Part IV. Actual APS and Case Studies
22.4 The Demand Planning Process of the Styrene Plastics Division
The goals of the Demand Planning in the Styrene Plastics Division are the
• influence of the market behavior,
• using of sales departments’ knowledge of customers’ ordering behaviour,
• coordination of decentralized regional sales departments,
• ability to react quickly to short-term market fluctuations and
• creation of a midrange demand plan in order to calculate a production and procurement plan.
To achieve these goals, input from marketing department, sales depart- ment and logistics department is needed. The marketing department adjusts market quantities and influences prices directly or indirectly. The sales de- partment is responsible for customer relationships and the “recording” of demands and placing the orders. Sales and logistics negotiate sales budgets to support “management by objectives.” The logistics department is respon- sible for the consolidation of the sales and marketing plans, the short-term adjustments and the creation of one single demand plan from the various plans.
416 Boris Reuter
The following plans are created:
• sales budget,
• rolling business forecast,
• marketing plan,
• sales plan and
• short-term adjustments.
Referring to the structure of Chap. 7 (p. 140) only the sales plan is sup- ported by a statistical forecasting method. The sales budget is the result of a negotiation process; the marketing plan is created in the marketing de- partment and entered manually; the rolling business forecast is also entered manually and based on the experience of the planner. Starting from statistical forecasting, the sales plan is revised and can be modified (revised judgmen- tal forecast). From the different plans, one consensus-based demand plan is calculated.
The monthly planning cycle consists of the following steps:
1. The sales plan is statistically forecasted and revised by the regional sales departments. The plan contains demands for each stock keeping unit and customer and determines the portions of each characteristic combination.
2. Marketing decides on the total quantities that should be sold to the mar- ket, neither to decrease the price by shipping too much nor to loose market share by selling too little.
3. The total of the sales plan quantities has to be adjusted to reach the total marketing quantity without changing the portions of the sales plan.
4. While the first three steps are related to months one through three , the aggregate rolling business forecast is made up for in months four through 12.
5. To release demands for 12 months to SNP (see Fig. 22.2), the first three months from the adjusted sales plan (step three) are combined with months four through 12 of the rolling business forecast.
6. Demands from the released demand plan are assigned to delivery loca- tions for an interval of 12 months.
7. Short-term adjustments can be done prior to the release to SNP.
An overview of the existing demand plans is shown in Fig. 22.2. In the following chart, the different plans are described in more detail.
Sales Budget
The sales budget is not a subject of the monthly planning cycle. The aim of the sales budgeting process is to negotiate the expected yearly quantities sold by a region or to a key account. The definition of the sales budget of the fol- lowing calendar year takes place once a year in October and remains valid for 12 months. The result agreed upon is yearly quantities at the product group
22 Demand Planning of Styrene Plastics 417
year month
Sales Budget (1x per year)
Rolling Business Forecast (monthly)
Marketing plan (monthly)
Sales Plan (monthly)
: Date of planning
Released Demand Plan (monthly) Short-Term Adjust-
ments (monthly)
Fig. 22.6. Overview of the different demand plans
and region level. The negotiating partners are the logistics department of the business unit and the responsible regional managers of the sales department.
The plan helps to keep track of the achievement of sales objectives and thus to shift priorities. It is also used to guide the promised product quantities – allocations – along the sales budgets. A special feature is an automated disaggregation procedure on the time structure that distributes the yearly quantity of the sales budget on monthly quantities. The distribution is based on the sales history of the previous year. This is done by the disaggregation rule of the sales budget based on the key figure “sales history.” Thus, seasonal effects are taken into account.
Rolling Business Forecast
In accordance to the midterm plans of the business unit, the logistics depart- ment enters the aggregate business forecast of the business unit manually.
The definition or update of the business forecast for the next 12 months is performed each month in the so-called rolling business forecast. The result is a highly aggregated plan for the complete business unit on a monthly ba- sis. The plan is needed to support 12 months production planning with SAP APO SNP. As the figures are highly aggregated on the product and customer structures, they have to be disaggregated for planning purposes. The sales budget is used as a reference key figure for the disaggregation.
418 Boris Reuter Marketing Plan
The aim of the marketing plan is to influence prices in the oligopoly market by short-term adjustments of the expected sales quantities for the different product families. The plan is made up for three months on a monthly basis.
Because of the global impact of the quantities sold to the market, a differen- tiation on regions or key accounts is not useful. The marketing department is responsible for the creation of a marketing plan. The quantities are used later in the adjustment process (see Short-term adjustments and plan con- solidation).
Sales Plan
The sales plan is the central object of forecast collection and input of the sales department. It contains the forecast quantities for the next three months on a monthly basis for each stock keeping unit and each customer. The planning complexity is reduced by an ABC-classification of customers and stock keep- ing units; thus, only the most important items and customers are planned manually, all others are forecasted automatically. The forecast process can be classified as a revised judgmental forecast (see Sect. 7.3). The sales planner is supported by the historical sales quantities of the last two years as well as by forecasts calculated with a simple moving average method, whereby more than six months of historical sales quantities are averaged. Because of the absence of more complex regular patterns, the calculated forecast gives an impression of the expected quantities. The variations in the historical time series, e. g. reduced sales caused by vacation, are easily identified by the plan- ner in a year-to-year comparison for each month and do not justify a more complex statistical model. Due to handling constraints, minimum delivery quantities have been taken into account for each combination of stock keep- ing unit and customer. Minimum quantities are automatically identified. If customer orders are already placed in a period, the forecasted quantities have to be at least as high as the ordered quantities, or else they are also identified.
Short-Term Adjustments and Plan Consolidation
Before the release of the demands to the finite capacity planning, various plans are consolidated and short-term adjustments are made.
One of the most important adjustments is matching the marketing plan with the sales plan. Here, the structure of the forecasted demand portions of the sales plan is combined with the quantities of the marketing plan. Con- sequently, the total quantity shipped to the market should not lead to an unwanted behavior, e. g. a decreasing price or loss of market share. Thus, quantities might change, but not portions. The calculation uses the disaggre- gation rule “based on a key figure,” where the reference key figure is the sales plan.
22 Demand Planning of Styrene Plastics 419 Another important adjustment is the adding of the delivery location. As shown above the forecast is made on stock keeping units and customers, whereas the sourcing location is missing. In the case of multiple production sites, customer demands could be produced at and shipped from different sites. Subject to transportation and production costs, the choice of the as- signed location has an impact on the cost structure. Of course, transfers between production sites are possible, but they cause transportation costs.
Thus, the assignment of demands depends on the product (single production site or multiple production sites) and customer site. If the planning region of the customer could be supplied from multiple locations, the ZIP-Code of the customer is used for assignment to the location. Those rules are also used for a first allocation within global ATP checks.
In between a planning cycle, demand figures may change. An instant reaction is supported by a direct communication process with the logistics department and a separate key figure for short-term adjustments to allow for the monitoring of changes.
After the adding of the location to the demand data, the release to Supply Network Planning is performed and production planning can be started.