capable-to-Promise ATP Check Forecast Check on Characteristic Level Allocation Check on Characteristic Level Rules-Based ATP Check on Characteristic Level Process / Process Step is Used in and Described for the Scenario Process / Process Step is Used in the Scenario Process / Process Step is Possible for the Scenario Fig. 2.9. Process Step Variants for Sales

For the ATP check almost the whole range of functionalities is used for

the different scenarios – though only a few alternatives are recommended per scenario. Figure 2.9 shows the process variants for sales. Depending on the representation of the forecast (see process step ‘forecast release’) the forecast check considers the characteristic values as well. The ATP check on characteristic level does not necessarily require rules-based ATP, but in most cases the option for a characteristic substitution is desired. • Production Process Step Variants

The main processes for production are production planning (i.e. the proc-

ess of creating the planned orders), detailed scheduling, production execu- tion with the conversion and processing of the production order and goods receipt. Depending on the configuration relevance and the master data (PDS or PPM), different possibilities exist for production planning to use either object dependencies or characteristic propagation in the PPM to con- trol the selection of the components and the capacity requirements of the planned orders. If block planning is used, the planned orders are scheduled into the matching blocks. Detailed scheduling offers the two functions of scheduling heuristics or the PP/DS optimiser. The planned order conver- sion is done automatically with the transfer to R/3 if the CDP- configuration relevance is used. With the use of conversion rules it can be ensured that planned orders are only converted if sales orders exist for VariaMntTCOownfiitghuVraCtion MTVOCw&ithPlVanCn&ingPliannning IinnaIcntaivcetivVeeVrseiorsnion MTVOCw&ithPlVanCn&ingPliannning AincAtivcetivVeeVrseiorsnion SSaalelessfrforommSStotocckk WWitihthCChhaararaccteterirsistitcicss Configure-to-Order MMillill with Propagation PSlaShnhenelifnlfLgLifiwefeiPthPlalSanhnneninlifngLgife SSaalelessOOrdrdeerr OOrireiennteteddPPlalannnniningg, 18 2 Characteristic Based Planning Overview them. Goods receipt finally may require a batch valuation if batch charac- teristics are used. Figure 2.10 shows the process variants for production. Process / Process / Process Step Variant Process Step Production Planning (without Characteristics) Production Planning with VC-Characteristics Production Planning Production Planning with CDP-Characteristics Block Planning Production Planning with Shelf Life Scheduling Heuristics Detailed Scheduling PP/DS Optimisation Order Conversion Order Conversion Order Conversion with Conversion Rule Goods Receipt Goods Receipt Goods Receipt with Batch Characteristics Process / Process Step is Used in and Described for the Scenario Process / Process Step is Used in the Scenario Process / Process Step is Possible for the Scenario

Fig. 2.10. Process Step Variants for Production Specific for CBP is the consideration of object dependencies and charac-

teristic propagation in production planning. Block planning covers differ- ent aspects which are all CBP relevant and has an impact on production planning as well. VVaarriiaaMnnttTCCOoownnffiiitgghuuVrraaCttiioonn MTVVOCCw&&ithPPllVaannCnn&iinnggPliiannnning IIinnnaaIccnttaiivvceetivVVeeeVrrsseiioorsnnion MTVVOCCw&&ithPPllVaannCnn&iinnggPliiannnning AAinccAttiivvceetivVVeeeVrrsseiioorsnnion SSSaaalleelesssffrrfoorommmSSSttootoccckkk WWiittihhthCCChhhaaarraaraccctteeterriirssisttiitccicsss Configure-to-Order MMMiillllill with Propagation PSSlaShhnheenelliffnlfLLgLiiffiweefeiPPthPllaalSannhnnneniinnlifnggLgife SSSaaalleelesssOOOrrddrdeeerr r OOrriireeiennntteetedddPPPllaalannnnnniinninggg, 3 Characteristics and Classes 3.1 Characteristics and Classes in R/3 Characteristics allow to attach attributes to a material, a routing, a work center, a batch or an order. These attributes are used for different purposes, and only some of them are relevant for characteristic based planning (CBP). For example the usage of material characteristics to search for ma- terials by their technical properties is rather targeted for production engi- neering, and the usage of work center characteristics to define resource al- ternatives for APO is not relevant for characteristics based planning either. For CBP we focus on characteristics that are used to describe the configu- ration of a product in the sales order and characteristics that describe the planning relevant properties of batches. Note 714929 provides additional information about the properties of characteristics in combination with planning in APO. • Overview Characteristics must be assigned to a class in order to use them in any SAP application. The usage of the characteristics is defined by the class type, though the same characteristic can be assigned to different classes of dif- ferent class types. Figure 3.1 sketches this interdependency. CharacteristicNMData Type (Numeric, Character, ...) Class Single Value / Multi Value Assignment Restricted ValuesNN11Class Type Organisational Area Fig. 3.1. Characteristics and Classes, 20 3 Characteristics and Classes Characteristics and classes are master data both in R/3 and in APO. • Characteristics The characteristics are created with transaction CT04 (both in R/3 and in APO) and contain the information about the data type, whether single or multiple values can be assigned and whether the values for the characteris- tic are restricted or not (checkbox ‘additional values’ on the values-view). Supported data types are character, numeric, date and time. • Classes Classes are created with transaction CL02 (both in R/3 and in APO). The key for the class is its name and its type, therefore the same class name can be used for different class types. Each class has a class type which restricts its usage. Table 3.1 lists the most important class types on R/3-side for characteristic based planning in APO and their usage. Table 3.1. Class Types in R/3 and their Usage Class Type Usage Assignment 018 Task List Routing 019 Work Center Work Center 023 Batch Classification Material Master 200 Material Classification Material Master 300 Variant Configuration Material Master For characteristics based planning only the class types 023 for batches and 300 for variant configuration are relevant. The class types 018 and 019 are used only in R/3 to valuate the master data. This information is used in the integration to APO either for block planning (class type 018, see chapter 8.3) or for the transfer of alternative resources (class type 019, see Dick- ersbach 2004). The material class for configurable products (class type 200) is used for a special case within the variant configuration scenario (see chapter 5). • Organisational Areas For the integration of classes and characteristics to APO the organisational area is used as the filter criterion. The assignment of the organisational area to the class is optional for most usages in R/3 but helpful to distin- guish those characteristics which are used for planning in APO and those which are only relevant for R/3. The customising for the organisational area is done with the path Cross-Application Components Classification System Classes Maintain Object Types and Class Types., 3.1 Characteristics and Classes in R/3 21 or with the transaction O1CL (in R/3 and in APO). Though the assignment of the organisational area to the classes and the characteristics is trans- ferred to APO, the transfer of the organisational area itself is not integrated and must therefore be created in APO as well. Fig. 3.2. Organisational Areas (R/3 and APO) The class type for material (table MARA) has to be selected before choos- ing the organisational area. The organisational area is class type specific. The assignment of both the class and the characteristics to the organisa- tional area takes place in the class. Assignment of the Characteristics to the Assignment of the Class to the Organisational Area (within the Class) Organisational Area Fig. 3.3. Assignment of the Class and the Characteristics to the Org. Area (R/3 and APO) • Assignment of Classes to the Master Data The assignment of the classes to the material master is done in the ‘classi- fication’-view. Only classes of one class type are displayed, but with F7 it is possible to switch between the class types. For variant configuration (class type 300) multiple classes can be assigned, for batch classification (class type 023) only one class is allowed. With the transaction CL24N it is possible to provide an overview about the class assignment to the material master and perform additional assignments as shown in figure 3.4., 22 3 Characteristics and Classes New Material Assignment Fig. 3.4. Class Assignment Overview Within the routing the classification is maintained from the operation via the menu path Extras Classification Operation and for the work center with the menu path Extras Classification. 3.2 Transfer of Characteristics and Classes to APO The transfer of classes and characteristics to APO is done via the CIF, and the organisational area is used as the selection criterion. For the transfer of classes and characteristics a separate integration model should be used (see note 714929), and the characteristics and classes should be transferred be- fore the materials are transferred. • Class Type Correspondence From a planning point of view only the R/3-class types 023 for batch clas- sification and 300 for variant configuration are relevant. The class type 300 has its correspondence on APO side with the same class type. Since planning with the CDP-configuration scheme is based on the APO-class type 400, a R/3-class with type 300 is transferred as a class of the type 300 and additionally (if the CDP-configuration scheme is active) as a class with type 400. For the use of batch classification the R/3-class type 023 is transferred as APO-class type 400 and additionally as class type 230, figure 3.5., 3.2 Transfer of Characteristics and Classes to APO 23

APO

Product Master Batch Class Type 300 Class Type 400 Class Type 230 R/3 Class Type 300 Class Type 023 Material Master Fig. 3.5. Class Types in R/3 and in APO The class type 230 is used in APO only for the valuation of the batches and is not assigned to the product master. • Assignment to the Product Master The assignment of the class to the product master is done automatically during CIF-transfer. The class is only assigned to the product master if the class contains characteristics. The class assignment can only be displayed in APO but not changed. Figure 3.6 shows how to display the class as- signment for further classes (if more than one class is assigned to the prod- uct). Fig. 3.6. Class Assignment to the Product Master in APO, 24 3 Characteristics and Classes • Transfer of Batches If batch classification is used in R/3 for CDP-configuration relevant prod- ucts, the integration of batches is required. The transfer of batches has to be selected explicitely in the CIF-model. The transfer of stocks is not suf- ficient to integrate the batches with their characteristic values with APO., 4 Configuration Scheme in APO For historical reasons there are two types of configuration schemes (for- merly known as class systems) in APO: Variant configuration (VC) and characteristic dependent planning (CDP). The VC-configuration scheme was developed mainly for the automotive and engineering industries and is closely linked to the variant configuration in R/3, whereas the CDP- configuration scheme was developed mainly for the mill industries. The main difference is that with the VC-configuration scheme only the sales orders have a configuration (i.e. characteristics with values). Other objects (e.g. planned orders) do not have an own configuration but a reference to the sales order via an internal key. Therefore the configuration that is dis- played for a planned order is the configuration of the sales order. The complete configuration – even for multiple levels – is defined at the entry of the sales order. Since only the sales order has a configuration, the make- to-order strategy is mandatory to keep the link between the sales order and the receipts during planning. With the CDP-configuration scheme sales orders, forecasts, planned or- ders, production orders and batches have an own configuration. Another difference is that the CDP-configuration scheme allows demands (sales or- ders, dependent demands) to define requirements for the characteristic val- ues of their supplies. Chapter 11.1.2 describes the logic for planning with the CDP-configuration scheme in detail. When using the CDP-configu- ration scheme, the characteristics values are considered in pegging. There- fore only few characteristics should be used. Note 526883 provides addi- tional information about the implications of using too many characteristics for CDP. Both configuration schemes offer advantages but in SCM 4.1 it is still only possible to use either the VC-configuration scheme or the CDP- configuration scheme per client. The customising path to define which configuration scheme is active for the client is APO Basis Settings Define Configuration Relevance. • Configuration Scheme and Configuration Relevance The term ‘configuration scheme’ refers to the setting of in the customising, whereas the ‘configuration relevance’ denotes whether the product actually, 26 4 Configuration Scheme in APO uses characteristics for planning. With APO 4.1 a product has either the configuration relevance as defined in the customising or none (e.g. when planning with shelf life). The outlook for further releases is to define the configuration relevance on product level (as a possibility). • Configuration Schemes and Functions Since APO 3.1 an effort has been started to harmonise the configuration schemes. Starting with the characteristic and class master data in APO 3.1, the product master was harmonised in APO 4.0 and the transactional data in APO 4.1. The latter two have been mostly internal changes, but with the result that more functions can be used with both configuration schemes. Table 4.1 provides an overview about the supported functions. Table 4.1. Configuration Scheme and Functions Function VC-Configu- CDP-Configu- ration Scheme ration Scheme Characteristic Based Forecasting Yes Yes Variant Configuration Yes Yes Multi-Level Variant Configuration Yes No Batch Selection No Yes Multi-Level ATP Yes No ATP with Characteristics No Yes Characteristic Values in Pegging No Yes Object Dependencies in the PDS Yes Yes Characteristic Propagation in the PPM No Yes Block Planning Yes Yes Shelf Life1 Yes Yes Planned Order Transfer Yes Automatic Conversion 1 Shelf Life Functionality is Independent of the Configuration Scheme For shelf life the product does not need any configuration relevance. Char- acteristics are only required for the batches, therefore the configuration scheme does not have any impact on the shelf life functionality. It is how- ever not possible to use shelf life for products with VC-configuration rele- vance. A CDP-configuration relevance on the other hand does not interfere with the shelf life functionality., 5 Make-to-Order with Variant Configuration 5.1 Scenario Description 5.1.1 Process Chain In some industries – e.g. in engineering – the finished products are config- urable (often up to a few hundred characteristics) and the number of units sold per product and characteristic combination is far from a commodity. In these cases the product is only produced after it is sold – i.e. after the sales order has been entered. The configuration for the product is main- tained in the sales order. The make-to-order strategy implies that the ATP check does not need to look for the availability of the finished product. The typical ATP method for this scenario is capable-to-promise (CTP) where planned orders are created for the product – if required, for several BOM levels including ca- pacity check and purchase requisitions. R/3-SD APO-ATP APO-PP/DS Sales Order Entry (Make-to-Order) Sales Order Creation Sales Order CTP Request Configuration Determination CTP Confirmation Planned Order Combination Creation Save Sales Order Detailed Scheduling Fig. 5.1. Process Chain for Make-to-Order with Variant Configuration, 28 5 Make-to-Order with Variant Configuration The focus for this scenario is on the sales side – there is no demand plan- ning, and few specifics for production planning and scheduling. Figure 5.1 shows the process flow for the sales order creation. For this scenario we assume that object dependencies are used and have to be integrated with APO. • Sales Order Creation and Assignment of Characteristic Values The process starts with the entry of a sales order. The sales order is created and configured as described in chapter 5.2. • ATP Check Since we are in a make-to-order scenario the finished product will not be available. There are three sensible options to perform an ATP check never- theless, of which capable-to-promise (CTP) is the most common. • Capable-to-Promise During the CTP-check a planned order is created and checked for capacity. The capacity check is either on a time-continuous and sequence dependent basis or based on a bucket capacity. The latter is available with APO 4.1 and the recommended alternative. The CTP-check can be applied for sev- eral levels, though with each level and with each finite resource the com- plexity increases. It is strongly recommended to keep the capacity check as simple as possible. Missing components might either be produced resp. procured or regarded as a constraint. This is controlled by the planning procedure of the product. This way both the production capacity and the material availability are checked. The object dependencies are considered as well. The assumption for this alternative is that both the production ca- pacity and the material availability might be bottlenecks. The use of CTP has major impacts and restrictions for production planning (especially if time-continuous CTP is used). For a more detailed description of the prop- erties and risks of CTP see note 426563 and Dickersbach (2004). • Multi-Level ATP Another option is the use of the multi-level ATP check. This option re- quires however that the key components are already available. Instead of the finished product the components for the finished product are checked (if required, for several levels). The demand for the components is calcu- lated using the PDS (the PPM could be used for multi-level ATP as well, but is not recommended for this scenario), and both the planned order du- ration and the object dependencies are considered. The assumption is that the production capacity is no bottleneck and that the components are, 5.1 Scenario Description 29 planned for. The use of multi-level ATP has impacts and restrictions for production planning. A limitation for the use of multi-level ATP is that the CDP-configuration scheme is not supported. A more detailed explanation of the multi-level ATP is provided in Dickersbach (2004). If planning of the assembly groups has a significant role within the business, the scenario ‘make-to-order with variant configuration and planning in an inactive ver- sion’ should be considered, therefore we do not assume the use of multi- level ATP for this scenario. • Checking Horizon The third option is the ATP check using the checking horizon. The sales order will be confirmed outside the checking horizon which represents the required lead time to produce the product. This setting is product specific and does not take the actual production capacity or the actual component availability into account. The assumption is that the check horizon pro- vides a sufficient time buffer to avoid material or capacity shortages. In most cases this leads either to unnecessarily late confirmations or causes problems in production and/or non-deliveries. With APO the two other methods provide usually better results, and therefore this option is not ex- plained any further. • Production Planning and Detailed Scheduling Production planning is done by the CTP check, but a detailed scheduling step still needs to be performed. 5.1.2 System Configuration Determinants The prerequisites for this scenario are the use of make-to-order assign- ments, class type 300 and the PDS master data object. The object depend- encies are transferred from R/3 and used for production planning in APO. Both the VC- and the CDP-configuration scheme are able to support this scenario, though it is rather unusual to use the CDP-configuration scheme. The exception are the mill industries. Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 5.2. System Configuration Determinants for MTO with Variant Configuration, 30 5 Make-to-Order with Variant Configuration If the CDP-configuration scheme is used, the planned orders are converted into production orders when they are transferred to R/3. Therefore the op- tion to transfer only production orders (i.e. planned orders with the conver- sion indicator) should be evaluated in this case. Another downside is the performance criticality of planning with many CDP-characteristics. The usage of a make-to-stock strategy in combination with the CDP- configuration scheme would be technically possible, though it is not ap- propriate for this scenario and the industries in target (engineering and automotive). This option will be mentioned as a special case within the scenario ‘configure-to-order with propagation’ in chapter 8. The make-to-order scenario with variant configuration requires the use of the PDS master data because it is not possible to integrate the object de- pendencies with the PPM. Though it would be possible to use a PPM in combination with the CDP-configuration scheme and model the object de- pendencies as macros in APO, this is a very circumstantial way. We define the integration of the object dependencies from R/3 as a part of the sce- nario, and therefore the PPM is incompatible with this scenario. 5.1.3 Configurable Material The standard material type for configurable materials is not the ‘finished product’ FERT but the ‘configurable material’ KMAT. This includes among others that an appropriate item category group (see chapter 5.2) is set and that the checkbox for variant configuration is set on the ‘basic data 2’- view, figure 5.3. Fig. 5.3. Checkbox ‘Material is Configurable’ (R/3) 5.1.4 Configuration Profile For the use of the variant configuration functionality in R/3 a configuration profile is required which defines the class types and the organisational ar- eas that are allowed for a material. The configuration profile is maintained with the transaction CU41., 5.2 Variant Configuration in the Sales Order 31 Configuration ProfileNMMaterialNN11Class Type Organisational Area Fig. 5.4. Configuration Profile (R/3) Only the class types 300 and 001 are allowed for the configuration profile. For a multi-level configuration and the configuration via material class ad- ditional settings in the detail of the configuration profile are necessary as described in the chapters 5.4 and 5.5. 5.2 Variant Configuration in the Sales Order The configuration of a product is performed as an assignment of character- istic values to the sales order. • Requirements Class and Item Category Group Prerequisites for the usage of the variant configuration functionality in the sales order are that the requirements class and item category support the variant configuration. Within the requirements class the indicator for con- figuration has at least to allow configuration. The standard item category for configuration is TAC. The item category is determined via the sales order type and the item category group as shown in figure 5.6. The customising paths for these entities are Sales and Distribution Sales Sales Documents Sales Document Item Define Item Categories resp. Define Item Category Groups and Assign Item Categories., 32 5 Make-to-Order with Variant Configuration Sales Order Type Item Category Group Item Category Fig. 5.5. Item Category for Variant Configuration (R/3) The standard item category group for configuration is 0002 and is main- tained in the ‘basic data 1’ and the ‘sales org. 2’ views of the material mas- ter as shown in figure 5.6. Fig. 5.6. Assignment of the Item Category Group to the Material Master (R/3) For the material type KMAT (configurable material) the item category group 0002 is set by default. • Configuration of the Sales Order The configuration of the sales order is performed in a follow-up screen af- ter the sales order item has been entered as shown in figure 5.7. ., 5.2 Variant Configuration in the Sales Order 33 Fig. 5.7. Configuration of the Sales Order (R/3) Depending on the configuration of the characteristics (i.e. whether the flag for additional values is set), the values are selected from the list or entered freely. It is possible to restrict the characteristic value combinations using variant tables. The assignment of the characteristic values is done before the ATP check is carried out. Not all characteristics have to be specified. Generally the use of unspecified characteristics is not recommended if the CDP-configuration scheme is used due to problems in pegging, production planning and ATP on characteristic level (see chapter 11.1). Since the variant configuration scenario is usually performed with a make-to-order strategy, the use of the sales order segment solves these problems. • Sales Order Representation in APO The sales order is transferred to APO with its configuration. In the product view (transaction /SAPAPO/RRP3) the green triangle indicates that a valua- tion exists. For the VC-configuration scheme in APO the order and the configuration details are shown in figure 5.8. The window on the left side shows the sales order segments. Green Triangle Sales Order Segments Fig. 5.8. Sales Order with VC-Configuration Relevance in the Sales Order Segment (APO), 34 5 Make-to-Order with Variant Configuration For variant configuration and the VC-configuration relevance the require- ments strategy ‘make-to-order’ is required. If the CDP-configuration scheme is used on APO side more details for the configuration exist – the valuations and the requirements. Fig. 5.9. Sales Order with CDP-Configuration Relevance (APO) The configuration of the sales order is transferred as valuation for CDP as shown in figure 5.9. The requirements (for the characteristic, see chap- ter 11) of the sales order are blank. 5.3 Object Dependencies 5.3.1 Structure and Usage of Object Dependencies Object dependencies are used in R/3 to influence the explosion of the BOM and the routing depending on characteristic values. The most com- mon types of object dependencies are selection conditions and procedures. Selection conditions control e.g. whether an item of a BOM is required de- pending on the value of a characteristic. Procedures on the other hand al- low to influence the value of a characteristic depending on the value of an- other characteristic. The object dependencies are either created as a separate entity with the transaction CU01 or within the BOM resp. routing itself. The creation of the object dependency as a separate entity has the advantage that it can be assigned to many different objects. The logic for the dependency is main- tained in the dependency editor as coding as shown in figure 5.10., 5.3 Object Dependencies 35 Fig. 5.10. Object Dependency (Selection Condition) (R/3) Optionally a dependency group can be assigned to the dependency. For some cases the use of the dependency groups SAPAPOACT and SAPAPOOPR is mandatory to ensure the correct integration to APO. These cases are described in chapter 5.4.4 about reference characteristics and 8.3 about block planning. Object dependency groups are maintained in cus- tomising with the path Logistics - General Variant Configuration Dependencies Main- tain Groups. In the example of figure 5.10 the selection condition is applied if the value of the characteristic is ‘BLACK’. The object dependency is assigned to the BOM resp. the routing as shown in figure 5.11. CTRL + F1 Fig. 5.11. Assignment of the Object Dependency to the BOM (R/3) Alternatively the object dependency can be maintained within the BOM or the routing as shown in figure 5.12. In this case the logic applies only to that master data., 36 5 Make-to-Order with Variant Configuration Fig. 5.12. Example for Object Dependencies (Selection) in the BOM (R/3) In this example the component CBP_COMP1 is selected if the value for the colour of the header product CBP_CAR_BIG is ‘RED’. • Object Variables Object variables define the BOM-level for which a selection condition or a statement of a procedure is applied. $SELF defines the current level, $PARENT one level above and $ROOT the top level – the level on which the configuration was entered. Only in the case of a multi-level configuration $ROOT and $PARENT will provide different results. • Syntax for Object Dependencies The syntax for the check and the assignment of characteristic values de- pends on the data type of the characteristic. Alphanumeric character values are assigned in quotation marks, numeric values without. Decimal places are maintained as a decimal point (e.g. 3.14). The operators ‘=’, ‘<>’, ‘AND’, ‘OR’, ‘NOT’ and ‘IF’ are supported among others. • Object Dependencies Check The object dependencies for a BOM or a routing (task list) can be checked in R/3 with transaction CU50 as shown in figure 5.13 without having to create a sales order., 5.3 Object Dependencies 37 Fig. 5.13. Check for the Object Dependencies in R/3 For object dependencies of the type procedure it is possible to create sepa- rate functions as described later in this chapter. • Integration of the Object Dependencies with APO Object dependencies within the BOM or the routing are transferred to APO – both for the VC- and the CDP-configuration relevance. This integration requires the master data object PDS. The PPM does not support the inte- gration of object dependencies. Figure 5.14 shows how the selection con- dition as maintained in the object dependencies (see figure 5.12) looks like in the PDS in APO. Fig. 5.14. Object Dependencies in the PDS in APO, 38 5 Make-to-Order with Variant Configuration The syntax of the object dependencies as displayed in the PDS is slightly different from R/3. Analogous to the object dependencies check in R/3 it is possible to simulate the impact of a configuration within APO. As shown in figure 5.15, for this purpose the ‘configuration’-tab within the PDS dis- play (transaction /SAPAPO/CURTO_SIMU) can be used. Maintain Configuration Apply Configuration Display Result Fig. 5.15. Simulation of a Configuration in APO For the display of the result one has to navigate to the respective objects (components, modes or activities) within the PDS. 5.3.2 Selection Conditions A selection condition controls whether an item of a BOM or an operation of a routing is used for a given configuration, e.g. if the value of the char- acteristic CBP_COLOUR is ‘BLACK’: CBP_COLOUR = ‘BLACK’ resp. if it is ‘BLACK’ or ‘GREY’: CBP_COLOUR = ‘BLACK’ OR CBP_COLOUR = ‘GREY’ The objects – BOM or routing – should be the maximal BOM resp. the maximal routing for this case, i.e. they should contain all the components resp. operations which might be necessary for any configuration. 5.3.3 Procedures Procedures are used to set the value of a characteristic e.g. depending on another characteristic value, 5.3 Object Dependencies 39 $SELF.CBP_COLOUR = ‘RED’ IF $ROOT.CBP_SIZE = ‘X’. or to propagate the characteristic value from the finished product to the component: $SELF.CBP_COLOUR = $ROOT.CBP_COLOUR. The syntax for multiple assignments is $SELF.CBP_COLOUR = ‘RED’ IF $ROOT.CBP_SIZE = ‘X’, $SELF.CBP_COLOUR = ‘BLUE’ IF $ROOT.CBP_SIZE = ‘Y’. 5.3.4 Procedures with Reference Characteristics Using reference characteristics it is possible to overrule master data set- tings by characteristic values, e.g. the quantity for a BOM item. Reference characteristics are ‘normal’ characteristics which have a link to the master data table and field on the ‘additional data’-view as shown in figure 5.16. Format of Field Overwrites Basic Data Settings Fig. 5.16. Reference Characteristic (R/3) The format and the description of the characteristic is taken from the refer- enced field. After saving the characteristic it is not possible to change the reference anymore. The reference characteristic has to be transferred to APO., 40 5 Make-to-Order with Variant Configuration Sales Order Assign the Value Class of the Characteristic Characteristic Determine the Value of the Manually Reference Characteristic by Reference Object Dependencies Characteristic

PDS

Characteristic Value of the Reference Characteristics Is Used For Object Dependency During Planned Order Creation Planned Order Planned Order with the Value of the Reference Characteristic Instead of the Standard Master Data Value Fig. 5.17. Scheme for the Use of Reference Characteristics Figure 5.17 shows the scheme for the use of reference characteristics. The reference characteristic is calculated by object dependencies, i.e. per de- fault the flag ‘not ready for input’ is set in the characteristic to inhibit a manual maintenance during characteristic value assignment in the sales or- der. The value of the reference characteristic is used during the planned order creation and replaces the standard value of the PDS for that order. Object dependencies have to exist in order to take the value of the refer- ence characteristic into account. • Dependency Group for Object Dependencies For dependencies for operations the dependency group SAPAPOOPR, for activities the dependency group SAPAPOACT has to be assigned to the de- pendency. The dependency group is maintained in the R/3 customising with the path Logistics General Variant Configuration Dependencies Main- tain Groups. This is necessary to map the dependency from the operation in R/3 to the right entity in APO – mode, operation or activity., 5.3 Object Dependencies 41 • Mapping of Reference Characteristics For the transfer of the reference characteristics to APO it is necessary to map the fields and tables as described in note 610873. The mapping table is found in the APO customising with the path APO Master Data Classification and Configuration Map Object Characteristics. Figure 5.18 shows an example for this mapping. Characteristic XREF_BOMRATIO2 in R/3 Mapping Table in APO Characteristic XREF_BOMRATIO2 in APO Fig. 5.18. Mapping of Reference Characteristics Table 5.1 lists correspondencies for some common fields as the BOM item quantity (STPOB-MENGE), the standard value for an operation1 (PLPO- VGW02) and the set-up group (PLPO-RFGRP). The set-up group can not be mapped without the use of variant functions (see also chapter 11.2.3). Table 5.1. Correspondence Between R/3 and APO-Fields R/3 APO Table Field Structure Field STPOB MENGE /SAPAPO/CULL_CFG_COMPONENT QUANT PLPO VGW021 /SAPAPO/CULL_CFG_MODE_PRODUCE DURVAR PLPO RFGRP /SAPAPO/CULL_CFG_ACT_ALL GROUP_ID2 PLPO RFSCH /SAPAPO/CULL_CFG_ACT_ALL ITEM_ID2 1 using the standard value key SAP1 for machine time 2 internal key; additional mapping with variant functions required The settings for the mapping have to be done before the transfer of the ref- erence characteristics since it is not possible to change the reference to the tables within the characteristic anymore. For the overruling of PDS com- ponents using reference characteristic note 815018 has to be considered., 42 5 Make-to-Order with Variant Configuration 5.3.5 Variant Functions Variant functions allow to model complex dependencies using function modules with ABAP coding. This way it is possible to use information that is not included in the characteristics. An example for the use of a variant function is given in chapter 11 for the characteristic based determination of the set-up group. The variant function references a function module (via the identical name) and is called by the object dependency. The variant function and the function module have to be created manually both in R/3 and in APO. Though the variant function and the function module are only evaluated in APO, the object dependencies are maintained in R/3 and for consistency reasons the variant function and the function module have to exist in R/3 as well. It is however sufficient to create a function module without source code in R/3 (especially if function modules and tables are used which are only available in APO) as long as no re-explosion of the master data is per- formed in R/3. Figure 5.19 visualises the structure of these entities: R/3 APO PDS Transfer Object Dependency Object Dependency Integrated into the PDS Create with CU01 Identical Name Variant Function Variant Function Create with CU65 Create with CU65 Link By Link By Identical Name Identical Name Function Module Function Module (Source Code Not Used) Create with SE37 Create with SE37 Fig. 5.19. Structure Overview for Variant Functions Note that the variant functions in R/3 and APO must have the same name. • Function Module Function modules are created with transaction SE37 and are assigned to a function group. The function group is created on the same screen via the menu path Goto Function Groups Create Group. The function module for the variant functions requires the import parameter GLOBALS LIKE CUOV_00 the tables, 5.3 Object Dependencies 43 QUERY LIKE CUOV_01 MATCH LIKE CUOV_01 and the exceptions

FAIL

INTERNAL_ERROR where the table QUERY contains the data for the import characteristics that is provided by the function resp. the object dependency to the function module and the table MATCH contains the data that will be passed back to the function and the object dependency. The structure CUOV_01 contains the fields VARNAM for the name of the characteristic, ATFOR for the for- mat of the value, ATWRT for the value as a characteristic and ATFLV for the numeric value. The online documentation provides additional information. • Variant Function The variant functions define the characteristics which are used as an input and those that will change their values. Variant functions are created with the transaction CU65, and figure 5.20 shows their maintenance. Output Characteristic Input Characteristic Fig. 5.20. Variant Functions (R/3 and APO) Note that the input characteristics require the flag ‘input parameter’. The maintenance of the variant function requires a status. For its use in the ob- ject dependencies the status has to be ‘1’ (released). In APO the statuses ‘released’, ‘in preparation’ and ‘locked’ might not exist as entries in the, 44 5 Make-to-Order with Variant Configuration table CUVFUN_ST. Note 601255 explains how to maintain the required en- tries. • Use of Variant Function in the Object Dependency To call a variant function VARFUNC in a procedure with the input charac- teristic INCHAR and the output characteristic OUTCHAR the following syn- tax is required: FUNCTION VARFUNC (INCHAR = $ROOT.INCHAR, OUTCHAR = $SELF.OUTCHAR) For the input characteristics the values have to be derived from the entry point of the valuation. 5.4 Multi-Level Configuration If the component for a configurable material is a configurable material as well (item category ‘N’ in the BOM required), the configuration for the component is either determined by object dependencies, by fix value as- signment from the material master or interactively assigned in the sales or- der. All the three options are supported by APO. For the interactive as- signment of the characteristic values the configuration profiles for both the finished product and the component must support the single or the multi- level BOM explosion, figure 5.21. Configuration Profile for Finished Product Single- or Multi-Level BOM Explosion Details Configuration Profile for Component Details Fig. 5.21. Configuration Profiles for Multi-Level Configuration (R/3) Unless a set is configured, the ‘planned order’ process should be selected. The characteristic value assignment in the sales order is performed per ma- terial as shown in figure 5.22., 5.4 Multi-Level Configuration 45 Configuration of the Finished Product Object Dependency Configuration of the Assembly Group Fig. 5.22. Multi-Level Characteristic Value Assignment in the Sales Order (R/3)

Object dependencies and fix values are taken into account. The valuations

for both the sales order and the dependent demand are transferred to APO, figure 5.23. Finished Product Component Fig. 5.23. Multi-Level Configuration in APO

The dependent demand is in the customer order segment of the sales order.

, 46 5 Make-to-Order with Variant Configuration 5.5 Configuration via Material Class An alternative approach to select components depending on the character- istic value of the finished product is the use of material classes for config- urable materials (class type 200) instead of object dependencies. The basic idea is that the components have characteristic values assigned (of the class type 200 since the components are not configurable) and are selected according to the characteristic values of the sales order for the finished product. Figure 5.24 provides an overview about the required settings. Value Material Material CPU1 CBP_PC CBP_PC_CPU01 Value BOM Material CPU2 CBP_PC_CPU02 Class Type 300 Class Type 200 CBP_PC_VC CBP_PC Characteristic CBP_CPU Fig. 5.24. Characteristic Assignments for Variant Configuration via Material Class The configuration of the sales order causes the selection of the component with the same characteristic value – i.e. if the sales order is configured with the value ‘CPU1’, then the component CBP_PC_CPU01 is selected. This is possible because the material class (type 200) is assigned to the BOM instead of the individual components. For the assignment of the material class to the BOM, the class must have the indicator to allow its use in the BOM, figure 5.25., 5.6 Material Variants 47 Class

BOM

Fig. 5.25. Class Assignment to the BOM (R/3) The material classes for configurable materials do not have to be trans- ferred to APO. During the transfer of the PDS the fix characteristic value of the component (that is set in the material master of the component) is interpreted as an object dependency and transformed into selection criteria as shown in figure 5.26. Fig. 5.26. PDS for Variant Configuration via Material Class (APO) No object dependencies are required in R/3 for this approach. 5.6 Material Variants The idea for the use of material variants is to produce common configura- tions of a configurable product as make-to-stock. This has the advantage of faster response to customer demands and offers the possibility to level fluctuations in the resource utilisation. • Master Data for Material Variants The material variants have a reference to a configurable product (for which a configuration profile must exist). The material variants are no configur-, 48 5 Make-to-Order with Variant Configuration able products (i.e. they are not of the type KMAT) nor do they have any class assignments. In the MRP3-view of the material master the link to the configurable product is maintained and the material is configured as shown in figure 5.27. Fig. 5.27. Definition of a Material Variant (R/3) To create a production version for the material variant it is necessary to create a reference to the BOM and to the routing of the configured mate- rial. For the BOM this is performed with the transaction CS40 as shown in figure 5.28. Indicator for Material Variant Fig. 5.28. Assignment of the Material Variant to the BOM of the Configured Material (R/3) Only the material variant is entered in this transaction. The link to the BOM is found via the referenced configurable material. In the routing of the configurable material (transaction CA02) the material variant is as- signed via the ‘material assignment’-button as shown in figure 5.29., 5.6 Material Variants 49 Fig. 5.29. Assignment of the Material Variant to the Routing of the Configured Material With these prerequisites the production version for the material variant is created and transferred as a PDS. The components which have a selection condition for different characteristic values are filtered during the transfer. • Material Substitution in the Sales Order During the sales order process the entered configuration is checked against existing material variants and – depending on the settings in the item cate- gory – either a notification appears or the product is automatically substi- tuted, figure 5.30. Original Item CBP_CAR Configuration as for Material Variant Substituted Item CBP_CAR_VAR Fig. 5.30. Substitution of the Configured Material by the Material Variant (R/3) With the substitution of the configurable product by the material variant the complete process is switched from make-to-order to make-to-stock., 50 5 Make-to-Order with Variant Configuration • Material Variants in APO The material variants in APO still have the link to the configured product as shown in figure 5.31. Fig. 5.31. Product Master for Material Variants in APO In the product view the triangle is displayed that indicates characteristics, but without colour. Though the configuration is stored, the characteristics are not used for planning for the material variants. Fig. 5.32. Product View for Material Variants in APO The PDS for the material variant is the same as the PDS of the configured product if the configuration was applied. The object dependencies are dis- played if they are applied for the configuration. 5.7 Configurable Material Variants For the configurable material variants the variant is not completely config- ured (i.e. not all characteristic values are maintained) in the material mas- ter – the missing configuration is performed in the sales order. The busi- ness case for the use of configurable material variants is very different, 5.7 Configurable Material Variants 51 from the use of ‘normal’ material variants. The motivation to use configur- able material variants is to reduce the effort of maintaining the characteris- tic values for the configuration and to reduce the effort of master data maintenance – especially for changes. The sales order is placed for the configurable material variant, and therefore the order is already pre-configured. Usually only a certain per- centage of the order is pre-configured, and the missing characteristic val- ues are maintained as for the normal variant configuration, see figure 5.33. Configuration in the Sales Order of the Material Variant Pre-Configuration via Material Variant Assignment Configuration in the Sales Order Fig. 5.33. Sales Order for the Configurable Material Variant Using configurable material variants no material substitution takes place nor would it be desirable. A typical case for the use of configurable mate- rial variants are the mill industries, where a quantity related property is maintained as a characteristic – e.g. the length of a tube. This case implies that the length has to be maintained as a configuration in the sales order. • Settings for the Use of Configurable Material Variants Though the material variant is configured in the sales order, the configur- able material variant is not of the type configurable material (KMAT) and the indicator for ‘material is configurable’ is not set. The difference is that the item category has to support configuration (standard item category TAC resp. item category group 0002) and that the requirements class has to contain the setting that the configuration is mandatory (standard require- ments class 046). If the configuration is not mandatory, the configuration screen does not appear since no configuration profile exists. If the CDP-configuration scheme is used (which is the usual case since configurable material variants are mostly used in the mill industries), for the integration of the PDS for the configurable material variant the BAdI /SAPAPO/CURTO_CREATE has to be implemented. In the method VARIANT_WITH_OWN_VALUATION the statement EV_DIFFERENT_COMPONENT_VALUES = ‘X’. needs to be added., 52 5 Make-to-Order with Variant Configuration • Configurable Material Variants in APO The configurable material variants look in APO the same as configurable products. In the product master the link to the configurable material exists like for the ‘normal’ material variants. 5.8 Variant Tables Variant tables are used to maintain non-linear dependencies between dif- ferent characteristics. Differing from normal tables, the variant tables can be used for the object dependencies (without the necessity to use variant functions). The variant tables are created in two steps: First the table is de- fined with the key characteristics as shown in figure 5.34. Definition as Key Fig. 5.34. Definition of the Variant Table (R/3 and APO) The table and its characteristics are defined with the transaction CU61. The maintenance of the values for the table is performed with the transaction CU60, figure 5.35. Fig. 5.35. Table Maintenance for Variant Tables (R/3 and APO) The syntax to assign a value from the variant table VARTAB to the charac- teristic CHAR depending on the value of the key characteristic KEYCHAR in the object dependencies is, 5.8 Variant Tables 53 TABLE VARTAB ( CHAR = CHAR, KEYCHAR = $SELF.KEYCHAR) The variant table has to be created in R/3 and in APO as well before trans- ferring the PDS. If the statuses for the variant table definition are missing, they have to be created explicitely as described in note 614280., 6 Make-to-Order with VC and Demand Planning 6.1 Make-to-Order with VC and Planning Overview The motivation to combine the variant configuration with the demand planning is to plan the demands for the assembly groups based on the de- mand for the finished product. This implies that demand planning has to include the characteristic values as planning levels – which is enabled by the use of characteristic based forecasting (CBF). Since the configurable materials have usually a make-to-order strategy, the forecast on finished product level is not used for the production of the finished product but for the production of the assembly groups (resp. the procurement of components). APO offers two alternative ways to apply CBF: One is to perform the production planning for the finished product in an inactive planning version and to transfer the dependent demand as fore- cast for the assembly group to the active version. The other alternative is to release the forecast with the characteristic value assignment to the plan- ning segment of the active version. Figure 6.1 shows the overview about these two scenarios again. Scenario Demand Planning Sales Production MTO with Variant Configuration Characteristic Based Multi-Level ATP or Production Planning with and Planning in Inactive Version Forecasting Capable-to-Promise Characteristics MTO with Variant Configuration Characteristic Based Forecast Check and/or Production Planning with and Planning in Active Version Forecasting Allocation Check Characteristics Fig. 6.1. Scenarios for Make-to-Order with Variant Configuration and Planning The first alternative is available since APO 3.0 and therefore the more of- ten used. The second alternative is available since APO 4.0 and requires the CDP-configuration scheme with its restrictions regarding planned or- der transfer and performance when using many characteristics (see chap- ter 11). The target industry for the first alternative is engineering, automo-, 56 6 Make-to-Order with VC and Demand Planning tive and high tech/consumer goods, while the target for the second alterna- tive is the mill industries. 6.2 Characteristic Based Forecasting Characteristic based forecasting (CBF) is a functionality of Demand Plan- ning. This leads us to a problem in the terminology because the term ‘characteristic’ is used in two ways: 1. characteristics as used within this book so far, i.e. an attribute to the product resp. the order and 2. characteristic as a basic object for Demand Planning which defines levels of planning (e.g. product, location, sales organisation etc.). The actual planning is performed for characteristic value combina- tions (CVCs), which are the master data for Demand Planning. For the purpose of distinction the first type of characteristics is called ‘con- figuration-characteristic’ and the second ‘info object-characteristic’ when- ever necessary. Consequently, in the case of CBF the demand planning is carried out on configuration-characteristic level, where the configuration- characteristic is used as info object-characteristic. Still not confused? • Planning Object Structure for CBF In the basic configuration for demand planning the main prerequisite is that the flag for CBF is set in the basic planning object structure. This con- trols that the three info object-characteristics 9AMV_PROF, 9AMV_TAB and 9AMV_ROW are included as shown in figure 6.2. Planning Object Structure1NPlanning Area Checkbox for CBF Characteristics Key Figures Base Unit Char. 9AMV_PROF Storage Bucket Profile Char. 9AMV_TAB Char. 9AMV_ROW Fig. 6.2. Planning Object Structure for CBF (APO), 6.2 Characteristic Based Forecasting 57 Those info object-characteristics provide the link to the CBF-profile and the CBF-table. The only difference for the planning area is that if a key figure shall have the option for fixing, the second key figure for the fixed values must be added to the planning area manually. • CBF-Profile and CBF-Table Overview CBF-profile and CBF-table are master data which are linked to the product and define the combination of the configuration-characteristics which may be used for this product in Demand Planning as info-object characteristic value combinations (in addition to the already existing, non-configuration based CVCs). Figure 6.3 provides an overview about the dependencies of CBF-profile and CBF-table to the other objects. N1N1Product CBF-Profile Planning Area

N

CVC Deselection optional CBF-Table CVC Characteristics Fig. 6.3. CBF-Profile and CBF-Table (APO) For each product a CBF-profile has to be created, but a product can only be assigned to one CBF-profile. One CBF-table can be assigned to more than one CBF-profile. Both the CBF-profile and the CBF-table are main- tained in the integrated CBF-profile maintenance. • CBF-Table In the CBF-profile maintenance with transaction /SAPAPO/IPM01 you can choose between a product, a planning area or a CBF-table. After choosing ‘CBF-table’, first the name for the CBF-table has to be defined and then the configuration-characteristics have to be assigned to the table as shown in figure 6.4., 58 6 Make-to-Order with VC and Demand Planning Create New CBF-Table Characteristics which are already Part of any CBF- Other Characteristics are Table are listed here. Selected with F4-Help. They are assigned to the CBF-Table with the Button. Fig. 6.4. CBF-Table – Assignment of Characteristics (APO)

The definition of the relevant characteristic value combinations (CVCs) in

the CBF-profile is done either interactively for explicit CVCs or is gener- ated from a subset of characteristic values as shown in figure 6.5: a) b) a) Create CVC Manually b) Generate CVC: Permutation of Selected Values Select Characteristic Values Generate Fig. 6.5. CBF-Table – Creation of Characteristic Value Combinations (APO)

The defined combination of values for configuration-characteristics will be

used as planning levels in Demand Planning, i.e. for those CVCs planning will be possible. The total number of CVCs in Demand Planning (per product) will be the number of the CVCs from the CBF-profile times the number of CVCs without consideration of configuration-characteristics., 6.2 Characteristic Based Forecasting 59 • CBF-Profile The CBF-profile is created with the same transaction /SAPAPO/IPM01. In this case ‘product’ has to be chosen and the product has to be specified. For the maintenance of the CBF-profile the following 5 steps must be per- formed: 1. create the CBF-profile 2. assign a planning area 3. activate the CBF-profile 4. assign a CBF-table 5. optionally: deselect CVCs from the CBF-table. Figure 6.6 shows the CBF-profile maintenance. Fig. 6.6. CBF-Profile Maintenance in APO The CBF-profile needs to be created before the characteristic value combi- nations are maintained. • Characteristic Value Combinations in DP By creating the CBF-profile and assigning it to the product the values for the info object-characteristics 9AMV_PROF (CBF-profile), 9AMV_TAB (CBF-table) and 9AMV_ROW (characteristic base) are populated. Fig- ure 6.7 shows the characteristic value combinations (CVC) for the plan- ning object structure as displayed with transaction /SAPAPO/MC62., 60 6 Make-to-Order with VC and Demand Planning Fig. 6.7. Characteristic Value Combination for Planning Object Structure The characteristic values for the CBF-profile, CBF-table and characteristic base provide the link to the characteristic profile. This way the configura- tion-characteristic values are planned for even though the info object- characteristic value combinations are different. If no CBF-profile exists, the CVCs are not created for a CBF-relevant planning object structure. • Master Data Changes For the change of the info-object characteristics, for example if a new product is added. In this case the difference to the normal process of char- acteristic value maintenance is that additionally the CBF-profile has to be created for the product. • Demand Planning with Configuration-Characteristics The characteristic values are used in planning and navigation just like normal characteristics – including drill down and selection in the shuffler. • Limitations for CBF CBF is only supported for the planning of finished products because the forecast consumption on characteristics level does not work with depend- ent demands. The number of key figures is limited to 20 including APODPDANT in the planning book when using CBF (the number of key figures for the planning area is not limited) and the disaggregation type ‘N’ (no disaggregation) is not supported. See note 484144 for more informa- tion. The results of CBF can only be used in PP/DS and not in SNP. 6.3 Forecast Release and Forecast Consumption In the following we describe the alternatives for the forecast release and the forecast consumption in general – i.e. not only the functions that are used in the two scenarios for make-to-order with variant configuration and planning, but also for the scenario ‘sales order oriented planning’., 6.3 Forecast Release and Forecast Consumption 61 • Forecast Release The release of the forecast needs to be done in different ways depending on the scenario. One of the main distinguishing features is whether for each CVC a separate forecast is released or whether the CVCs are aggre- gated and the forecast is released only with the class information. If sepa- rate forecasts per CVC are released, the forecast consumption is performed on CVC-level as well. To control the creation of separate forecasts per CVC a consumption group has to be defined and assigned to the product master and maintained in the release settings. For the scenario using the VC-configuration scheme the forecast is released in an aggregated way, for the scenario using the CDP-configuration scheme and the scenario ‘sales order oriented planning’ using descriptive characteristics the consumption group is applied to create separate forecasts per CVC. Another differenti- ator is whether the forecast is released to the planning segment or to the make-to-stock segment, which is controlled by the requirements strategy in the product master or as a setting in the release job. If a make-to-stock sce- nario is used, the release has to be done to the make-to-stock segment, which requires a BAdI for the scenarios with CDP-configuration relevance (there is no make-to-stock scenario for the VC-configuration relevance, though the forecast is released to the make-to-stock segment but to an in- active version). Note that an integration of planned and production orders with configuration with a make-to-stock strategy is only possible with IS Mill. Figure 6.8 provides an overview about the different options. Demand Planning Characteristic Based Forecasting Sales Order MTO with VC MTO with VC Oriented Planning and Planning in and Planning in Active Version Inactive Version Consumption Consumption Consumption Group Group Group BAdI Forecast with Forecast with Descriptive Forecast with Forecast with Class but no Characteristic CDP-Values CDP-Values Values Values Make-to-Stock Segment Planning Segment Planning Segment Active Version Inactive Version Fig. 6.8. Forecast Release Options, 62 6 Make-to-Order with VC and Demand Planning If CBF is used, then for the release with consumption group to the active version the CDP-configuration relevance is required. The release to the inactive version is done with the VC-configuration relevance. Figure 6.9 shows the different representation of the characteristics in the forecast – depending whether only descriptive characteristics, the CDP- configuration relevance or the VC-configuration relevance is used. VC-Configuration Relevance CDP-Configuration Relevance Characteristic Details Descriptive Characteristics Fig. 6.9. Characteristic Details for Forecasts – Different Configuration Relevance (APO) The green triangle as the indicator for characteristics is displayed for all the three kinds of the forecast. • Consumption Group After the release, the forecast is usually on the level of the product and the location only. This is the level the forecast consumption takes place as well. With the consumption group it is possible to have separate lines per characteristic value combination for a forecast. For the forecast consump- tion the characteristics which are used in demand planning are assigned to fields from the sales order. The consumption group is maintained with the transaction /SAPAPO/CSP1 per planning area and assigned to the product master and the release settings. For the scenario ‘sales order oriented planning’ the ‘normal’ info object- characteristics from DP are used as descriptive characteristics. In this ex- ample we want to consume the forecast on customer level. Therefore the characteristic CUSTOMER has been included into the basic planning object structure. This characteristic has to be assigned now to the field KUNNR of the sales order. Figure 6.10 shows the settings for the consumption group for this example., 6.3 Forecast Release and Forecast Consumption 63 Fig. 6.10. Consumption Group for Descriptive Characteristics (APO) For CBF the consumption group is used as well, but not the info object- characteristics are now chosen but the configuration-characteristics. In the ‘DP characteristic assignment’-view the configuration-characteristics that have been maintained in the CBF-profile are available, figure 6.11. The field catalogue assignment is carried out automatically. Adopt Characteristics Fig. 6.11. Consumption Group for CBF (APO) The forecast receives its configuration only via the vehicle of the descrip- tive characteristics. If forecast with CDP-characteristic relevance is used, 64 6 Make-to-Order with VC and Demand Planning for planning or for the ATP check, the consumption group has always to be defined for the forecast release. • Assignment of the Consumption Group The consumption group has to be assigned to the ‘demand’-view of the product master as shown in figure 6.12 and to the settings for the release of the forecast. Fig. 6.12. Assignment of the Consumption Group to the Product Master (APO) The assignment of the consumption group to the release settings (transac- tion /SAPAPO/MC90) is shown in figure 6.13. Fig. 6.13. Assignment of the Consumption Group to the Forecast Release Settings (APO) The forecast is defined by locationproduct, planning segment and the de- scriptive characteristics. Therefore the consumption group is often required as well when accessing the forecast (e.g. for deletion reports), but not for the reorganisation with transaction /SAPAPO/MD74. • Release with Consumption Group to the Make-to-Stock Segment Though this is not part of any of the selected scenarios, it might be re- quired to release the forecast with CDP-configuration relevance to the, 6.4 Scenario Description for Planning in Inactive Version 65 make-to-stock segment (and have separate forecasts per CVC for planning and forecast consumption). This bears some limitations – e.g. it is not pos- sible to transfer the planned orders to R/3 unless they are converted to pro- duction orders (each planned order with CDP-configuration relevance is converted into a production order in R/3). The setting to prevent the trans- fer of planned orders is on client level. Therefore this option should only be considered if planned orders are not required in R/3. From a configuration point of view it is required to activate the BAdI /SAPAPO/DM_BADI_CONF and add following line of coding to the method CONFR_FILL: None of the selected scenarios requires this option. 6.4 Scenario Description for Planning in Inactive Version 6.4.1 Process Chain for Planning in Inactive Version The basic idea of this scenario is to use the demand planning information for the configured finished products to calculate the dependent demand for the non-configured components. This scenario is limited to a single-level configuration. The calculation of the dependent demand takes place in an inactive version, and the dependent demand from the inactive version is transferred as forecast for the components into the active version. Figure 6.14 shows the process chain for this scenario., 66 6 Make-to-Order with VC and Demand Planning R/3-SD APO-DP APO-ATP APO-PP/DS Sales Order Creation Demand Planning Sales Order Entry Characteristic Based (Make-to-Order) Forecasting Production Planning in Inactive Version Release Forecast Production Planning to Inactive Version in Inactive Version Transformation of Dependent Demand to Forecast Forecast Consumption by Dependent Demand Planned Orders for Semi-Finished Production Planning for Semi-Finished Sales Order ML-ATP Request Configuration Determination Confirmation Planned Order Save Sales Order Combination Creation for Finished Production Planning Fig. 6.14. Process Chain for MTO with VC and Planning in Inactive Version Since this scenario was developed for the VC-configuration scheme we will describe the steps for the VC-configuration scheme only. Though it is technically possible to apply the CDP-configuration scheme (with minor deviations, especially for the release of the forecast) for this scenario as well, this would be rather unusual. Demand planning is performed using CBF on finished product level in- cluding the characteristic values of the finished product. The forecast is re- leased to an inactive planning version without any consumption group. For the first BOM-level production planning is performed in the inactive ver- sion, and the dependent demands are transformed as forecasts (i.e. planned independent requirements) for the assembly groups into the active version. These forecasts are the basis for the production or procurement of the as- sembly groups resp. components. If a sales order is placed, the forecast is consumed by the dependent demand. The inactive planning version is used because with the VC-configu- ration scheme no forecast consumption is possible on characteristic level for the finished product. A release of the forecast to the active planning version is possible as well but would lead to an imprecise forecast con- sumption., 6.4 Scenario Description for Planning in Inactive Version 67 • Demand Planning with Characteristics The finished products are planned for on characteristic value level in De- mand Planning using the CBF functionality. As shown in figure 6.15, the only difference is that the planning takes place on configuration- characteristic level as well. Info Object-Characteristic Configuration-Characteristic Fig. 6.15. Demand Planning with CBF (APO) From a user’s point of view the configuration-characteristics behave the same as the info-object characteristics. • Forecast Release In the next step the forecast is released into an inactive version. There are two points of attention: - there must not be a consumption group neither in the settings for the release (transaction /SAPAPO/MC90) nor in the product master and - the strategy for planning without final assembly (e.g. standard strat- egy 30) has to be maintained either in the settings for the release or in the product master. As shown in the log of the release, only one forecast was released per loca- tionproduct. Fig. 6.16. Log for the Release of the Forecast (APO) The forecast contains the information about the class but not about the characteristic values. If more than one forecast exists within the same, 68 6 Make-to-Order with VC and Demand Planning bucket but with different characteristic values, still only one forecast is re- leased as shown in figure 6.17. Fig. 6.17. Forecast after Release (APO) Note that the values for the characteristics are blank and that there is only one forecast per bucket. • Production Planning in the Inactive Version For the production planning in the inactive version the standard heuristics are used. Since the planning procedure for multi-level ATP contains the heuristic SAP_PP_CTP which is not suited to cover demands, the produc- tion planning heuristic SAP_PP_002 must be included into the MRP heuris- tic explicitely. As a prerequisite the PDS master data with the object dependencies is transferred from R/3. The result of the production planning run are planned orders which have the class assigned but no characteristic values as shown in figure 6.18 – analogous to the forecast., 6.4 Scenario Description for Planning in Inactive Version 69 Fig. 6.18. Planned Order with VC-Configuration Relevance (Without Configuration) The dependent demand for the components reflects however the character- istic values of the forecast. In this example the characteristic values of the forecast were 12 each for the colours ‘BLACK’, ‘BLUE’ and ‘RED’. Com- ponent 1 is required for all colours, component 2 only for ‘RED’ and com- ponent 3 only for ‘BLUE’. Figure 6.19 shows the dependent demand of the planned order. Fig. 6.19. Dependent Demand Reflecting the Characteristic Values of the Forecast The characteristic values of the demand are taken into account because during the planned order creation the characteristic values are read from demand planning. The usage of a requirements strategy for planning with- out final assembly is required for this – else the characteristic values are not read from demand planning. The planned orders are not transferred to R/3 because they are in an inactive version. • Transformation of the Dependent Demand to a Forecast The purpose of the previous steps was to calculate the expected dependent demand for the assembly groups based on the forecast for the finished, 70 6 Make-to-Order with VC and Demand Planning products. The dependent demand for the assembly groups in the inactive version is transformed to a forecast for the assembly groups in the active version as described in the next chapter. • Sales Order Creation The sales order is created and configured as described in chapter 5 for make-to-order with variant configuration. • ATP Check and Production Planning Sensible alternatives for the ATP check are either the multi-level ATP or CTP. The purpose of the whole scenario is to create a better demand plan for the components which implies that the components might be bottle- necks. The basic properties of these and the resulting planned order crea- tion is described in Dickersbach (2004). • Forecast Consumption by Dependent Demand The ATP check creates planned orders either via CTP or multi-level ATP. The dependent demand of the planned orders consume the forecast for the assembly groups. 6.4.2 System Configuration Determinants For the make-to-order scenario with variant configuration and planning in an inactive version the VC-configuration scheme is required. On R/3-side a make-to-order strategy and a variant configuration class is required, fig- ure 6.20. Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 6.20. Determinants for MTO with VC and Planning in an Inactive Version Since object dependencies are used, again the PDS master data object is mandatory. The PDS with object dependencies is transferred for the fin- ished products. Though these will be used in an inactive version, it is not necessary to assign them explicitely to the inactive version., 6.4 Scenario Description for Planning in Inactive Version 71 6.4.3 Transformation of the Dependent Demand Since the purpose of the previous steps is to calculate the demand for the assembly groups in order to trigger production, the dependent demand has to be transformed into a forecast for the assembly groups. The more tedi- ous way to do so would be to transfer the dependent demand to a key fig- ure in Demand Planning and release the key figure content to SNP into the active version. The shortcut for this process is to use the transaction /SAPAPO/DMP2 as shown in figure 6.21. Fig. 6.21. Transform Dependent Demand to Forecast (APO) The product selection has to contain the components. Only those compo- nents are selected which have a strategy for planning with final assembly (e.g. standard strategy 20) and are flagged for assembly planning. After a successful transformation the log displays the generated forecasts as shown in figure 6.22. Fig. 6.22. Log for the Generation of Forecasts (APO), 72 6 Make-to-Order with VC and Demand Planning The result of this step is the forecast for the components in the active ver- sion as shown in figure 6.23. Inactive Version Active Version Fig. 6.23. Transformation of the Dependant Demand to Forecast (APO) The forecast is consumed by the dependent demand as described in the next chapter. 6.4.4 Forecast Consumption by the Dependent Demand The prerequisites for the forecast consumption by the dependent demand on assembly group level is that the components are planned as make-to- stock and that the - the requirements strategy allows planning with final assembly, - the checkbox for assembly planning is selected in the product mas- ter and - the dependent demands are collective requirements as shown in figure 6.24., 6.4 Scenario Description for Planning in Inactive Version 73 Fig. 6.24. Demand and Strategy Settings for the Component in the Product Master (APO) Note that the ATP categories AY for the dependent demand and AU, AV and AX for the order reservation have to be included into the category group of the strategy. Fig. 6.25. Forecast Consumption for the Component (APO) The consumption situation overview (transaction /SAPAPO/DMP1) shows that the forecast is consumed by the dependent demand of the planned or- der on the assembly group level as shown in figure 6.25., 74 6 Make-to-Order with VC and Demand Planning 6.4.5 Component Logic The usage of the component logic is an optional functionality which is only possible for the VC-configuration scheme. Demand Planning takes great care about consistency of the data across the different planning levels. In some cases this consistency is not desired – an example would be the planning for PCs where the components – CPUs, hard discs and others – are modelled as characteristics. These com- ponents can be used in different PCs. Since the customer demand is stronger related to the technical properties than to the PC-model, demand planning on characteristic level might be desired. Now one PC might have more than one hard disc, and to allow a planning for more hard discs than PCs an info object-characteristic value for the difference is introduced: Increase Forecast on Detailed Level By 5 Delta to Balance Increase Fig. 6.26. Balance for Overplanning (APO) Instead of aggregating the forecast figure on the detailed level this level gets balanced. The overplanned characteristic value – in this example 55 for ‘100 GB’ instead of 50 – is used for the object dependencies during the PDS explosion. As result the planned order has a dependent demand of 55 for the selected component, figure 6.27. Fig. 6.27. Dependent Requirements in the Planned Order Using Component Logic (APO) • Settings in the CBF-Profile for the Use of the Component Logic The prerequisite for this behaviour is that in the CBF-profile the settings for the component logic are made. The maintenance of these settings is more stable with the old transaction /SAPAPO/DPC1 as shown in fig- ure 6.28 step by step., 6.5 Scenario Description for Planning in Active Version 75 Fig. 6.28. Maintenance of the Component Logic in the CBF-Profile (APO) When entering the CBF-profile, make sure that the ‘valid from’-entry is maintained. 6.5 Scenario Description for Planning in Active Version 6.5.1 Process Chain for Planning in Active Version This scenario was developed for and requires the use of the CDP- configuration scheme. The main difference to the previous scenario (plan- ning in an inactive version) is that no inactive planning version is required. The planning of the finished product in order to determine the dependent demand for the assembly groups resp. the components is done in the plan- ning segment of the active version. The reason for using the planning seg- ment is that the production for the finished product is in most cases only started when a sales order is placed. Another reason is that planned orders (and production orders) with CDP-configuration relevance in the make-to- stock segment can not be integrated with R/3 unless IS Mill is used., 76 6 Make-to-Order with VC and Demand Planning R/3-SD APO-DP APO-ATP APO-PP/DS Sales Order Creation Demand Planning Sales Order Entry Characteristic Based (Make-to-Order) Forecasting Release Forecast to Planning Segment Production Planning Forecast Consumption on Characteristic Level Sales Order Forecast Check Configuration Production Planning (CDP-Configuration) Save Sales Order Planned Order Creation Fig. 6.29. Process Chain for MTO with VC and Planning in Active Version The consequence of planning in the planning segment is that CTP can not be used any more as an ATP method (unless the resources are regarded as infinite). The planned orders in the planning segment would block the re- source capacity for the CTP-planned orders in the sales order segment. On the other hand, due to the use of the CDP-configuration scheme separate forecast elements are created for each characteristic value combination. This allows the use of the forecast check on characteristic level. The proc- ess steps for the scenario ‘make-to-order with variant configuration and planning in the active version’ are shown in figure 6.29. Compared to the scenario ‘make-to-order with variant configuration and planning in the in- active version’, the forecast consumption is done on finished product level and no data copy between planning versions is required. • Demand Planning with Characteristics The finished products are planned on characteristic value in Demand Plan- ning using the CBF functionality as in the previous scenario. • Forecast Release The forecast is now released to the active version and the consumption group needs to be maintained in the product master and in the release set- tings as described in chapter 6.3. An example for the log for the release of the forecast is shown in figure 6.30., 6.5 Scenario Description for Planning in Active Version 77 Fig. 6.30. Log for the Forecast Release with CDP-Configuration Relevance (APO) Note that for each locationproduct and characteristic value combination an entry is listed in the log. Figure 6.31 shows the representation of the fore- cast in the product view. Same Date Fig. 6.31. Forecast with CDP-Configuration Relevance (APO) With the CDP-configuration relevance the forecasts now have a valuation. The requirements are blank. • Production Planning Production planning is performed with the usual heuristics – no change in the set-up of the production planning is required for planning with the CDP-configuration relevance. Chapter 11 describes the properties in detail. Figure 6.32 shows the result of a production planning run in the product view., 78 6 Make-to-Order with VC and Demand Planning Fig. 6.32. Planned Orders with CDP-Configuration Relevance (APO) In the pegging overview (transaction /SAPAPO/PEG1) it can be easily checked that only elements with the same valuations are pegged. The production planning in the planning segment is required to create the receipt elements for the assembly groups resp. the components. The production planning in the planning segment might provide a feedback about the feasibility of the forecast – which is especially valuable for if the forecast check is used in the ATP. Another implication of the forecast check is that production planning is required for the sales orders. • Sales Order Creation For variant configuration make-to-order is usually applied, though make- to-stock would be possible in combination with the CDP-configuration scheme and the IS Mill solution as well. This case is described in chapter 8 about the scenario ‘configure-to-order with propagation’. The configura- tion of the sales order is performed as in the processes before. • ATP Check The most suitable way to perform the ATP check in this scenario is the fo- recast check. The multi-level ATP check does not support the CDP- configuration scheme and CTP is not an option because it might lead to a double capacity consumption in combination with planning in the planning segment. The use of the forecast check implies that the forecasted quanti- ties are feasible. The creation of the actual receipts needs to be done by the production planning run afterwards. As an alternative or in combination with the forecast check an allocation check is possible. Both options – the forecast check on characteristic level and the allocation check on charac- teristic level – are described in the next chapter., 6.5 Scenario Description for Planning in Active Version 79 6.5.2 System Configuration Determinants For variant configuration with planning in the active version the CDP- configuration scheme and the PDS master data are mandatory. On R/3-side the make-to-order strategy is common, though make-to-stock is possible with some limitations as well (see chapter 8). The class type 300 is re- quired on R/3-side. Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 6.33. Determinants for MTO with VC and Planning in the Active Version The VC-configuration scheme can not be used because this scenario relies on separate forecasts per characteristics value combination. The use of the PDS is mandatory for the integration of object dependencies from R/3 – if this is not required, the PPM could be used as well. 6.5.3 Forecast Check on Characteristic Level The forecast is released on detailed level using the consumption group and therefore the forecast check is performed on detailed level as well. To check the forecast for make-to-order in the planning segment the check mode needs to contain the assignment mode ‘2’ (assign customer require- ments to planning without assembly). • Rules-Based ATP with Forecast Check Generally it is possible to use rules-based ATP to perform a characteristic value substitution during the forecast check. There are however limitations regarding the use of rules-based ATP in combination with make-to-order as described in note 453921 – mainly due to the copying of the account as- signment to the sub-items. Especially for the configuration with the CDP- configuration scheme there are additional problems in setting the configu- ration for the sub-items in R/3. Since the forecast check is used within the described scenarios only for make-to-order this option is not explained any further., 80 6 Make-to-Order with VC and Demand Planning 6.5.4 Allocation Check on Characteristic Level The allocation check is usually performed on any field combination that is within the field object catalogue. Since the characteristics are freely de- fined, they are not part of the field object catalogue. To perform the alloca- tion check nevertheless on characteristic level, the CBF-table is used in- stead of the field object catalogue. In the product allocation group the checkbox ‘use configuration characteristics’ has to be selected to allow the assignment of the CBF-table to the allocation group, figure 6.34. Fig. 6.34. Product Allocation Group for Allocation Check with CBF (APO) To select the characteristic for the CBF-table via F4, the view with the values for the configuration objects (instead of the field catalogue) has to be selected. Fig. 6.35. Connection of the Allocation Group to the Planning Area (APO) For the connection to the planning area an internal key for the CBF-table is generated. This has to be assigned to the characteristic 9AMV_TAB of the planning area (resp. the planning object structure) as shown in figure 6.35., 6.5 Scenario Description for Planning in Active Version 81 6.5.5 Forecast Consumption on Characteristic Level For this scenario the consumption of a forecast in the planning segment by a sales order in the sales order segment is required. Therefore the check mode needs to have the assignment mode ‘assign customer requirements to planning without final assembly’ and the requirements strategy 30 has to be assigned. Figure 6.36 shows the forecast consumption on characteristic level. Forecast Consumption Overview on Characteristic Level Fig. 6.36. Forecast Consumption on Characteristic Level (APO), 7 Sales from Stock with Characteristics 7.1 Scenario Description 7.1.1 Process Chain In some cases the result of the production might differ depending on un- controllable factors. For example in chemical production processes (e.g. in the High Tech industry) there is only a statistical probability to get the de- sired output. Another example for the use of this functionality is the expi- ration of certain products. The idea for this scenario is a pure sales from stock, i.e. no planned re- ceipts are considered. This is a hard restriction, because planned receipts with CDP-configuration relevance in the make-to-stock segment are not integrated with R/3 (unless IS Mill is used). Figure 7.1 shows the process chain for this scenario. R/3-SD APO-ATP APO-DP APO-PP/DS Forecasting (w/o Characteristics) Production Forecast Release Production Planning Demand Planning Detailed Scheduling Sales Order Creation Sales Order Entry Production Execution Rules-Based ATP Batch Selection with Characteristics GR as Batch Save Sales Order Fig. 7.1. Process Chain for Sales from Stock, 84 7 Sales from Stock with Characteristics The demand planning and the production planning is performed without the consideration of characteristics. The characteristic values are only maintained at goods receipt for the batch. Neither planned orders, produc- tion orders nor forecasts have any configuration (unless IS Mill is used). • Sales Order Creation The sales order receives its requirements for the characteristic via the batch selection. Differing from the usual way of using the batch selection func- tionality, only the characteristics of the batches are used (and transferred to APO) and not the batch numbers. • ATP Check with Characteristics Within the ATP check only those elements are used for confirmation which have a configuration that matches the requirements of the sales or- der. Usually rules-based ATP is used to allow a substitution of the charac- teristic values (e.g. if multiple values or an interval are maintained as batch selection criteria). This way it is ensured that the sales order has only a single value as requirement (or none if it is not confirmed). The implication of planning without characteristics is that the scope of the ATP check in the sales order has to be restricted to stock categories. A characteristics value substitution using rules-based ATP is common. • Production Planning Production planning is performed without the consideration of characteris- tics. Neither the forecast nor the planned orders have any characteristics as shown in figure 7.2. Fig. 7.2. Product View for the Sales from Stock Scenario (APO) For interactively created planned orders it is possible to maintain a con- figuration. A pop-up requests the maintenance of a characteristic value, it is however possible to leave the valuation blank. If many batches with CDP-configuration exist, this might have implica- tions on the performance of the production planning run., 7.1 Scenario Description 85 • Pegging With the CDP-configuration scheme supplies are pegged to demands if the valuations of the supplies match the ‘requirements’ of the demands (or are the same as the valuation of the demand). This behaviour is described in chapter 11 in detail. Demands without ‘requirements’ or valuations – like the forecast in this scenario – are pegged to supplies of any valuation as shown in figure 7.3. Fig. 7.3. Pegging Overview – Valuated Batches vs. Non-Valuated Requirements Sales orders without confirmation are another type of demand without ‘re- quirements’. Since this scenario includes demands without ‘requirements’ and supplies without valuations, the pegging situation might be rather sub- optimal (as explained in chapter 11) up to not very helpful. The impact of this is that the alert monitor might not be very helpful either. • Process Variant: Characteristics in Planning In combination with the IS Mill solution it would be possible to use char- acteristic values in planning as well. In this case CBF would provide fore- casts with valuations that would lead to planned orders with valuations – and these are considered for the characteristics based ATP check. Still the scenario has to be a make-to-stock scenario (else neither stock nor planned receipts are considered). 7.1.2 System Configuration Determinants The material has to be assigned to a batch class (type 023) and the CDP- configuration relevance is required in APO. This scenario makes only sense for a make-to-stock strategy. Because this scenario is independent of production (resp. production is not taken into consideration), it does not matter whether the PDS or the PPM is used., 86 7 Sales from Stock with Characteristics Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 7.4. Determinants for Sales from Stock with Characteristics If the expiry date is used as a characteristic, all dates should be aggregated e.g. to monthly values (foods is not the target industry). Since each charac- teristic value requires an own time stream, it is necessary to keep the num- bers of characteristic values low in order to keep the performance accept- able. 7.2 Batch Selection in the Sales Order A different way (than configuration) to assign characteristic values to the sales order is using the batch selection functionality. For this the class type 023 for batches and the CDP-configuration relevance in APO are required. The values of the batch selection are transferred to APO as requirements and not as valuations (differing from the configuration). In APO the ATP check is performed on characteristic level. The characteristic values of the batches that were used for confirmation are used for the sub-items in the R/3 sales order. • Value Assignment Before ATP The characteristic value assignment is performed using the batch selection. To control that the batch search strategy is carried out before the ATP check and that the values are transferred to APO, an entry in the table TVBX_MAT_CBATP (in R/3) is necessary. This entry is set either via RFC- call if the characteristic view is maintained in APO (see chapter 7.5) or manually. • Search Strategy The value proposals for the sales order are maintained in the search strat- egy. Depending on the values of the key fields that are defined in the ac- cess sequence (e.g. customer and material), characteristic values are pro- posed. Figure 7.5 shows the maintenance of the batch search strategy with transaction VCH1., 7.2 Batch Selection in the Sales Order 87 F7 Fig. 7.5. Batch Search Strategy (R/3) It is either possible to maintain multiple values or an interval – not both (else there will be an error in the integration). Characteristic with optional values should not be used in order to avoid missing requirements. The search strategy is maintained per strategy type. Figure 7.6 provides an overview about the entities ‘access sequence’, ‘strategy type’ and ‘search procedure’ which are required for the determination of the charac- teristic values. Sales Area Sales Organisation Division Search Procedure

M

Channel N Strategy Type Access Sequence Fig. 7.6. Batch Search Strategy Determination (R/3) The search procedure is assigned to the sales area via the customising path Logistics – General Batch Management Batch Determination and Batch Check Batch Search Procedure Allocation and Activation Allo- cate SD Search Procedure/Activate Check. The access sequence uses the condition technique and is defined with the customising path, 88 7 Sales from Stock with Characteristics Logistics – General Batch Management Batch Determination and Batch Check Access Sequences Define Sales and Distribution Access Sequence For the strategy type the customising path is Logistics – General Batch Management Batch Determination and Batch Check Strategy Type Define Sales and Distribution Strategy Type and for the search procedure … Define Sales and Distribution Search Procedure. • Assignment of Characteristic Values The characteristic value assignment is carried out either in the background or as a dialogue depending on the settings in the detail of the search strat- egy as shown in figure 7.7. Fig. 7.7. Batch Determination in Background or Dialogue (R/3) The characteristic value assignment of the sales order is displayed with the icon as shown in figure 7.8. Fig. 7.8. Display Batch Characteristic Value Assignment (R/3), 7.3 ATP with Characteristics 89 Make sure that the characteristics are copied into the sales order (and are not just referenced by the strategy). The batch selection characteristic val- ues are transferred as requirements to the sales order in APO as figure 7.9 shows. Fig. 7.9. Sales Order with Requirements in APO In this case a discrete characteristic with multiple valuations without rules- based ATP was used. 7.3 ATP with Characteristics 7.3.1 Characteristic View As a standard the ATP check (as product check) is carried out on product and location level, optionally on sublocation (i.e. storage location) and ver- sion (i.e. batch) level as well. For the ATP with characteristics the ATP check is additionally performed on the level of the characteristic value combination which might have a negative effect on performance. Fig. 7.10. ATP Time Series with Characteristics (APO), 90 7 Sales from Stock with Characteristics The elements in the ATP time series which have ATP relevant characteris- tic valuations are displayed with the green triangle as shown in figure 7.10. ATP with characteristics requires the CDP-configuration relevance and class type 023 for batch classification. The ATP check with characteristics including the rules-based characteristic substitution is only released for SCM 4.1. • Characteristic View ATP on characteristics level is activated by the creation of the characteris- tic view per product and characteristic. Not all characteristics of the class need to be relevant for ATP. The characteristic view is maintained with transaction /SAPAPO/ATPCH01, figure 7.11. Fig. 7.11. Characteristic View (APO) • ATP Check with Characteristics With the activation of the characteristic view the ATP check is performed on characteristic value level, i.e. only the receipts with the same character- istic values are used for the confirmation of the requirement (i.e. sales or- der). Receipts without characteristic values are regarded as inappropriate for a requirement with characteristic values. If accidentally no characteristic values are assigned to a requirement it is possible to get an overconfirmation, since the ATP check is carried out on locationproduct, sublocation, batch and characteristic level independently. 7.3.2 Rules-Based ATP with Characteristic Substitution Within the rules for rules-based ATP (transaction /SAPAPO/RBA04) it is possible to maintain a characteristic substitution procedure (i.e. a list of characteristic values which can be substituted) as shown in figure 7.12., 7.3 ATP with Characteristics 91 Rule Characteristic Substitution Procedure Fig. 7.12. Rules Maintenance for Characteristic Substitution (APO) The logic for the substitution is the same as with location and product sub- stitution as long as the requirement has only a single characteristic valua- tion (i.e. no multiple values for the same characteristic). • ATP Check with Characteristics The ATP check on characteristic level is the special feature within this process. In most cases not only one characteristic value is suitable for a re- quest, therefore rules-based ATP offers the possibility to substitute charac- teristic values. The result of a rules-based ATP check with characteristic substitution is displayed in figure 7.13. Characteristic Values Fig. 7.13. Result of a Rules-Based ATP Check (APO) In this case the request for 20 pieces was first checked with the lower qual- ity B and for the missing with the higher quality A. Figure 7.14 shows the result of the ATP check in the sales order on R/3-side., 92 7 Sales from Stock with Characteristics Display Selection Criteria Fig. 7.14. Sales Order in R/3 with Selection Criteria The two different characteristic values for the confirmation are represented by sub-items. Each of the sub-items has its own configuration (i.e. re- quirement) according to the result of the ATP check. The item itself has still the initial value resp. values as determined by the selection criteria. In APO only the sub-items are displayed with their confirmed quantities and requirements, figure 7.15. Fig. 7.15. Sales Order in APO • Multiple Characteristics If multiple characteristics are used the time series are created per charac- teristic value combination. The batch selection criteria is however still maintained per characteristic, and so is the characteristic substitution pro- cedure within the rules for rules-based ATP., 7.4 Delivery with Characteristics 93 Rule / Characteristic Substitution Procedure Result of the Rules-Based ATP Fig. 7.16. Rules-Based ATP with Multiple Characteristic Substitutions. The consequence is that a controlled substitution of characteristic combi- nations (e.g. A/BRIGHT or B/DARK) is not possible. 7.4 Delivery with Characteristics The valuations for the sub-items which have been determined in the rules- based ATP check for the sales order are used for the ATP check in the de- livery as well. Since the ATP check in the sales order was already re- stricted to inventories the probability of a missing availability is compara- tively low. In case of a shortage it is however not possible to apply the rules-based ATP in the delivery in order to change the characteristic valua- tions. For the delivery the same batch search strategy is used as for the sales order. Figure 7.17 shows the delivery with the assigned batches. Fig. 7.17. Delivery (R/3) The delivery is transferred to APO, as shown in figure 7.18., 94 7 Sales from Stock with Characteristics Fig. 7.18. Delivery in APO For each batch split a separate line with its own requirement is created., 8 Configure-to-Order with Propagation 8.1 Scenario Description 8.1.1 Process Chain

The specific feature of the scenario ‘configure-to-order with propagation’

is that it is a make-to-order process for the finished product and a make-to- stock process for the assembly groups. Additionally the assembly groups have characteristics which are propagated from the finished product and are usually handled in batches. Figure 8.1 shows the process chain for the scenario ‘configure-to-order with propagation’. R/3-SD APO-ATP APO-PP/DS Sales Order Taking Block Planning Sales Order Entry Block Planning for Finished Sales Order CTP Request Planned Order Configuration Determination Creation (Finished) Optional Forecast Generation for Semi-Finished Optional Forecast Consumption (Semi-Finished) ? Optional Planned Order Production Planning Save Sales Order Confirmation Combination Creation (Semi-Fin.) for Semi-Finished Detailed Scheduling Optional Steps Apply Only If Semi-Finished Product Does Not Have Any Batch Characteristics

Fig. 8.1. Process Chain for Configure-to-Order with Propagation

, 96 8 Configure-to-Order with Propagation This scenario is mainly used for the mill industries. The mill industries is a collective term for the metal, the paper, the wood, the textile and the foil industries. The peculiarity in those areas is that the material flow is diver- gent from the key material – e.g. the steel coil or the paper reel. These key materials have usually a long lead time and are handled in (huge) batches. The finished products are usually produced as make-to-order, and the availability is frequently restricted by the inventory of the key material and the capacity of the key resources. The key material is often produced in- house but might be externally procured as well. In many of the industries – e.g. the steel industry – the key materials have batch characteristics to de- scribe their grade, thickness, length and other properties which are usually only known at the time of the goods receipt. An example for these are steel coils. On the other hand the requirements for the key component cover a range of properties and therefore it is usually possible to use key compo- nents of different properties for the same finished product. In this case the ‘requirements’ (as a property within CDP, see also chapter 11) of the de- pendent demand are applied to select an appropriate batch. Depending whether the components have batch characteristics or not, the optional steps can be applied. For components with batch characteris- tics they do not make any sense. The scenario relies on the CTP check in the sales order which triggers the production. Often block planning is used to provide a rough cut scheduling for the planned orders, but it is neither mandatory for this scenario to use block planning nor is block planning re- stricted to this scenario. In this case we assume that the block planning takes place on the finished product level – depending on the production re- quirements it might be any other level as well. A subsequent detailed scheduling step is mandatory. For the integration of the valuated planned and production orders into the make-to-stock segment for the key components the IS Mill solution is required on R/3 side. • Assignment of Characteristic Values The configuration of the sales order is performed as described in chapter 5. For most cases this is sufficient, for others (e.g. cables) additionally a batch selection as described in chapter 7 can be carried out if requirements are needed for the sales order – e.g. because the consideration of appropri- ate batches takes place already on finished product level. Another case is the modelling of tolerances (e.g. for the thickness of steel sheets) as re- quirements. In these cases the variant configuration must be compatible with the batch selection (i.e. the valuation must be within the require- ments) to avoid errors in further processing., 8.1 Scenario Description 97 • Production Planning & Characteristic Propagation The most common characteristic-related property of the planned order for the finished product is the propagation of characteristic values from the finished product to the dependent demand. For the PPMs this functionality has to be configured explicitely, for the PDS the propagation of a charac- teristic value is performed as a default. If the characteristic value has to be manipulated (e.g. if the finished product is ‘RED’, the component should be ‘BLUE’), object dependencies as described in chapter 5 have to be used. • Requirements for the Dependent Demand with the PDS For some business cases it is necessary to use requirements for the charac- teristic values in the dependent demand (e.g. cables because the length is modelled as a characteristic). Using the PPM it is possible to define re- quirements for characteristic values. With APO 4.1 this is not possible within the PDS. There is however a way to model requirements when planning with the PDS as well by defining the requirements for the planned order directly (instead of defining it in the PDS). For this purpose the BAdI /SAPAPO/RRP_SRC_EXIT for the planned order creation has to be used. The method RRP_USEX_PLORD_CREATE allows the definition and assignment of the requirements. If planning is performed with requirements, specific production plan- ning heuristics are required. The normal production planning heuristic does not provide sufficient results. • Optional: Planning for Semi-Finished Using Block Planning In some cases – e.g. for paper industry – the key component (the pulp) does not necessarily have any batch characteristics. In combination with block planning it is possible to create independent requirements for these components based on the blocks. The optional steps cover this case. 8.1.2 System Configuration Determinants As soon as batch characteristic are used, the CDP-class system is manda- tory. Since APO was used for mill industries before the PDS master data object existed, both PPM and PDS can be used in this case. It is however recommended to use the PDS for new implementations because the PPM will not be developed any further. If the PPM is used, the logic for the processing of the characteristic values has to be maintained within the PPMs on APO side. Figure 8.2 provides an overview about the settings., 98 8 Configure-to-Order with Propagation Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 8.2. Determinants for Configure-to-Order with Propagation Usually a make-to-order requirements strategy is used for the finished product and make-to-stock for the key components. In some cases a make- to-stock strategy is used for the finished product as well, and the sales or- der is connected with the receipt by compatible characteristic values. Since the CDP-configuration scheme is used, all elements have their own valua- tions. For the assignment of the valuation to the sales order the finished product must be configurable and classes of both types (300 and 023) have to be assigned to the material master in R/3. Both classes must contain the same APO relevant characteristics. Figure 8.3 shows the product master in APO with the classes of the types 300 and 400. Fig. 8.3. Class Assignment for the Finished Product (APO) Another important prerequisite is the use of the IS Mill solution on R/3- side in order to allow planned resp. production orders for the semi-finished products with valuation but without a sales order assignment., 8.2 Characteristic Propagation within the PPM 99 8.2 Characteristic Propagation within the PPM The propagation of characteristics is possible with both the PDS and the PPM. For the PDS this is already explained in chapter 5. Historically the characteristic propagation as described in this chapter was developed and used for mill industries. Currently this is the only scenario where charac- teristics propagation within the PPM is used – the use of the PDS is rec- ommended for new implementations. For new implementations the PDS normally covers the requirements for the scenario with the advantage of improved R/3-integration. Characteristic propagation in the PPM is only possible for the CDP-configuration relevance. Due to the divergent material flow, in the mill industries stock is usually kept for the raw material (e.g. the steel coil or the paper reel), which is used to produce many different products. The finished product is usually produced for the individual sales order. Therefore one of the key functions is to propagate the valuations of the sales order to the dependent demand of the planned resp. the production order. The settings for the propagation of the valuations are maintained within the PPM with the transaction /SAPAPO/SCC05. The maintenance for the characteristic propagation settings is activated with the menu path plan characteristics propagation use and changed with the menu path goto characteristic propagation: overview as shown in figure 8.4. Activate Characteristic Propagation Change Characteristic Propagation Fig. 8.4. Activate and Change the Characteristic Propagation in the PPM (APO) The entry screen for the characteristic propagation maintenance provides an overview about the input and output products and which classes are as- signed to the products or to the operation, the activity or the plan itself., 100 8 Configure-to-Order with Propagation Fig. 8.5. Characteristic Propagation Maintenance Screen in the PPM (APO) The classification is set by default for all classified products and block- planned activities (the product and the resource master are checked for rea- sonable proposals). If components are added later, the class proposal has to be triggered interactively within the characteristic propagation mainte- nance screen with the menu path propagation propose classes. It is possi- ble to use the classification additionally on plan and on operation level to change settings according to the valuations. It is possible to assign a class interactively to any of these levels. In this case the flag ‘only rules/parameters’ is set automatically. This flag controls that the characteristic values are only used for the rules but no characteris- tic values will be adopted to the planned order. Therefore, if a class as- signment is expected but does not show, check the master data instead of maintaining it interactively. For the classified output product the standard valuation and the standard requirements are defined. The standard valuation is used if the demand does not have any valuation, and the standard requirement of the output product defines for which valuation the PPM might be used. Figure 8.6 shows the principle of the characteristic propagation from the valuation of the output node to the valuation of the input node. Sales Order Valuation ‘Green’ Propagation Planned Order Valuation ‘Green’ Planned Order ‘Use Valuation’ Green Propagation within the Order ‘Set Valuation’ Dependent Demand Valuation ‘Green’ Fig. 8.6. Propagation of Valuations, 8.2 Characteristic Propagation within the PPM 101 For the propagation of the valuations of the output product to the compo- nents, on the output product level the flag ‘use valuation’ has to be set in the ‘valuations’-view and on the component level the according flag ‘set valuation’ has to be set as shown in figure 8.7. Output Product Input Product Fig. 8.7. Required Setting for Characteristic Propagation (APO) The standard requirements for the output product are used as a selection criterion for the PPM – i.e. to cover a demand, the valuation of the demand must match the requirements of the output product. In this case the valuation is propagated with the same value. It is how- ever also possible to apply a more complex logic using macros. The re- quirements for the component are maintained as standard requirements. It is also possible to change the requirements depending on the valuation of the output product, e.g. with the use of macros. These macros have to be created in APO, no integration with the object dependencies from R/3 is possible. Requirements set in macros must be kept compatible with the valuations without system support. This functionality originates from APO 3.x when requirements were mandatory for pegging. With APO 4.1 this should only be used in exceptional cases. • Characteristics as Selection Criteria Characteristic value can be applied in the PPM to perform a selection of the component, the mode (i.e. the resource) and the set-up group. The logic is quite similar for all of these selections. Therefore only the selection of the set-up group is explained in detail. In the example shown in figure 8.8 the set-up group XX1 is selected if the value of the operation is ‘GREEN’ and the set-up group XX2 if the value is ‘RED’., 102 8 Configure-to-Order with Propagation Input Component Activity Operation Fig. 8.8. Characteristic Based Set-Up Group Determination (APO) The logic for the selection is maintained as rules on operation level. The rule itself is an identifier which combines the object to which it relates (in this case the activity of the type ‘S’) with the set-up group. The selection conditions are maintained as ‘requirements’ in the details of the rule. The selection of components and modes is performed analogously on the input material level resp. the activity level. 8.3 Block Planning 8.3.1 Motivation for Block Planning The idea of block planning is to reserve capacity on key resources for or- ders with certain properties. Especially in the mill industries the valuation of the output product might have a significant impact on the production – for instance for the coating of steel the type of the coating is substantial for the production planning, since the effort for the set-up between different coatings is considerable. Other examples are the paper production that must use up a large amount of pulp or steel processing where huge quanti- ties have to be processed within a short time (e.g. due to solidification)., 8.3 Block Planning 103 The sequence of the pulp or steel qualities – the cycles – can be roughly planned or is known from experience. Since the sales orders and – in the make-to-order environment – the planned orders are not yet available, with block planning the resource capacity is reserved for certain characteristic values of the operations. The length of these blocks are defined by the planner, often using a reference cycle and adjusted interactively. 8.3.2 Block Definition in the Resource For block planning a class of the type 400 (R/3: 300 or 023) in APO is re- quired. This class needs to contain the relevant characteristic or character- istics and is assigned to the resource master (model independently). Within R/3 only the characteristics are required, the class can be defined in APO if no meaningful class can be integrated. The blocks themselves are defined version dependently, figure 8.9. Fig. 8.9. Block Definition in the Resource (APO) The intervals are the blocks and define the requirements of the resource for a time interval. The block basis definitions are the identifiers for the re- quirement of a block. These are either directly or via a reference cycle as- signed to the intervals as shown in figure 8.10. The use of reference cycles is common but optional., 104 8 Configure-to-Order with Propagation Block Basis Definition Requirement Reference Cycle Interval Fig. 8.10. Entities for the Block Maintenance The block basis definitions including the requirements and the reference cycles are resource independent, i.e. they can be used for all resources with the same class. Figure 8.11 shows the screens for the maintenance of the rules resp. block basis definition and the reference cycle. Block Basis Definition Requirements Reference Cycle Fig. 8.11. Block Basis Definition and Reference Cycle (APO) The blocks are defined by the assignment of the block basis definitions and/or the reference cycles to the intervals, figure 8.12., 8.3 Block Planning 105 Insert Reference Cycle Insert Rule Move Block Additional Block Properties Fig. 8.12. Block Definition via Intervals (APO) In the ‘intervals’-view additional properties of the blocks are maintained, e.g. a fixing of the block, whether the scheduling has to be performed at the block limits and block tolerances (i.e. how much an activity might ex- ceed the block limits). The fixing of the block causes that no planning into the block is allowed and the activities inside the block are not scheduled any more (though the activities do not have the status ‘fixed’). The block sizes are changed either by overwriting the dates of the blocks or via drag and drop in the graphical overview (see figure 8.12). Block planning can not be used for secondary resources. • Visualisation of the Blocks in the Planning Board The blocks are displayed per default as thin lines within the resource. With a set-get parameter in the user settings (/SAPAPO/CDPS_BLOCK = 1) it is possible to change the display of the blocks to boxes. The colour of these lines is defined in the maintenance for the rules resp. block basic defini- tions. Fig. 8.13. Block Planning in the Planning Board (APO), 106 8 Configure-to-Order with Propagation With right mouse click on the blocks it is possible to branch into the block maintenance. An extended block maintenance is possible by calling the heuristic SAP_CDPBP_03 from the planning board. Fig. 8.14. Extended Block Planning (APO) As a prerequisite the heuristic SAP_CDPBP_03 has to be assigned to the heuristic profile of the planning board. 8.3.3 Activity Valuation The principle of the block planning functionality is to define requirements for the blocks on the resource and to valuate the activities. The definition of the requirements for the blocks is quite independent of the scenario, but how to valuate the activities depends on the master data (PDS or PPM) and whether the product is configurable or not. For the use of PPMs the CDP- configuration scheme is required, for PDS both VC- and CDP-configu- ration scheme are supported (with APO 4.1). • Valuation of Activities via Characteristic Propagation in the PPM For a configurable product we do have a valuation at the demand node (usually the sales order) which will cause a valuation of the supply node of the planned order. To propagate the valuation of the output product to the activity, the activity has to be classified in the characteristic propagation settings within the PPM and the flag ‘set valuation’ must be set, fig- ure 8.15. The class of the activity has to contain the characteristic that is used for block planning. Make sure that the indicator for ‘only rules/ pa- rameters’ is initial (otherwise the activity will not be valuated)., 8.3 Block Planning 107 Fig. 8.15. Valuation of the Activity (APO) Again it is possible to apply a more complex logic for the propagation us- ing macros. The rules (the button in the top right area) can be used for the selection of modes within the activity, e.g. to exclude modes for certain values of the characteristic. All of these definitions have to be maintained manually in APO. If the demand node does not have any valuation – i.e. if the product of the master output is not CDP-configuration relevant – it is possible to de- fine a standard valuation for the activity. In this case block planning is not carried out according to the configuration resp. the valuation of the fin- ished product but according to the product itself (or to be more precise: the PPM). For simple cases with fix valuation (i.e. no characteristic propaga- tion, just standard valuation) it is possible to set these values during the CIF transfer as described in note 495825. • Valuation of Activities via Object Dependencies With the PDS the activity is valuated per default with the characteristic values of the order resp. the output node if the resource is relevant for block planning. Neither class assignment nor object dependencies are re- quired for this. If a more complex logic is required – e.g. to set the value of the block planning characteristic depending on the value of another charac- teristic – object dependencies (procedures) have to be used. These object dependencies must be assigned to the dependency group SAPAPOACT. This dependency group is used to control that the object dependency is as- signed to the activity (and not to the operation or the mode, see chapter 5). As an example the value of the characteristic CBP_COLOUR (which is used for block planning) for the activity is set to ‘BLACK’ if the value of the quality for the order is ‘A’: $SELF.CDP_COLOUR = ‘BLACK’ if $ROOT.CBP_QUALITY = ‘A’., 108 8 Configure-to-Order with Propagation The object dependency is assigned to the operation within the routing. The corresponding PDS has the object dependency assigned to the activity. Fig. 8.16. Object Dependencies in the PDS (APO) If block planning is to be used for non-configurable products, a fix valua- tion has to be assigned to the operation of the routing in R/3. For this pur- pose a class of type 018 has to be created and assigned to the operation (with the path extras classification operations within the routing) as shown in figure 8.17. Fig. 8.17. Valuation of the Operation for Non-Configurable Materials (R/3) The class has to contain the characteristic that is used for block planning. No additional class transfer to APO is required. However, to transfer the valuation to APO the method USE_OPERATION_CLASSIFICATION of the BAdI CUSLNTRTO_ADDIN on R/3-side has to be implemented (by adding the coding line EV_TRUE = ‘X’.) and activated. The valuation of the operation is displayed in the PDS with transaction /SAPAPO/CURTO_SIMU as shown in figure 8.18 in the details of the block planning indicator., 8.3 Block Planning 109 Fig. 8.18. Valuation of the Activity for Non-Configurable Products in the PDS (APO) Note that the block planning class is the class of the resource in APO and not the class of the routing in R/3. The block planning settings for the re- source have to be maintained before integrating the PDS to APO. Per default only the activities of the type ‘production’ are valuated. If it is required to valuate the set-up activities as well, this must be done via BAdI. For this purpose the method BLOCK_PLANNING_RELEVANCE of the BAdI /SAPAPO/CURTO_CREATE has to be implemented. 8.3.3 Block Planning Functions Block planning was originally developed for the need of the mill industries but is used by other industries as well – both for configurable and for non- configurable products. Note 528189 describes the block planning specific scheduling options (planning at the block limits, fixing of blocks) and gives additional information about the properties and limitations. 8.3.4 Assembly Planning Based on Block Definitions Usually the production strategy in the mill industries is only make-to-order for the finished product and make-to-stock for the key assembly groups. The planning for the assembly groups might be performed consumption based or via forecast. The functionality described in this chapter offers a help for the generation of the forecast. The generation of forecasts based on the block definitions in block plan- ning makes only sense if the input material is neither configurable nor sub- ject to planning relevant batch characteristics. The forecast is generated for the input material of the products which use the block planning. The basis for this is the PDS resp. the PPM of the product which is produced on the resource. Either one product per resource or one product per block is specified to provide the relation between productive time and base quan-, 110 8 Configure-to-Order with Propagation tity. This specification is maintained in the ‘planned independent require- ment’-view of the block definitions within the resource, figure 8.19. Fig. 8.19. Settings for the Forecast Generation Within the Block of the Resource (APO) The quantity for the forecast is calculated using the PDS resp. PPM of the specified product. Note that in the PDS selection conditions resp. in the PPM rules have to exist for all the requirement values of the blocks else the forecast generation terminates with an error. For example, if one block has a requirement that includes ‘RED’ and ‘ORANGE’, both for ‘RED’ and ‘ORANGE’ selection conditions resp. rules have to exist. The forecast gen- eration itself is called with the transaction /SAPAPO/BLRG01. Figure 8.20 shows the result screen. Fig. 8.20. Forecast Generation from Blocks (APO), 8.4 Goods Receipt with Batch Characteristics 111 The forecast generation does not support any characteristics. Though it is possible to enhance the forecast with characteristics using the BAdI /SAPAPO/BLRG_BADI_EX, the forecast consumption for dependent de- mand does not consider characteristics in any way. 8.4 Goods Receipt with Batch Characteristics If a batch class is assigned to the material master and the checkbox for batch management in the ‘work scheduling’-view is selected, batch num- bers are created for each goods receipt and the valuation of the batch is re- quested via pop-up as shown in figure 8.21. Automated Batch Number Assignment Pop-Up Manual Characteristic Value Assignment Fig. 8.21. Batch Valuation at Goods Receipt (R/3) The batches and their valuations are transferred to APO (if the transfer of batches is active in the CIF model) as shown in the product view in fig- ure 8.22. Fig. 8.22. Batches with Characteristics in APO, 112 8 Configure-to-Order with Propagation For the use of batch characteristics in APO the CDP-configuration rele- vance is mandatory. The ‘version’-view of the product view provides a more detailed over- view about the batches and their valuations, figure 8.23. Fig. 8.23. Version View of the Product View (APO) • Deactivation of Batches Since batches are a different entity than inventory and the batches still ex- ist even when the inventory has been issued long ago, it is necessary to de- activate batches that are not needed any more with the report RVBDEACT (or to archive the batches in R/3) in order to ensure system performance., 9 Planning with Shelf Life 9.1 Scenario Description 9.1.1 Process Chain

The planning with shelf life is mainly relevant for industries with perish-

able products and a make-to-stock production – e.g. foods and pharma. In these cases planning has to ensure that the products still have a certain shelf life at their demand date. R/3-SD APO-DP APO-ATP APO-PP/DS Demand Planning Forecasting Production Forecast Release Production Planning Sales Order Taking Exception Handling for Shelf Life Sales Order Entry Planned Orders Detailed Scheduling Production Execution ATP Check Batch Selection GR as Batch(w/o Shelf Life) Save Sales Order Fig. 9.1. Process Chain for Planning with Shelf Life, 114 9 Planning with Shelf Life Since shelf life planning in APO is limited to the locationproduct-level, in a straightforward model the shelf life limitations will be lost with each production step and each stock transfer. Nevertheless it is often possible to limit the shelf life planning problem to one critical locationproduct and use customer reports to identify possible problems on other levels and treat these as exceptions. In the following we assume that the shelf life problem for planning is limited to one locationproduct only. Figure 9.1 shows a sample process for the shelf life scenario. Demand planning is performed assuming that the shelf life requirements will be met – i.e. without any consideration of the shelf life. The main spe- cific of the process is that the shelf life restrictions are considered in pro- duction planning and scheduling. The sales order might transfer a customer specific shelf life requirement, but the shelf life is not considered in the ATP check. • ATP Check It is not possible with standard ATP methods to take the shelf life informa- tion into account for the ATP check – neither with the ATP time series nor with the ATP check within the pegging network (the setting ‘characteristic evaluation’ in the check mode). A possible solution on customer project basis might be to enhance the ABAP class for CTP and to perform the ATP check within the pegging network. Other ATP functions as ‘full delivery’ might violate the con- straints nevertheless. • Alternative Approach: Reporting In some cases the problem of expiring inventories can be solved as well by a report to list the batches which are expired or will expire soon. If the shelf life is long compared to the planning horizon, this approach should be considered as an alternative. 9.1.2 System Configuration Determinants The prerequisite for the use of shelf life are batch characteristics (type 023) in R/3. The shelf life functionality does not require any configuration rele- vance for the product. Therefore this scenario is independent of the con- figuration scheme, but shelf life planning is not compatible with the VC- configuration relevance (i.e. if the VC-configuration scheme is used, the product must not be configurable). The indicator for shelf life planning has to be set in the product master as well., 9.1 Scenario Description 115 Shelf life planning is usually only applied in a make-to-stock environment, though it would be technically possible to use it with a make-to-order strategy as well. The shelf life functionality is independent of the master data object – i.e. both the PDS and the PPM can be used. Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 9.2. Determinants for Shelf Life Planning 9.1.3 Shelf Life Settings in the Material Master In the material master the indicator for batch management has to be set and the shelf life data in the ‘plant data / storage 1’ view needs to be main- tained. R/3: Plant Data/ Storage 1-View APO: Properties-View (Location Independent) Fig. 9.3. Material Master Settings for Shelf Life The value for the total shelf life is transferred from the R/3 material master to the APO product master and triggers the setting for ‘planning with shelf life’. Note that this information is location independent in APO. The re- quired minimum shelf life and – if needed – the required maximum shelf life have to be maintained manually resp. via user-exit., 116 9 Planning with Shelf Life Since APO 4.0 it is possible to plan with location dependent shelf life data as well. The flag ‘planning with shelf life’ triggers that shelf life data is displayed on the location dependent demand view of the product, fig- ure 9.4. Fig. 9.4. Location Dependent Shelf Life in the ‘Demand’-View of the Product Master Additionally the classes and characteristics for shelf life have to be as- signed to the material master in R/3, see chapter 9.3. The assignment is neither required in APO nor is it transferred to the product master. Characteristics are only required for the integration of the batch infor- mation from R/3. Apart from this the shelf life planning is not based on characteristics (neither technically nor in display – the shelf life informa- tion is stored differently in live cache and the products are not configura- tion relevant, though a batch class is assigned in R/3. 9.2 Shelf Life Functionality Especially in foods industries, but in other businesses as well (pharmaceu- tical industries for instance) shelf life is an issue which needs to be consid- ered in planning. The shelf life functionality in APO covers two aspects of this, the shelf life itself (i.e. the expiration) and the maturity. This informa- tion is used in pegging, where the shelf life and the maturity of a supply element are checked with the required shelf life and – optionally – with the required maximum shelf life of a demand element. Figure 9.5 illustrates this logic. The additional constraints for pegging defined by the shelf life function are Rqmt. Date ≥ Avail. Date + Maturity (9.1) Rqmt. Date + Req. Max Shelf Life ≥ Avail. Date + Shelf Life (9.2) Rqmt. Date + Req. Min Shelf Life ≤ Avail. Date + Shelf Life (9.3) In the first case in figure 9.5 the constraint in formula 9.1 (and 9.2) is vio- lated, in case 2 constraint 9.2 and in case 4 constraint 9.3., 9.2 Shelf Life Functionality 117 1 Req. Min. Req. Max. Order Shelf Life Shelf Life Order Maturity Shelf Life Time 2 Req. Min. Req. Max. Order Shelf Life Shelf Life Order Maturity Shelf Life Time 3 Req. Min. Req. Max. Order Shelf Life Shelf Life Peggin g Order Maturity Shelf Life Time 4 Req. Min. Req. Max. Order Shelf Life Shelf Life Order Maturity Shelf Life Time Fig. 9.5. Shelf Life Dates

Shelf life is activated by setting the according flag in the ‘attributes’ view

of the product master. The durations for maturity, shelf life, required shelf life and required maximum shelf life are maintained in the ‘attributes’ view as described in 9.1.3.

The information regarding shelf life is displayed in the product view, if

shelf life is activated. Figure 9.6 shows the product view for an example as in figure 9.5. Receipt Maturity Shelf Life Requirement Required Min. Shelf Life Required Max. Shelf Life Fig. 9.6. Display of Shelf Life Information in Product View (APO), 118 9 Planning with Shelf Life The difference between shelf life and a maximum pegging constraint is that the goods receipt date of the batches is taken into account, i.e. the ma- turity and the shelf life are adjusted with proceeding time, figure 9.7. Time 1 Req. Min. Time 2 Req. Min. Order Shelf Life Order Shelf Life Stock Stock Maturity Shelf Life Shelf Life Fig. 9.7. Adjustment of the Shelf Life with Time 9.3 Characteristics for Shelf Life Characteristics and object dependencies are mainly required to store the shelf life and the maturity of batches. A second usage is the transfer of cus- tomer specific required shelf life. • Characteristics and Classes for Shelf Life For shelf life a class of type 023 is required. As described in the notes 391018 and 751392 this class should contain the following standard charac- teristics • LOBM_VERAB • LOBM_VFDAT • LOBM_HSDAT. The characteristic LOBM_HSDAT has to be created according to note 394229 as shown in figure 9.8 (for R/3 4.7 create only the characteristic and do not carry out any of the other actions). Fig. 9.8. Create Characteristic LOBM_HSDAT (R/3), 9.3 Characteristics for Shelf Life 119 Additionally the new characteristics • LOBM_APO_SL_MIN: Min. shelf life/maturity in sec., type NUM 15,3 • LOBM_APO_SL_MAX: Max. shelf life/maturity in sec., type NUM 15,3 • LOBM_APO_SL_UTC: Reference time stamp, type NUM 15,0. have to be created and assigned to the class. These characteristics should have the flag ‘not ready for input’ and ‘no display’. The naming conven- tion of the new characteristics is essential. These characteristics are either explicitely created or with the report Z_ADD_LOBM_APO_SL_CHARS as in note 602174. After the report has run, the characteristics have to be up- dated from client 000 with transaction BMSM. If configured materials are used together with shelf life, additionally the characteristics SL_MIN_DAYS, SL_MIN_DAYS2, SL_MAX_DAYS and SL_MAX_DAYS2 have to be created and assigned to the variant configura- tion class (type 300) to transfer the shelf life requirements as explained in note 391018. Only the new characteristics have to be transferred to APO, e.g. by assigning the organisational area only to the new characteristics only as shown in figure 9.9. Fig. 9.9. Class for Shelf Life (R/3) For some of the characteristics object dependencies are required as de- scribed in note 391018. These are created either interactively or with trans- action BMSM as well – provided that no object dependencies exist already. The old object dependencies are deleted either manually or with the report Z_DEL_SL_OBJECT_DEPS as described in note 604757. • Transfer of Characteristics and Classes to APO Different from the other cases, the characteristics LOBM_APO_SL_MIN, LOBM_APO_SL_MAX and LOBM_APO_SL_UTC are only assigned to the class type 230 in APO and not to the class type 400 (nor is it possible to assign these characteristics to the class type 400). Differing from R/3, no, 120 9 Planning with Shelf Life class is assigned to the product and the configuration relevance indicator is set to ‘not relevant for configuration’. Make sure that the indicator for material configuration is not set in the ‘basic data 2’-view (unless configuration is used for another purpose – in this case the VC-configuration scheme is not compatible with the shelf life functionality). • Characteristic Customising in APO For the display of the shelf life relevant data of the batches it is necessary that the organisational area of the characteristics LOBM_APO_SL_MIN, LOBM_APO_SL_MAX and LOBM_APO_SL_UTC is assigned to the class types 230 (all organisational areas should be maintained in APO). This as- signment is possible with transaction O1CL analogous as described in chapter 3 and shown in figure 9.10. Fig. 9.10. Assignment of Organisational Areas to Class Type for Batches (APO) • Re-Evaluation of Batches If batches already exist before the configuration of the shelf life settings (resp. before the change of the classes), the batches need to be re- evaluated. This is possible with the report ZVBRELB as described in note 78235. 9.4 Customer Specific Shelf Life Requirements Without batch selection the ‘minimum required shelf life date’ is calcu- lated as the confirmed date plus the required minimum shelf life as main- tained in the product master. The batch selection is an optional step and only necessary if customer specific required minimum shelf life exist or if the shelf life of the batches has to be checked for the deliveries. For the maintenance of a customer, 9.4 Customer Specific Shelf Life Requirements 121 specific required shelf life value the class SHELF_LIFE_SEL of the type 023 for batches has to be created additionally as described in note 483576, figure 9.11. Fig. 9.11. Class SHELF_LIFE_SEL for the Calculation of the Required Shelf Life (R/3) It is not required to assign the class to the material, nor must any of these characteristics be transferred to APO. For the dependencies the function LOBM_UBD and the object depend- encies LOBM_UBD have to be created as described in note 33396. The function is created with transaction CU65 and shown in figure 9.12: Fig. 9.12. Object Dependency for the Transfer of Customer Specific Requirements (R/3), 122 9 Planning with Shelf Life The object dependency itself is created with transaction CU01 and has to contain the coding as shown in figure 9.13. Fig. 9.13. Object Dependency LOBM_UBD (R/3) For the assignment of the object dependency LOBM_UBD to the character- istic LOBM_RLZ it is necessary to execute the transaction BMSM. The re- sult can be checked in transaction CT04 for the maintenance of characteris- tics, figure 9.14. Fig. 9.14. Assignment of Object Dependency to Characteristic (R/3) The search strategy is created for the strategy type SD01 with the key cus- tomer and material as described in chapter 5. The class SHELF_LIFE_SEL is assigned, figure 9.15., 9.4 Customer Specific Shelf Life Requirements 123 Fig. 9.15. Search Strategy for the Transfer of Customer Specific Requirements (R/3) The value for the characteristic LOBM_RLZ (remaining shelf life for batch) represents the customer specific shelf life requirement. If the configuration is done correctly, the required minimum expiration date is calculated and stored in the sales order as shown in figure 9.16. Fig. 9.16. Calculation of the Customer Specific Minimum Expiration Date (APO) To transfer the required shelf life date to APO the user exit CIFSLS03 has to be activated (transaction SMOD) with the example coding as provided in note 483576, figure 9.17., 124 9 Planning with Shelf Life Enhancement / User-Exit CIFSLS03 Include ZXCIFU16 Function Module EXIT_SAPLSLS_002 Fig. 9.17. User-Exit CIFSLS03 for the Transfer of the Required Shelf Life (R/3) Additionally the user-exit has to be assigned to a customer project with transaction CMOD and the project has to be activated. 9.5 Production Planning and Scheduling with Shelf Life Neither the sales order nor the forecast nor the planned or production or- ders have classes or characteristics assigned. Figure 9.18 shows the prod- uct view for a product with shelf life. Required Shelf Life of Requirement < Shelf Life of Receipt Fig. 9.18. Product View for Shelf Life (APO) Note that the indicator for ‘planning with shelf life’ triggers the inclusion of the columns ‘minimum required shelf life date’ and ‘required maximum, 9.6 Goods Receipt with Shelf Life Characteristics 125 shelf life date’ into the product view. The ‘minimum required shelf life date’ of the requirement has to be before the ‘required maximum shelf life date’ of the receipt to be shelf life compliant. • Production Planning Another impact of the shelf life design, since it is related to the pegging area (i.e. one locationproduct), is that shelf life is not propagated, if the lo- cationproduct is processed in production or transported to another location. To consider the shelf life restrictions in production planning, the heuristic SAP_PP_SL001 has to be applied. This heuristic is a variant of the heuristic SAP_PP_002 as shown in figure 9.19. Fig. 9.19. Production Planning Heuristic SAP_PP_SL001 (APO) Shelf life alerts are only calculated, if the flag for shelf life is set in the product master. Though the SNP optimiser considers shelf life with some restrictions according to note 579556, its use is rather the exception. Since none of the other SNP functions support shelf life, it will be extremely difficult to in- terpret the result. The restriction that shelf life is limited to the location- product level remains for the SNP optimiser as well. • PP/DS Optimisation If the pegging relationship exists before calling the optimiser, the optimiser considers the pegging constraints. If no pegging exists, the optimiser ig- nores the shelf life constraints (i.e. the ‘pegging’ by the optimiser does not take shelf life into account). 9.6 Goods Receipt with Shelf Life Characteristics At the goods receipt the production date has to be maintained to provide the shelf life information for the batch, figure 9.20., 126 9 Planning with Shelf Life Goods Receipt Manual Entry Proposal Batch Valuation Fig. 9.20. Goods Receipt for Batches with Shelf Life (R/3) The expiration date is calculated from the ‘total shelf life’-value in the ma- terial master. This date is a proposal which can be overruled. 9.7 Shelf Life Alerts To support the planning with shelf life there are specific alert types for shelf life within the PP/DS alert profile. Figure 9.21 shows these alert types. Fig. 9.21. Alert Profile for Shelf Life Planning (APO) The alert types ‘maturation time not completed’, ‘shelf life too short’ and ‘shelf life too long’ indicate a violation of the pegging constraints. These, 9.7 Shelf Life Alerts 127 kinds of violations are of course only possible when fix pegging is used. Depending whether requirements or receipts are chosen, the alert monitor will list the requirement or the receipt element. The alert type ‘receipt without pegging relationship with expiration date’ indicates that a receipt element is not pegged – probably due to incompatible shelf life properties. Figure 9.22 shows a planning situation and the resulting alerts. Fig. 9.22. Shelf Life Alerts (APO) The flag for planning with shelf life in the product master is a prerequisite for the generation of shelf life alerts., 10 Sales Order Oriented Planning 10.1 Scenario Description 10.1.1 Process Chain Many companies use neither a pure make-to-order nor a pure make-to- stock strategy but a mixture – either by using lot sizes for a make-to-order production or by planning for customers (with commitments to the cus- tomers) in a make-to-stock production. The idea of this scenario is to com- bine the advantage of the make-to-stock production (e.g. the consideration of inventories) with the planning without final assembly (production is only triggered if sales orders are available) and a requirements planning for individual customers. R/3-SD APO-ATP APO-DP APO-PP/DS Sales Order Taking Demand Planning Forecasting Sales Order Entry Release with

ATP

Consumption Group Production Save Sales Order Forecast Consumption on Descriptive Characteristic Level Forecast Adjustment Production Planning Detailed Scheduling Order Conversion if Pegged to Sales Order Fig. 10.1. Process Chain for Sales Order Oriented Planning, 130 10 Sales Order Oriented Planning The scenario for sales order oriented planning contains three building blocks which differ from a typical make-to-stock scenario: • Forecast consumption on customer or customer group level, • adjustment of the forecast and the planned orders in the short term ac- cording to the sales order demand and • planned order conversion depending on existing sales order require- ments. • Demand Planning and Forecast Consumption Demand planning is performed on multiple levels – one of them is usually the customer or customer group level. The consumption of the forecast however does not take this level into account unless descriptive character- istics are used. In many cases the forecast consumption on location and product level is sufficient or even desired because deviances can be levelled this way. The business assumption for using the forecast consumption on customer level is that if one customer exceeds the forecast, the other customers will still place sales orders for their planned demand. • Sales Order Entry and ATP Check For the use of descriptive characteristics in APO neither characteristics nor classes are required in R/3. Therefore no characteristic value assignment takes place during sales order taking. There are no functional restrictions for the ATP check in this scenario – a simple product check would be the most straightforward choice. • Forecast Adjustment In the short term horizon the forecast is reduced and the planned orders are adjusted as well in order not to block the resource capacity. The motivation for this step is the assumption that no more sales orders will be placed for this horizon. This is an optional step, and whether the assumption is ap- propriate or not depends on the individual case. • Production Planning and Order Conversion For production planning and detailed scheduling there are no peculiarities. The production – i.e. the planned order conversion into a production order – is however only triggered if the planned order is pegged to a sales order. • Sales Order Oriented Planning for Make-to-Order It is also possible to apply this process for a make-to-order strategy. This might be favourable if no pure make-to-order is used. An example for this, 10.2 Descriptive Characteristics 131 is pharmaceutical production where the difference between many finished products lies mainly in the package and neither having inventory nor re- packaging of goods are an exception. The limitation for this process is that the ATP check is imprecise (checking horizon) or the result of the ATP check has to be corrected manually. The posting of goods from stock to the customer order assignment is an interactive process as well. 10.1.2 System Configuration Determinants Sales order oriented planning relies on descriptive characteristics to ensure that the forecast for a customer or a customer group is not consumed by a different customer resp. customer group. For descriptive characteristics no class assignment is required – neither in R/3 nor in APO. Figure 10.2 shows the system configuration determinants. Requirements Class Class Type (R/3) Configuration Scheme Master Data Account Assignment Make-to-Stock Batch (023) CDP PPM Make-to-Order Variant Config. (300) VC PDS Required / Supported Supported but Unusual Not Supported Fig. 10.2. System Configuration Determinants for Sales Order Oriented Planning The sales order oriented planning scenario is smoother for make-to-stock production, but as previously described it is also possible to apply this sce- nario for make-to-order with some restrictions. No configuration relevance is required, but technically both types of configuration relevance might be used. For the same reason the use of the class types is indifferent to this scenario. 10.2 Descriptive Characteristics The standard way of forecast consumption is on location and product level with the consequence that a more differentiated demand planning – for ex- ample on customer resp. customer group level or for different order types (e.g. express orders) – is undermined by an aggregated forecast consump- tion. With the use of descriptive characteristics it is possible to restrict the forecast consumption to location, product and descriptive characteristic. The prerequisite for this functionality is the definition of a consumption group as described in chapter 4.2. No changes to the planning area or plan-, 132 10 Sales Order Oriented Planning ning object structure are required, and no characteristics have to be as- signed to the product master. The use of descriptive characteristics is en- tirely within APO, so no settings in R/3 are required. Note that only those characteristics can be used for this purpose that are used in the sales order as well. • Representation of the Forecast For each characteristic combination a separate forecast is created. This means that there are more than one forecast orders for the same location- product at the same date. The descriptive characteristics use the ‚normal’ characteristic functionality for the display – the green triangle. The details are displayed by double-clicking on this triangle. Forecast Release Fig. 10.3. Forecast with Descriptive Characteristics (APO) In this example the forecast was performed on customer level. • Forecast Consumption The representation of the sales order in APO is similar as for the forecast – the green triangle exists, but the sales order contains only the descriptive characteristics as shown in figure 10.4. Fig. 10.4. Sales Order with Descriptive Characteristics (APO), 10.3 Forecast Adjustment 133 The forecast is consumed on the most detailed level, i.e. locationproduct and descriptive characteristic (in this case the customer). If the sales order for a characteristic value combination (here: customer X) exceeds the fore- cast quantity, the forecast for other characteristic value combinations (he- re: customer Y) is not consumed. A requirements strategy for planning with final assembly is used. Figure 10.5 displays the consumption situation (transaction /SAPAPO/DMP1) and shows that the values of the descriptive characteristics are taken into account. Fig. 10.5. Forecast Consumption for Descriptive Characteristics (APO) Using the ‘characteristic/value’ button it is possible to filter the forecast situation for certain characteristic values. • Release Back to Demand Planning with Descriptive Characteristics The release back to Demand Planning is used to provide the demand plan- ner with an information about the feasibility of the demand plan. The in- formation about the descriptive characteristic value is only available for the requirements (i.e. the forecasts and the sales orders) and not for the re- ceipts. Using the pegging between the forecast and the planned order the according receipts are determined and transferred back to DP with the de- scriptive characteristic information. This is only possible for planned re- ceipts and not for pegged stock elements. 10.3 Forecast Adjustment The forecast is adjusted in order to have only sales orders as demands in the short-term. Two horizons are introduced for this: • the ascertainment horizon, which is the basis for a detailed report- ing of non-consumed forecast (in order to provide a help for man- ual intervention), and • the adjustment horizon, in which the forecast is reduced by the ad- justment heuristic. The planned orders are reduced accordingly. The benefit of the forecast adjustment is that the capacity consumption on the resources is released and therefore the opportunity is given to improve the scheduling. This adjustment is only relevant for the short term and, 134 10 Sales Order Oriented Planning should only be applied if the assumption is valid that within this horizon usually no sales orders are placed. The definition of the horizon for the ascertainment (i.e. reporting) and the adjustment is done in the product master in the PP/DS-view as shown in figure 10.6. Fig. 10.6. Settings in the Product Master for Ascertainment & Adjustment Horizons (APO) We try to explain the significance of the ascertainment and the adjustment horizons with an example. Figure 10.7 shows the initial situation in the product view. The forecast is planned for the customers X and Y.

X X Y Y X Y Y X

Descriptive Characteristic Value Fig. 10.7. Example for Sales Order Oriented Planning (APO) In this example we have still forecast in the adjustment horizon. The val- ues of the descriptive characteristics are displayed in the white circles. In contrast to the use of ‘real’ characteristics it is not possible to display the value of the descriptive characteristics as a separate column in the product view. • Ascertainment Horizon Since forecasting is performed on customer resp. customer group level and for each of these a different forecast element is created, it is more confus- ing to keep the overview about the consumption situation. Additionally Adjustment Horizon Ascertainment Horizon, 10.4 Planned Order Conversion with Conversion Rules 135 aggregation usually levels the errors in forecasting and this is now missing. Therefore a report was developed to list the non-consumed forecast includ- ing the customer information within the ascertainment horizon as shown in figure 10.8 Fig. 10.8. Result of the Ascertainment Report (APO) This report is included in the heuristic SAP_PP_014 and called via back- ground production planning with the transaction /SAPAPO/CDPSB0. • Adjustment The adjustment heuristic SAP_PP_015 reduces the forecast and the planned orders within the adjustment horizon. Figure 10.9 shows the result of the adjustment heuristic based on the initial situation of figure 10.7. Result Log of the Adjustment Heuristic Adjustment of the Planned Order from 20 to 2 Reduction of the Forecast from 20 to 2 (2 are Consumed by Sales Order 10134) Fig. 10.9. Forecast Adjustment (APO) The forecast is only reduced to its consumed value and not completely de- leted. 10.4 Planned Order Conversion with Conversion Rules The conversion rule is defined in customising with the path APO Supply Chain Planning PP/DS Transfer to Execution Define Conversion Rule. and defines whether an ATP check and/or a requirements check is per- formed at the conversion of a planned order into a production order. The default conversion rule is defined in the global parameters and default val-, 136 10 Sales Order Oriented Planning ues for PP/DS (transaction /SAPAPO/RRPCUST1) and optionally in the product master. The entry in the product master overrules the entry in the global parameters and default values, figure 10.10. Global Parameters & Default Values Conversion Rule Product Master Fig. 10.10. Conversion Rule (APO) The ATP check is performed in APO for the components of the planned order. The requirements check on the other hand looks at the pegging of the planned order and issues a warning if it is not pegged to a sales order. Depending whether the planned order quantity is partially pegged to a sales order or only pegged to forecast different warning messages pop-up for manual conversion: Warning if Planned Order is Not Pegged to Any Sales Order Warning if Planned Order is Pegged to Forecast As Well Fig. 10.11. Warning for Conversion (APO) If not the full quantity of the planned order is pegged – e.g. due to lot sizes – no warning is issued. The background conversion is not performed if a warning exists., 11 Production Planning and Scheduling 11.1 Production Planning with Characteristics 11.1.1 Production Planning with the VC-Configuration The usage of the VC-configuration relevance is often linked with the use of object dependencies in R/3 and requires the PDS master data. For the products that are planned with the VC-configuration relevance, the charac- teristic values are not relevant for pegging. The connection between the requirement and the receipt is ensured by the sales order segment. There- fore variant configuration requires a make-to-order requirements strategy for a consistent process. The VC-configuration relevance does not have any impact on production planning. Figure 11.1 shows a planned order in APO for a product with the VC-configuration relevance. Fig. 11.1. Planned Order (VC-Configuration Relevance) Though the planned order has characteristic values assigned, these are only a reference to the characteristic values of the sales order. It is possible to construct examples in R/3 with production orders that have an individual configuration of their own, but that is rather unusual., 138 11 Production Planning and Scheduling 11.1.2 Production Planning with the CDP-Configuration With the CDP-configuration relevance the orders have valuations and re- quirements which contain the values of the characteristics. The valuations in APO are integrated with the configuration in R/3 or the batch classifica- tion (for batches). The requirements in APO are integrated with the batch selection criteria in R/3. Both the demand and the supply nodes (i.e. the requirements and the receipts) of the locationproduct have valuations – or at least they should. The valuations describe the value of the object itself while the requirements of a demand represent the conditions for a receipt to be pegged. If no supply with the same valuation exists, supplies with matching valuations according to the requirements of the demand node are considered. Only if no satisfactory supply exists, no pegging takes place. This property is mainly used within the mill scenario for the assignment of batches. Figure 11.2 illustrates this behaviour for nominal characteristics. In case 3 where no matching supply exists a new receipt with the valuation of the demand is created by production planning. The valuations of the demand are used when a planned order is created in PP/DS. Valuation Demand Valuation Demand Valuation Demand Green Green Green Requirement Requirement Requirement {Green, Blue} {Green, Blue} {Green, Blue} Green Blue Red Blue Red Black Green Batch Batch Batch Batch Batch Batch Planned 4711 4712 4715 4712 4715 4716 Order 1 Check For Existing Supply Else 2 Check For Supply which Else 3 Create Supply with the Same with the Same Valuation Matches the Requirement Valuation Fig. 11.2. Pegging for CDP-Configuration Relevance Demands or supplies without a valuation resp. with incomplete require- ments (unspecified characteristics) might cause problems in the further course of the planning – especially if ATP on characteristics level is used. A sub-optimal pegging is likely, in production planning an incorrect pro- duction plan might result and in the ATP overconfirmations might occur. Note 526883 describes these risks in more detail. Though pegging takes the CDP-configuration into consideration, the pegging relations are not optimised – even if the configuration is complete for all demands and supplies (see also note 526883). It might as well hap- pen that demands are not covered though a different assignment of sup-, 11.1 Production Planning with Characteristics 139 plies would allow this – and this might even happen to demands which have been covered before, as figure 11.3 shows. Net requirements calcula- tion within production planning and pegging are separate steps with differ- ent rules. Demand Blue B Demand {Green, Blue} Demand Red A Red A {Red, Blue} {Red, Blue} Green Blue Green Blue Batch Batch Batch Batch 4711 4712 4711 4712 Fig. 11.3. Suboptimal Pegging and Stealing of Supplies In this example two batches are available, one with the evaluation ‘GREEN’ and one with the evaluation ‘BLUE’. The demand A has the re- quirement for ‘RED’ and ‘BLUE’, therefore the blue batch is pegged to the demand. Now a second demand B with the requirement for green and blue is created. Though it would be possible to cover both requirements if the green batch was pegged to demand B, a new pegging (which is triggered by the creation of a new demand node) might assign the blue batch to the new demand B with the consequence that demand A is not covered any more. Up to APO 4.0 pegging arcs were only created if the valuation of the re- ceipt element did match the requirements of the demand element. With APO 4.1 pegging takes place between identical valuations as well. There- fore the requirements are optional and should only be used if necessary. • Production Planning The standard production planning heuristics (e.g. SAP_PP_002) considers the CDP-configuration, therefore no change in the production planning is required. The valuation of the newly created planned order is the valuation of the demand node. If the demand node does not have any valuation (even if it does have requirements) the newly created planned order will not have any valuation either. The performance requirements for production planning with CDP- configuration are much higher than without. This is a limitation for this process. Figure 11.4 shows a planned order in APO for the CDP- configuration relevance., 140 11 Production Planning and Scheduling Fig. 11.4. Planned Order in APO (CDP-Configuration Relevance) A rather substantial difference to the VC-configuration relevance is that planned orders are automatically converted into production orders when they are sent to R/3 if they have nodes with CDP-configuration relevance. The reason for this is that a planned order can not have a configuration of its own in R/3. • CDP in a Make-to-Stock Environment Differing from the VC-configuration relevance, production planning with CDP-configuration relevance can be used in a make-to-stock environment as well. In this case for the integration of the planned and production or- ders to R/3 the IS Mill solution is required however. 11.2 Detailed Scheduling with Characteristics 11.2.1 Impact of Characteristics on Detailed Scheduling The characteristics information is either included in the pegging informa- tion (CDP-configuration and shelf life) or linked via customer order seg- ments (VC-configuration). The impact of characteristics on scheduling has two aspects. One is that scheduling considers the characteristics in the sense that scheduling is able to create results that are feasible in terms of matching configurations. The customer order assignment is always considered – i.e. it is not possible to assign receipts to requirements of different customer order segments. For the CDP-configuration relevance in a make-to-stock environment the pegging assignment between receipts and requirements might change if new demands or supplies are created or re-scheduled. The, 11.2 Detailed Scheduling with Characteristics 141 pegging in this case might be sub-optimal as described in this chapter ear- lier. The planning strategies within the scheduling heuristics control whether pegging and therefore the dependencies between receipts and re- quirements are considered. The optimiser does consider customer order assignments and the CDP- configuration like the pegging does – independent of an existing pegging arc at the starting point. The shelf life constraints however are only consid- ered if pegging already exists at the initial situation. The other aspect of detailed scheduling with characteristics is how the characteristics are applied as a criterion for scheduling. In APO there are two different possibilities for that. One is to use the characteristic values as a sorting criterion for the scheduling sequence within the scheduling heu- ristics, the other is to reflect the different characteristic values by different set-up between the activities. 11.2.2 Characteristics as Sorting Criteria for Heuristics The strategy profile and the scheduling heuristic SAP001 allow to define a sequence for scheduling both in interactive and in batch mode. For differ- ent criteria it is possible to define an ascending or a descending order – other sequences for sorting are not possible. If another sequence is re- quired, a workaround using a different (new) field has to be used. Many common fields are available as sorting criteria but no characteristics since the characteristics are master data and implementation specific. It is how- ever possible to define the characteristics as customer field and use them for sorting. • Characteristic Based Sequence Number The most convenient way to use characteristics values as sorting criteria is by setting the value of the ‘sequence’-field in the operation via object de- pendencies as described in chapter 5.3.4. The reference characteristic has to influence the field SEQNR in APO. This field is part of the structure /SAPAPO/CULL_CFG_OPERATION. • Characteristics as Sorting Criteria for CDP-Configuration Relevance Another way to use characteristics to influence the sequencing in the scheduling heuristics is to make the characteristics available as sorting cri- teria. This is however only possible for the CDP-configuration relevance. The method GET_ORDDATA of the BAdI /SAPAPO/CDPS_ORDDATA has to be implemented and a field for the characteristics must be appended as customer field to the structure CI_ORDCUS. For this field the domain and, 142 11 Production Planning and Scheduling the data element have to be created. The structure CI_ORDCUS itself is part of another structure as shown in figure 11.5. All entries are on APO side. BAdI /SAPAPO/CDPS_ORDDATA Parameter Structure ET_ORDERS_EXT /SAPAPO/CDPS_ORDER_EXT_STR Structure CI_ORDCUS Data Type Domain Data Element Field INT1 ZXX_SIZE_DOMAIN ZXX_SIZE CHARNAME New Elements Existing Elements Fig. 11.5. Structure of the Enhancement to Include Characteristics as Sorting Criteria

Figure 11.5 shows an example where the characteristic CBP_SIZE is used

as sorting criterion and the domain ZXX_SIZE_DOMAIN and the data ele- ment ZXX_SIZE have been created. The domain and the data element are created with the transaction SE11. The creation of the data element is ex- plained in figure 11.6. Text Maintenance (Short, Medium & Long) Fig. 11.6. Data Element (APO)

The field for the characteristic CBP_SIZE – in this case called CBP_SIZE as

well – is assigned with the data element to the structure CI_ORDCUS. The structure is either created using the transaction SE11 or from the display of, 11.2 Detailed Scheduling with Characteristics 143 the top-level structure /SAPAPO/CDPS_ORDER_EXT_STR. In this example we defined the field CBP_SIZE in the structure CI_ORDCUS with the con- sequence that this field is available as a sorting criterion in the strategy (transaction /SAPAPO/CDPSC1) as shown in figure 11.7. Sorting Criterion 'Size' (Characteristic) for Scheduling Fig. 11.7. Characteristic as Sorting Criterion in the Strategy Profile (APO) For display and performance reasons this characteristic should be included into the operation as well. The inclusion is done in the customising of the graphical objects in the planning board customising (transaction /SAPAPO/CDPSC2). Figure 11.8 shows the relevant steps to select the field for display. Characteristic 'Size' for Display Fig. 11.8. Planning Board Customising for the Display of Characteristics (APO), 144 11 Production Planning and Scheduling Finally the BAdI /SAPAPO/CDPS_ORDDATA has to be implemented in or- der to provide the planned order resp. production order with the character- istic value. An example for the usage of the characteristic CBP_SIZE (taken from the output node of the order) is listed among the implementation samples in the appendix. 11.2.3 Characteristics for Set-Up Group Determination The use of sequence dependent set-up is a common way in APO to influ- ence the sequence. As described in Dickersbach (2004), sequence depend- ent set-up is modelled in APO by assigning a set-up group resp. a set-up key to the operation and defining the set-up duration between the set-up groups resp. keys in a set-up matrix. One or more set-up keys are assigned optionally to a set-up group to refine the set-up duration for certain set-up groups. The set-up group or – if set-up keys are used - the set-up key is as- signed to the PDS resp. the PPM on operation level. Though there is no integration, the set-up group corresponds to the set-up group category in R/3 and the set-up key to the set-up group key in R/3. The set-up matrix defines the set-up duration between the set-up groups resp. the set-up keys and is assigned to the resource as shown in figure 11.9. Set-Up Matrix PDS / PPM Set-Up Group A from / toXYXX- 2YBY31Res. 4711AABBAASet-Up Fig. 11.9. Calculation of Sequence Dependent Set-Up (Dickersbach 2004) Within the PPM the set-up group can be controlled depending on the value of a characteristic as described in chapter 8.2. To control the set-up group via object dependencies in the PDS it is necessary to use variant functions because the set-up group is stored for the PDS as an internal key. There- fore a reference characteristic has to be used in APO and the value of the reference characteristic (the internal key) has to be calculated via a func- tion module as described in chapter 5.3.5. The structure for the set-up, 11.3 Planned Order Integration 145 group and the set-up key in the PDS is /SAPAPO/CULL_CFG_ACT_ALL, the fields are GROUP_ID for the set-up group and ITEM_ID for the set-up key. • Example How to Control the Set-Up Group with Object Dependencies In the following we describe an example how to control the set-up group based on the value of the characteristic CBP_COLOUR. The reference char- acteristic is XREF_ROUTINGSETUP, the variant function and the function module are ZXX_SETUP. The use of variant functions is defined in chap- ter 5 (e.g. in figure 5.20). The object dependency has to call the variant function ZXX_SETUP: FUNCTION ZXX_SETUP (CBP_COLOUR = $ROOT.CBP_ COLOUR, XREF_ROUTINGSETUP = $SELF.XREF_ROUTINGSETUP) The dependency group SAPAPOACT has to be assigned to the object de- pendency, and the object dependency has to be assigned to the operation of the routing. The function module ZXX_SETUP and the variant function ZXX_SETUP need to be created in APO as described in chapter 5.3.5. An example coding for the function module is listed as an implementation sample in the appendix. To enable the sequence dependent set-up calculation the flag for se- quence dependent set-up has to be set. This is done per default during the transfer of the PDS if the set-up group is set in the routing in R/3. If no set- up group is maintained in the routing, the relevance for sequence depend- ent set-up must be set explicitely via a reference characteristic in APO. The field that has to be referenced for this case is IS_SETUP_ACT of the structure /SAPAPO/CULL_CFG_ACT_SETUP. 11.3 Planned Order Integration For the VC-configuration relevance the configuration of the planned order exists only as a reference to the configuration of the sales order, whereas a planned order with CDP-configuration relevance contains its own configu- ration without reference to the sales order. However, in the case of a prod- uct with CDP-configuration relevance and a make-to-order strategy the planned order references the sales order (in order to avoid inconsistencies at the integration with R/3). In R/3 the planned order may only have configuration as a reference to the sales order, therefore only planned orders with VC-configuration rele-, 146 11 Production Planning and Scheduling vance characteristics are transferred to R/3 as planned orders. If the planned order contains CDP-configuration relevance (at the output node), the planned order is automatically converted into a production order when transferred to R/3. Figure 11.10 shows the configuration of a planned order in R/3 that was transferred from APO (using the VC-configuration rele- vance). Fig. 11.10. Planned Order in R/3 (VC-Configuration Relevance in APO) The production order in R/3 is configurable – 11.11 shows the valuation of the production order. Fig. 11.11. Production Order in R/3, 11.4 Display of Characteristics in the Product View 147 It does not make a difference for the production order whether the VC- or the CDP-configuration relevance is used in APO. 11.4 Display of Characteristics in the Product View It is possible to display the characteristic values in the product view using the characteristic visualisation profile. The characteristic visualisation pro- file is defined with the customising path APO Supply Chain Planning PP/DS Product Planning Table Define Visualisation Profiles. As shown in figure 11.12. Fig. 11.12. Characteristic Visualisation Profile (APO) The values of up to ten characteristics can be displayed this way (the char- acteristics are assigned to the visualisation profile and column number). The characteristic visualisation profile is assigned to the ‘orders’ view of the user settings as shown in figure 11.13. Fig. 11.13. Assignment of the Visualisation Profile to the User Settings (APO), 148 11 Production Planning and Scheduling The result of this assignment is that a new column is created with the char- acteristic value as shown in figure 11.14. Characteristic Value Fig. 11.14. Product View with the Display of Characteristic Values (APO) This is a convenient way to provide an overview about the characteristic values of different orders. For the CDP-configuration relevance it is additionally possible to filter orders via characteristics when entering the product view or within the product view using the triangle in the icon bar.,

References

Literature Dickersbach, J. Th.: Supply Chain Management with APO. Springer-Verlag, Berlin Heidelberg 2004 Eversheim, W.: Organisation in der Produktionstechnik Band 1. VDI-Verlag, Düsseldorf 1996 Knolmayer, G., Mertens, P., Zeier, A.: Supply Chain Management Based on SAP Systems. Springer-Verlag, Berlin Heidelberg 2002 OSS Notes 33396 Batch determination.: Selection with remaining life LOBM_RLZ 78235 Revaluating reference characteristics in batch classification 391018 Shelf life R/3 APO, settings in R/3 426563 CTP: Settings, system behaviour and performance 453921 X: char. value, batch no., copy acc. assignment to RBATP item 483576 Required shelf life per customer/material without configuration, 150 References 484144 Information on characteristics-based forecasting 495825 CDP definitions for block planning via user exit 526883 CDP: Unspecified characteristics in planning and pegging 528189 Additional information on block planning 551124 APO: Finite scheduling with MRP heuristic 601255 Variant functions cannot be created 602174 Standard LOBM_APO_SL* characteristics do not exist in client 604757 Incorrect variant functions and object dependencies 614280 Variant tables cannot be created 714929 Changing characteristic and class master data for CDP 751392 Shelf life R/3 APO, settings in R/3 system 815018 Change of component using dependency,

Abbreviations

APO Advanced Planner and Optimiser ATP Available-to-Promise BAdI Business Add-In BOM Bill of Material CBF Characteristic Based Forecasting CBP Characteristic Based Planning CDP Characteristic Dependent Planning CIF Core Interface CTM Capable-to-Match CTP Capable-to-Promise CVC Characteristic Value Combination DP Demand Planning GR Goods Receipt IS Industry Solution MLATP Multi-Level ATP MTO Make-to-Order MTS Make-to-Stock PDS Production Data Structure PIR Planned Independent Requirement PP/DS Production Planning and Detailed Scheduling PPM Production Process Model RBATP Rules-Based ATP RTO Run-Time Object SCM Supply Chain Management SNP Supply Network Planning TP/VS Transportation Planning and Vehicle Scheduling VC Variant Configuration,

Implementation Samples

For the detailed description about how to implement a function at several points in the book coding was required – either as object dependency, as BAdI or as a function module. In chapter 11.2 a BAdI and a function mod- ule are mentioned which require a more extensive coding. Examples for this coding are listed in this chapter. • BADI to Include Characteristics as Sorting Criteria As mentioned in chapter 11.2.2 it is possible to include characteristics as sorting criteria into the scheduling strategy and into the scheduling heuris- tic using the BAdI /SAPAPO/CDPS_ORDDATA. An example for the usage of the characteristic CBP_SIZE (taken from the output node of the order) is listed here: method /SAPAPO/IF_EX_CDPS_ORDDATA~GET_ORDDATA. DATA: lv_charact_id TYPE /sapapo/mc01ch_id, ls_gen_params TYPE /sapapo/om_gen_params, lv_simsession TYPE /sapapo/om_simsession, ls_exclude_exports TYPE /sapapo/om_getdata_options, lt_orders TYPE /sapapo/om_ordid_tab, lt_outputs TYPE /sapapo/om_io_tab, lt_val TYPE /sapapo/om_charact_val_tab, lt_rc TYPE /sapapo/om_lc_rc_tab. FIELD-SYMBOLS: LIKE LINE OF it_orders, LIKE LINE OF it_outputs, LIKE LINE OF et_orders_ext, LIKE LINE OF lt_val. * get internal ID of characteristic: CALL METHOD /sapapo/cl_mc01_ccv_struct=>charname_as_id

EXPORTING

ic_charname = 'CBP_SIZE'

IMPORTING

en_charid = lv_charact_id

EXCEPTIONS

OTHERS = 2. CHECK sy-subrc IS INITIAL., 154 Implementation Samples LOOP AT it_orders ASSIGNING . APPEND -orderid TO lt_orders. ENDLOOP. " AT it_orders SORT lt_orders. DELETE ADJACENT DUPLICATES FROM lt_orders. * get charact. data of order output nodes from liveCache: CALL FUNCTION '/SAPAPO/RRP_SIMSESSION_GET'

IMPORTING

ev_simsession = lv_simsession es_gen_params = ls_gen_params. CLEAR ls_exclude_exports WITH 'X'. CLEAR: ls_exclude_exports-get_charact_outnode. CALL FUNCTION '/SAPAPO/OM_ORDER_GET_DATA'

EXPORTING

is_gen_params = ls_gen_params iv_simsession = lv_simsession it_order = lt_orders[] is_exclude_exports = ls_exclude_exports

IMPORTING

* et_charact_val_acts = * et_charact_req_inpnode = et_charact_val_outnode = lt_val[] et_rc = lt_rc[]

EXCEPTIONS

lc_connect_failed = 1 lc_com_error = 2 lc_appl_error = 3. CHECK sy-subrc IS INITIAL. SORT lt_val BY object_id object_type position_no line_no charact_id. * provide sorting value for orders from characteristics: SORT lt_outputs BY orderid position_no line_no. LOOP AT et_orders_ext ASSIGNING . * get output for order (assuming only 1 output): READ TABLE it_outputs ASSIGNING WITH KEY orderid = -ordid BINARY SEARCH., Implementation Samples 155 CHECK sy-subrc IS INITIAL. * get characteristic value for output: READ TABLE lt_val ASSIGNING WITH KEY object_id = -orderid position_no = -position_no line_no = -line_no charact_id = lv_charact_id. CHECK sy-subrc IS INITIAL. -CBP_SIZE = -quan_value. ENDLOOP. " AT et_orders_ext endmethod. Special cases as by-products are ignored in this example. • Function Module to Determine the Set-Up Group Identifier Not all fields have a simple correspondence between R/3 and APO. For the use of reference characteristics this implies that variant functions might be required to determine the APO internal identifiers for some fields. An ex- ample coding for the function module (called by the variant function) to determine the identifier for the set-up group is listed in the following as mentioned in chapter 11.2.3. In this case the set-up group is set to ‘XX2’ if the value of CBP_COLOUR is ‘RED’. FUNCTION ZXX_SETUP. *"- *"*"Local Interface: *" IMPORTING *" REFERENCE(GLOBALS) LIKE CUOV_00 STRUCTURE CUOV_00 *" TABLES *" QUERY STRUCTURE CUOV_01 *" MATCH STRUCTURE CUOV_01 *" EXCEPTIONS *" FAIL *" INTERNAL_ERROR *"- data: lv_apo_group_id TYPE /SAPAPO/CDPS_SETUP_ID, lv_apo_ITEM_ID TYPE /SAPAPO/CDPS_SETUP_ID, lv_SETUP_GROUP TYPE /SAPAPO/CDPS_SETUP_GROUP, lv_SETUP_ITEM TYPE /SAPAPO/CDPS_SETUP_ITEM, ls_CUOV_01 TYPE CUOV_01, lv_SETUP_ID TYPE /SAPAPO/CDPS_SETUP_ID., 156 Implementation Samples READ TABLE QUERY WITH KEY VARNAM = 'CBP_COLOUR' INTO ls_CUOV_01. CHECK sy-subrc = 0. IF ls_CUOV_01-atwrt = 'RED'. lv_SETUP_GROUP = 'XX2'. ENDIF. CALL FUNCTION '/SAPAPO/DM_SETUP_ID_READ'

EXPORTING

i_locid = 'LjGhTBmHSKQa3r65jI0OyW' i_setup_group = lv_setup_group i_setup_item = lv_setup_item

IMPORTING

e_setup_id = lv_setup_id

EXCEPTIONS

not_found = 1 OTHERS = 2. CALL FUNCTION '/SAPAPO/DM_SETUP_IDS_GET'

EXPORTING

i_setup_id = lv_setup_id

IMPORTING

e_setup_group_id = lv_apo_group_id e_setup_item_id = lv_apo_item_id

EXCEPTIONS

not_found = 1 OTHERS = 2. CLEAR ls_CUOV_01. ls_CUOV_01-varnam = 'XREF_ROUTINGSETUP_GROUP'. ls_CUOV_01-atflv = lv_apo_group_id. ls_CUOV_01-atfor = 'NUM'. APPEND ls_CUOV_01 to MATCH. ENDFUNCTION. Since the internal key for the set-up group is location dependent, the func- tion /SAPAPO/DM_SETUP_ID_READ needs the GUID of the location as an input. The GUID is stored in the table /SAPAPO/LOC.,

Transactions

For the quick access of some functions this chapter provides a list of useful transactions for planning with characteristics. Most of these have been ex- plained in the text. Characteristics and Classes System Description of the Transaction Transaction R/3, APO Characteristics CT04 R/3, APO Class CL02 R/3, APO Assignment of Class to Product CL24N R/3, APO Organisational Area O1CL Variant Configuration System Description of the Transaction Transaction R/3 Configuration Profile CU41 R/3 Object Dependency CU01 R/3 Object Dependencies Check CU50 APO PDS Display & Config. Simulation /SAPAPO/CURTO_SIMU R/3, APO Variant Functions CU65 R/3 BOM Assignment to Material Variant CS40 R/3 Routing Assignment to Mat. Variant CA02 R/3, APO Variant Table – Definition CU61 R/3, APO Variant Table – Data Maintenance CU60 Batch Selection and ATP with Characteristics System Description of the Transaction Transaction R/3 Batch Selection Search Strategy VCH1 R/3 Batch Customising OCHA R/3 Sort Sequence for Characteristics CU70 APO Rules /SAPAPO/RBA04 APO Characteristic View (ATP) /SAPAPO/ATPCH01, 158 Transactions Characteristic Based Forecasting System Description of the Transaction Transaction APO CBF- Profile Maintenance /SAPAPO/IPM01 APO CBF- Profile Maintenance (old) /SAPAPO/DPC1 APO Characteristic Value Combinations /SAPAPO/MC62 APO Consumption Group /SAPAPO/CSP1 APO Forecast Release /SAPAPO/MC90 APO Forecast Reorganisation /SAPAPO/MD74 APO Transform. of Dependent Demand /SAPAPO/DMP2 APO Consumption Overview /SAPAPO/DMP1 Block Planning System Description of the Transaction Transaction APO Resource /SAPAPO/RES01 APO PPM /SAPAPO/SCC05 APO Forecast Generation from Blocks /SAPAPO/BLRG01 Shelf Life System Description of the Transaction Transaction R/3 Update Shelf-Life Characteristics BMSM Production Planning and Detailed Scheduling System Description of the Transaction Transaction APO Product View /SAPAPO/RRP3 APO Pegging Overview /SAPAPO/PEG1 APO Global Parameters and Default Values /SAPAPO/RRPCUST1 APO Background Planning /SAPAPO/CDPSB0 APO Scheduling Strategy Profile /SAPAPO/CDPSC1 APO Planning Board Customising /SAPAPO/CDPSC2 APO PP/DS Heuristics /SAPAPO/CDPSC11,

Index

Account Assignment 10 Characteristic View 89 f. Adjustment 130, 133 ff., 135 Checking Horizon 29 Alerts 126 f. Class 19 ff., 46, 118 ff. Allocation Check 80 Class Type 11, 20, 22 f., 98, 102, Ascertainment 134 ff. 107 Assembly Group 109 Component Logic 74 f. Assembly Planning 71, 72 Configurable Material 30 Assignment of Characteristic Configurable Material Variant Values 16, 88, 96 50 ff. ATP Check 16, 70, 78, 114 Configuration 32, 38, 46, 96 ATP with Characteristics 13, 26, Configuration Determinants 9 ff., 84, 86, 89 ff. 12, 14, 29, 70, 79, 85, 97, 114, BAdI 51, 65, 108, 109, 111, 141, Configuration Profile 30 f., 44 144 Configuration Relevance 25, 61, Batch 2, 24, 111 f. 75, 83, 137, 138 ff., 145 f. Batch Search Strategy 86 ff., 122 Configuration Scheme 11, 13, Batch Selection 26, 86 ff., 93, 96, 25 f., 66, 98 120 Configure-to-Order with Propa- Block Definition 102 ff., 109 gation 7, 95 ff. Block Planning 26, 102 ff. Consumption Group 61, 62 f., 67, BOM 35 f., 46, 48, 66 131 Conversion Rule 136 CBF 26, 56 ff., CTM 13 CBF-Profile 57, 59, 74 CTP 28, 114 CBF-Table 57 f., 80 CDP 25, 138 ff. Delivery 93 f. Characteristic 1, 19 ff., 118 ff., Demand Planning 4, 15, 67, 114, 141 f., 147 130 Characteristic Propagation 26, Dependency Group 35, 40, 107 96, 98 ff. Descriptive Characteristics Characteristic Substitution 90 f. 131 ff. Characteristic Value Combina- Detailed Scheduling 17, 140 ff. tion 57, 58, 59 f., 160 Index Field Catalogue 63, 80 Planned Order 69, 78, 137, 140, Forecast 62, 68, 132 146 Forecast Adjustment 130, 133 ff., Planned Order Conversion 17, 135 26, 130, 135 f. Forecast Check 79 f. Planning Area 57 Forecast Consumption 16, 60, 72, Planning Object Structure 56 81, 130, 132 f. Planning Segment 61, 76 Forecast Generation 109 ff. PP/DS-Optimisation 125, 141 Forecast Release 16, 60, 61 ff., PPM 11, 98 ff., 106, 144 67, 76 Procedures 38, 39 Production 4, 17 Goods Receipt 17, 111, 125 Production Order 146 Production Planning 17, 68, 84, Industry Focus 9 96, 124 f., 137 ff., 139 Info Object 56, 62 IS Mill 14, 61, 75, 83, 96 Rapid Planning Matrix 13 Item Category 31 f., 49, 51 Reference Characteristics 39 f. Reference Characteristic Map- Make-to-Order 10 ping 41 f. Make-to-Order with Variant Release 16, 60, 61 ff., 67, 76 Configuration 6, 27 ff. Requirements 25, 89, 97, 138 f. Make-to-Order with Variant Requirements Class 10, 31 f., 51 Configuration and Planning 7, Requirements Strategy 10, 61, 55 ff., 65 ff., 75 ff. 67, 69, 71, 72, 73 Make-to-Stock 10 Routing 48 f., 107 f. Make-to-Stock Segment 61, 64 RTO 11 Material Class 46 Rules-Based ATP 79, 90 f. Material Type 30, 47 Material Variants 13, 47 ff. Sales 4, 16 Multi-Level ATP 13, 26, 28 Sales from Stock with Charac- Multi-Level Configuration 26, teristics 7, 83 ff. 44 ff. Sales Order 33, 34, 89, 92 Sales Order Oriented Planning 8, Object Dependencies 26, 34 ff., 61, 62, 129 ff. 37, 107, 121, 145 Scenario 5 Object Dependencies Check 36, Scheduling Heuristic 140, 141 ff. 50 Search Strategy 86 ff., 122 Object Variables 36 Selection Condition 38, 101 Organisational Area 20 f., 120 Set-Up Group 41, 101, 144 f. Shelf Life 8, 14, 26, 113 ff. PDS 11, 37, 47, 107 f., 144 Pegging 26, 78, 85, 125, 138 f., Index 161 System Configuration Determi- nants 9 ff., 12, 14, 29, 70, 79, 85, 97, 114, 131 TP/VS 13 Transformation of Dependent Demand 69, 71 f. Variant Configuration 1, 25, 26, 31 ff., 137 Variant Functions 42 ff., 121, 145 Variant Tables 52 f. Visualisation 105, 147 f. Visualisation Profile 147]

15