Spare Parts Planning (SPP) is the planning module of the SAP SPM solution integrated within the Supply Chain Management (SCM) Business Suite of SAP. It has been specifically designed to meet the requirements of the spare parts business, among which the most significant ones are listed below. Many other industries can also take advantage of the numerous SPP advanced planning features.
• Nowadays customers expect more and more a high service level regarding the service parts.
• Aftermarket companies are typically dealing with a very high number of SKU’s( Stock Keeping Unit ). Several hundreds of thousands active product references are not unusual.
• The distribution network, and as a consequence the planning processes, are more complex than in a normal supply chain. Dynamic distribution networks, horizontal moves and re-manufacturing processes, among others, are standard attributes of service parts supply chains.
• Single customers can typically be supplied by multiple distribution centers depending on the parts and their dynamic stocking policies.
• High variability of the parts portfolio in term of volume and demand patterns, physical dimensions, price.
Service parts planning is concerned with the forecasting , inventory planning , procurement and distribution of service parts to the customer facing location in order to keep the target service levels . The process for service parts planning are shown in figure 1.1 and range from capturing of the demand to the planning of the procurement and stock transfer.
Fig 1.1 Overview of service Parts Planning Processes
Bill-of-Distribution (BOD): a New Planning
One cannot talk about SAP Spare Parts Planning without presenting what the BOD concept is. The well-known BOM (Bill-of-Materials) which specifies the list of parts used to build a product has now its distribution counterpart; the list containing the structure and the locations used to distribute a product is saved as a new master data element and called Bill-of-Distribution (BOD). This concept is central in the Service Parts Planning solution and all the planning functionalities, without any exception, are linked to this main product attribute. This feature acts as a decoupling point between the planning processes and the physical distribution network which, in the aftermarket, typically requires regular modifications. This BOD-based architecture allows the planning supply chain model to be dynamic and adaptive to the market conditions. This flexible feature actually assumes that each part belongs temporarily to a pre-defined distribution network, and that this network can be, transparently for a user, replaced by another one neither without harming any planning process nor without the need for any manual cumbersome maintenance.
Figure: Examples of BOD's
Contract Packager Concept
In many service parts companies activities like packaging and/or repackaging within the network are outsourced. SAP SPP enables an easy activation or change between several contract packagers at location product level. The BOD is then extended to the packager which can be defined on product location level. This additional network function is transparently integrated within the planning calculations (e.g. lead times, deployment…).
A Starting Cycle
There is no single starting point in the functional flow of SPP, but if there were one, it would be somewhere on the cycle represented on the following figure. On one hand, the sales orders of the service parts business are captured on the right location in SPP (Capture Demand function) whereas onthe other hand the Stocking/ Destocking feature determines which one is the right location.
Capture Demand records each order item on the best facing location depending on defined rules and dynamic decisions (e.g. Stocking/Destocking). The rules can be static and very straightforward like a geographical assignment based on the closest delivering plant, or can be much more advanced and supported by dynamic rules together with gATP, the Global Available-To-Promise tool of SAP. The idea that an order can be assigned to different locations is directly impacting the operational flow; it also means that a single order can potentially be delivered from diverse sources to the final customer. This is where the second half-cycle, Stocking/Destocking, answers the question: “Where should an order supplied from to reach a certain customer?” This rule-based tool decides automatically for each part where it needs to be stocked or not, and by consequence where it needs to be delivered from. The dynamic nature of an aftermarket supply chain requires that kind of decision much more than for a traditional supply chain for which a single plant is classically delivering all the customers in a given geographical zone around it. In order to optimize the distribution of the numerous SKU’s of an aftermarket organization, this automated “relocation” is a must. Although very flexible, the rules at the origin of the stocking decisions are mainly based on the combination of volume and cost. Of course additional rules critical to the industry are foreseen like for instance the exclusion of items based on regulations or hazardous conditions. The usual heavy, expensive and slow moving parts of the automotive industry like the body shells or the engines are not forgotten by this process: a Procure-to-Order scenario can also be automatically assigned (no stocked location in the BOD) when the service level becomes less critical compared to the inventory carrying cost or the obsolescence risks.
Balancing the Demand
When Stocking/Destocking stops the planning (destocking decision) of a product location, the associated demand history is transferred to another facing location of the BOD in order to still generate the right level of forecast for the customer base even though no longer from the same location. The demand history, and consequently the service, is simply switched from a warehouse to another in order to optimally deliver the parts. This mechanism, called demand realignment, is following the same principle as the one used during capture demand and ensures that each sales order is always optimally assigned within the supply chain. Next to the Stocking/Destocking decisions, other planning processes can trigger the realignment of the demand history like a modification of the BOD assignment, a promotion or a supersession chain.
Balanced History to Balanced Forecast
The sections above explain why, how and when the demand history is captured and realigned depending on the planning decisions. But this process would be meaningless without an ultimate goal which is the generation of the forecast. The guarantee that the demand is at any given moment in time recorded on the right location signifies that the forecast is generated always in the locations where the goods will be physically delivered from. The SPP suite contains nine predefined forecasting models (based on the realigned demand history), which are capable of accurately forecasting (if possible) the future demand of parts according to different types of demand patterns such as constant, trend, seasonal, intermittent and exponentially declining. Some models are part of the eminent algorithms known by all within the supply chain planning world while others have been specifically designed by aftermarket industry insiders or adapted from existing models. The forecast process is by far the most comprehensive (not to say complex) of the solution and encompasses, among others, advanced tools like an automatic model selection, composite forecasting run, rough and fine tuning of model parameters, outlier corrections, trend limitations, parameters inheritance, tripping, triggs tracking, adaptive initialization of models, disaggregation throughout the BOD, manual and automated approval, recalculation in the past for optimized forecast error calculation.
Economic Order Quantity (EOQ) and Safety Stock
As briefly noted in the previous section, SPP provides a recalculation of the forecast in the past methodology which gives as an advantage to always have an accurate forecast error evaluation (standard deviation) regardless of the forecast model change, realignment of demand history or other modifications of network lead times.
This standard deviation is directly used by the EOQ and safety stock module which generates these two values in a time-phased manner, and most important, in an interdependent way. The algorithm is based on the optimization of the total stockholding and procurement cost depending on the forecast, forecast error, target service level, lead times and probability distributions. An automated segmentation of the parts portfolio is executed regarding both the target service levels and the statistical distributions used. The first one is defined based on flexible rules in order to meet the service policies of the company while the second one is optimally assigned depending on the demand pattern of the part location. Additionally the service level is also dependent on the lifecycle of the parts which allows using a higher service level during the demand peaks, as shown on the Figure 1.2: adaptive target service level.SPP inventory planning is based on the normal probability distribution as well as the Poisson distribution which is particularly important to deal accurately with the abundant slow movers/lumpy demand items.
Figure 1.2: adaptive target service level
Supersession in a Nutshell
Supersession planning handles the replacements (complex or simple) and discontinuations of parts during their lifecycle. Typically the supersession chains create a lot of confusion within the aftermarket industry – as well as others – mainly due to the fact that the chains can contain a huge number replacement and that they have an impact on all the departments, from the warehouse to the senior management. SPP gives the visibility regarding the supersession chains and handles them efficiently in each planning process, thanks for instance to the demand realignment.
From Tactical Planning to Operational Planning
So far the main tactical planning functions of SPP have been described. The outcome of these activities are used by the operational planning tasks in order to, on one hand, generate the appropriate procurement proposal, and on the other hand to manage the movements of parts inside the distribution network from the suppliers to the customers. The first action is performed by the module “Distribution Requirements Planning” (DRP) which uses the BOD as the main structure to roll up the demand from facing locations to entry points in the network. Its calculations are based on the forecast, external demands, EOQ and safety stock values as well as on data coming from the operational system (ERP) such as the stock figures. DRP provides also the following interesting features:
• Various planning Horizons and stability options which ensures the alignment of the planning flow and the supplier contracts
• Anticipated demand coverage (DRP provides the possibility to smooth seasonal demand patterns by pulling ahead peak season demand into lower seasons)
• Pre-season safety stock planning
• Product group procurement optimization
• Supplier shutdown handling
• Virtual orders consolidation (for slow movers)
• Planning of remanufactured products
• Advanced approval rules
Through the BOD
The creation of the Stock Transfer Orders (STO) are generated by two processes: the first one considers the vertical top down movements throughout the BOD (deployment) while the other one takes care of the horizontal requirements coming from potential unbalances inside the network. Deployment differs in some ways from the classical APO deployment. Among others the fact that the deployment is done day by day without any creation of orders in the future or the use of priority tiers for the demand in case of shortage to guarantee an optimized use of the available supply. Next to that, the deployment can also generate expedite shipments when the business conditions require it. As every planner knows, planning is mainly dealing with uncertainties. When these uncertainties unbalance inventories, the inventory balancing tool of SPP provides an automated way to overcome these situations by generating horizontal movements within a BOD. Although this flow results into more expensive operations, it can be very useful when the cost of a stockout is higher than an additional STO. Of course inventory balancing provides the possibility to define the balancing areas based on the regulations or geographical zones for instance, and can be easily used as an analysis tool thanks to its embedded approval functions.
Life Cycle Management
Given the huge number of SKU’s, the warranty agreements and the demand patterns of typical aftermarket parts, surplus and obsolescence issues are common. In addition to the supersession planning, SPP provides a Surplus and Obsolescence tool which handles the life cycle of the parts (end of life). Old parts produced in the eighties are no longer stored forever thanks to an automated determination of the surplus stock based on the long-term forecast and the retention policies of the companies. Moreover SPP determines when is the best time to remove a part from stock depending, among others, on the market responsiveness. Since the obsolete stock is an important source of cash loss within the parts business, the potential return of using such a tool is very high.