Explaining Automotive Make-to-Order Scenarios

In the dynamic and competitive automotive industry, manufacturers are constantly challenged by the need to manage high volumes of production while meeting the individual requirements of their customers. The automotive make-to-order (MTO) process is a critical component for companies aiming to deliver customized vehicles efficiently and effectively. This chapter introduces the SAP Solution for the Automotive Make-to-Order Process, an integrative solution that spans across SAP Integrated Process and Product Engineering and embedded PP/DS.

Integrated Product and Process Engineering (iPPE) is an integral part of SAP S/4HANA designed to manage the complete product lifecycle, especially for products with many variants. While iPPE also covers modeling from the early engineering stages and enables collaboration thought the product lifecycle, this course will focus on its functionalities supporting the production process of high-variant and high-volume products.

iPPE offers an integrated master data model for production of vehicles such as the bill of materials in the Product Structure (PVS), routing data using the Process Structure (ACT) and the production line structures through Factory Layout (FLO). To utilize iPPE effectively for production of configurable products like vehicles, integration with S/4HANA components like engineering change management, the classification system, and variant configuration is required as well. This chapter will explain the fundamentals of the integrated master data model in iPPE which is the foundation of production planning with embedded (Production Planning/Detailed Scheduling) in SAP S/4HANA.

Product Variant Structure (PVS) offers the possibility for a redundancy-free description of products or families of products with numerous variants. In this context Product Variant Structure (PVS) functions as the so called super-BOM, combining all possible variants into one unified model, to which then material master data is linked.

PVS consists of three fundamental elements that are used to model the configurable product: the access node, view nodes, and structure nodes.

The mandatory access node defines the product to be built - for example, a car - and contains access variants for fully configurable vehicles and/or their variants.

The optional view nodes, allows for the implementation of logical groupings without material assignments and as such they can be used to create multi-level hierarchies. They are typically used to model parts of a vehicle like the engine, the wheels, or the seats. View nodes do not influence planning directly and are ignored during BOM explosions.

Structure nodes define the components used to build the product. Components for the structure nodes are entered as structure node variants, and these nodes may be connected directly to the access node or through view nodes. Structure nodes utilize object dependencies like the so-called selection conditions, that dynamically control which components need to be selected. Additionally, structure nodes can contain multiple component variants, with selection conditions that determine which component variants are selected. Object dependencies are local selection conditions within PVS and are vital for the correct explosion of the structure.

The hierarchical structure in PVS begins at the access node, followed by the optional view node, then the structure node where object dependencies and components are defined, and finally, the assembly node at the lowest level. It is important to note that all node types except the assembly node must be assigned a variant class.

Assembly nodes represent single-level BOM structures, involving a parent assembly with a list of child components. These nodes do not support object dependencies, meaning any linked component will always be selected once added. Multiple parents can be assigned to a single assembly node.

In the iPPE Process Structure (ACT), production processes are modeled using various node types to allow the modeling of a super-routing.

The header node is mandatory, representing the highest level of the process model, also known as line routing, and stores the header data and activity sequence. Grouping nodes are optional and are used to group several activities and operations into a line sub routing, enabling the creation of multi-level process hierarchies. Line operations, which are also optional and assigned to the line subrouting, group activities for joint execution.

Activities are the basic and lowest level elements in the process structure and they describe how and where a single process step is executed. The activity sequences can be modeled within the process structure, although the actual sequence used in production planning is determined via line balancing which will be covered later in this course.

Attributes of activity nodes include job description and capacity requirements and they can be used in multiple operations and grouping nodes. Modes are used to assign activity to a resource. Resources are used to describe the respective process step and they can either be a human resource or a machine performing the assembly operations.

In this section, we delve into the Factory Layout (FLO), which is used to model the factory or production plant. FLO objects encompass various elements such as lines, work stations, and alternative lines, linking production line resources to these structures.

FLO objects are organized into a hierarchical structure called a Line Net, allowing for both single-line and multi-line configurations which are crucial for planning the flow of finished items through the plant by considering factors such as takt times and the longest path. The line net is at the top of the Factory Layout object hierarchy and serves as the header node and contains the Line Balance information. Beneath this is the production line, which must include a line resource for scheduling purposes, as well as potentially containing a Supply Area and an Action Point used for tasks like backflushing and tracking.

Production Lines can be subdivided into Line Segments, although these segments do not have resources and thus cannot be scheduled independently. Line Segments can, however, include supply areas and an action point of their own. The smallest divisions are Work Areas, which help distinguish inventory supply locations within a plant using supply areas but do not support scheduling or reporting.

Line Resources in the FLO system play a critical role in several operational aspects, including capacity planning in repetitive manufacturing, component demand planning using the rapid planning matrix, and the determination of supply areas for PVS components. They also assist in backflushing and line balancing.

As stated in the beginning of this chapter, iPPE offers an integrated data model combining master data for the bill-of-materials, the factory layout and the process structure. In this section, will look into the integration points of Product Variant Structure (PVS), Activity Structure (ACT), and Factory Layout (FLO).

The different iPPE application objects are linked with each other in the following way:

• Product Variant Structure is related to the Activity Structure by linking materials to activities in the structure nodes or structure node variants.
• Activity Structure is related to the Factory Layout by linking the activities to a location in the plant to map out at which station of the plant a certain activity will be performed.

The integrated master data model of iPPE makes it possible to determine exactly when and where in the factory which components are required. This allows for fine-tuned scheduling and inventory control. Components are assigned to activities, and activities are linked to line objects through the line balance, dictating precise timing for each part in the production cycle.

To achieve proper line balance, it is essential to create a model mix, which includes a list of products and their proportions in production. The system calculates weighted durations based on this mix, adjusting labor requirements to match the average labor content of the assembled products. Should discrepancies arise between load and capacity, adjustments can be made by altering the model mix, reallocating operations, changing takt time, or modifying capacity and activity durations. By managing integration efficiently, production lines can maintain harmony between workflow, resource allocation, and overall production capacity, enhancing productivity and reducing bottlenecks.

In production planning several conflicting demands need to be managed when trying to derive an optimal production schedule. These come from customers who want their individually configured cars produced and delivered quickly while for production it is easier to build the same car in the same volume every day. Therefore, line capacity must be utilized efficiently which becomes even more difficult when scheduling across multiple production line.

To solve this challenge, Model Mix Planning and Sequencing in embedded PP/DS helps to determine the optimal production schedule considering restrictions that can be defined in PP/DS.

To determine the production schedule, model mix planning and sequencing are used to schedule the production orders for vehicles in different time frames:

• Model Mix Planning is conducted for mid-term planning and vehicles are scheduled in daily or shift buckets which represent the capacity of cars that can be built. To assign vehicle demands firm orders and/or forecasts are considered.
• Sequencing is conducted in the short term by scheduling vehicles to individual takt. Due to the rolling horizon, the production orders that were first planned with Model Mix Planning will eventually be planned with sequencing which takes the orders that are in a bucket and assigns them to a specific takt. By assigning the order to a takt, order start and end time can be calculated.

User defined restrictions help determine the optimal allocation of orders to takts. The following restriction types can be created in PP/DS:

• Quantity restrictions define the minimum or maximum of a characteristic that can be built (for example, only 90 cars with VF9 can be built per shift)
• Spacing restrictions define the minimum or maximum break between similar characteristics (for example, after each convertible, there must be at least two hard top vehicles)
• 'K in M' or density restrictions defines the proportion of a certain characteristics (for example, in each group of five cars, max two must have air conditioning)
• Block restrictions defines the minimum or maximum group of a certain characteristic (for example, try to schedule five red cars in row)
• Even distribution tries to spread a certain characteristic across the period (for example, schedule vehicles with VF9 evenly across the day)
• Position restrictions schedules a characteristic in a specific takt (for example, schedule the prototype car in the first takt)

When we refer to production orders being scheduled by Model Mix Planning and Sequencing it is important to note that these are not classic planned orders, but so-called matrix planned orders for which requirements planning is carried out using the Rapid Planning Matrix.

After the production schedule is determined via Model Mix Planning and Sequencing, the requirements need to be calculated for these orders. While MRP in S/4HANA is a great choice for many scenarios, it does not perform well under the specific conditions usually found in the production of configurable vehicles: high variability of individually configured orders with high volume of data, complex structures and frequent changes by engineering and customers. The Rapid Planning Matrix (RPM) is built to address this issue. It is a product-related database that is stored in-memory and provides two matrices for requirements planning:

The component matrix maps all required components that stored as rows of in matrix to the orders which are stored as columns of the matrix. The component matrix is populated in three steps. In the first step, the assembly view of the PVS is converted into a flat list of master data that includes material numbers, quantities, selection conditions, and effectivity parameters. In the second step, the assembly orders are included (due date, quantities, and configuration). Explosion of the BOM represents the third step and required components are indicated as an X for the respective order. As one cell in the matrix is represented by one bit, the RPM makes very efficient use of memory to store the component requirements for all orders.

The activity matrix for calculating the required activities per order follows the similar logic like the component variant matrix: it stores the orders as columns of the matrix and the activities as the rows of the matrix. This way is allows for the calculation of actual resource requirements per station, aggregation of all activities which are carried out at a station, matching of actual working hours with working hours available and the production scheduler can adjust resource availability to requirements

Depending on the purpose there the matrices can be read and interpreted horizontally or vertically. Reading the component matrix vertically allows for the determination of components per order, the BOM, and the component requirements. Writing to the matrix vertically allows for MRP, backflush, and impulse activities, as well as sequencing activities and the generation of sequenced JIT calls.

Reading the component matrix horizontally, that is, per activity/component enables the determination of all component requirements. Reporting functionality based on this includes a stock-requirements lists, coverage calculations, delivery scheduling, and pull lists. Time buckets can be created for component requirements.

The component and activity matrices show the exact requirements per order. Through the assignment of activities to the line, the matrices can also determine exactly when these inputs are required. If a vehicle needs a certain component, these inputs (resource and material) will be required when the vehicle reaches the appropriate workstation.

Action Points: These serve as Production Confirmation and can report production statuses. They can be integrated with the Production Action Handler for enhanced shop-floor monitoring, functioning as both reporting and tracking points.

For costing purposes, a so-called product cost collector is used. While each vehicle produced in this MTO-Scenario is managed by sales order and after production the vehicle is posted to sales order stock, the product cost collector is generic at material level. Consequently, all vehicle costs are captured in one cost collector and can only be analyzed on a period basis rather than for every individual vehicle.