Operations Management / Supply Chain Management

Module 03.01 Key Concept: Demand Management and Forecasting

Demand Management and associated planning and execution of supply to support demand, requires intimate knowledge of customer needs.  This means that communication throughout the supply chain is key.  The extent and frequency of communication will depend on the characteristics of each industry / market segment / customer.  Consumer product markets typically require daily interaction through fully linked systems while process industries may rely predominantly daily interaction with inter-company organizations and less frequent contact with 3rd party customers.  In the latter case, the customer base likely includes other intermediate suppliers rather than end-product customers.  Take for example BASF – remember their commercials “We don’t make the cars, we make the cars ……er (faster, safer, etc.).”

Forecasts are developed in Demand Management to predict the quantities and timing of what might occur in the marketplace.  We will explore different forecasting methods later in this Module.

Supply plans are then developed to specify how the firm will respond to these forecasts.  These plans may look significantly different from the forecast–because of other considerations (operational efficiency and cost, staffing, capacity, raw material / component availability etc.)  The authors of the text propose: “managers can’t be held responsible for forecast errors, but should be held responsible for failing to execute the plan.”  There is a common process developed by W. Edwards Deming (W. Edwards Deming Institute) with four simple steps:

 Plan > Do > Check > Act

You first create a plan (demand plan, supply plan, distribution plan, etc.).  Then you implement the plan and monitor (metrics) the results.  Finally, you evaluate the results and take action by changing the plan, acquiring / eliminating resources, etc. to improve subsequent results.

When we evaluate forecasting methodologies we will also evaluate different approaches to monitoring and dealing with forecast-to-actual discrepancies.

The forecasting process involves much more than just the estimation of future demand.  The forecast also needs to take into consideration the intended use of the forecast, the methodology for aggregating and dis-aggregating the forecast, and assumptions about future conditions.  All of this depends on the type of business and the intended level of Manufacturing Planning and Control for its usage.

The forecast information and technique must match the intended application.

  • Strategic Planning requires highly aggregated estimates for determination of long-term capacity / resource needs.
  • Sales & Operation Planning requires more detailed forecasts in terms of product families and time periods.
  • Detailed Planning and Control require highly detailed forecasts, to cover a short period of time


Forecasts in Strategic Planning:

Forecasts for Strategic Planning are usually presented in general terms (sales dollars, tons, hours).  Financial forecasts are shown in pro-forma income statements, balance sheets and other such financial reports.

The aggregation level may be related to broad indicators (gross national product, income).  Causal models and regression/correlation analysis are typical tools.  Qualitative forecasting methods may be frequently employed due to the longer-range planning horizon involved.  Managerial insight is critical and top management involvement is intense.

Forecasts are generally prepared annually and covers a period of years


Forecasts in Sales & Operations Planning:

Forecasts for Sales and Operations Planning are usually presented in aggregate measures (dollars, units).  Dollarized forecasts are used in concert with the business plan to predict future financial performance, check actual performance vs. plan and develop corrective actions from a financial standpoint.  Unit forecast are used in the development of production plans at the family level with subsequent establishment of inventory and backlog plans (depending on make-to-stock or make-to-order strategy respectively).

The aggregation level is related to product families (common family measurement).  Forecasts are typically generated by summing forecasts for individual products.  Managerial involvement is moderate and limited to adjustment of aggregate values.  The forecast are generally prepared several times each year and covers a period of several months to a year or more depending on the planning horizon for major raw materials / components and finished goods.


Forecasts in Detailed Planning:

Forecasts for Detailed Planning are usually presented in terms of individual products (units).  However, in an assemble-to-order environment, forecasts may be created for product different product modules to drive planning for sub-assemblies / components.

Forecasts are typically generated by mathematical procedures, often using software.  Projection techniques are common.  The assumption is that the past is a valid predictor of the future.  Top Managerial involvement is minimal.  Forecasts are updated frequently (the text says “almost constantly”.) and cover a period of days or weeks – depending on the Master Production Scheduling Process and production / procurement lead times.  As stated for Sales & Operations Planning, forecasts must extend far enough in the future to cover planning for all raw materials, intermediates / sub-assemblies and finished products.

There are two basic types demand:

  • Independent Demand – Customer demand that is not directly influenced by the actions of the firm (e.g. customer orders)
  • Dependent Demand – Demand that is driven by the plans and activities of the firm (e.g. components, warehouse demand)

Although there are exceptions, normally, independent demand is forecasted and dependent demand is not.  Instead, dependent demand is calculated / derived from some type of follow-on process such as Material Requirements Planning, Distribution Requirements Planning, Kanban, etc.

Let’s look at a practical example of Independent vs. Dependent Demand.  The following slides show some familiar products from a familiar global company – McDonalds.  McDonalds has many different product families but we will use one for illustration purposes.

Take a look at the following slide and think about the definitions for independent and dependent demand.


Consider the following questions:

Is the Happy Meal independent or dependent demand?  What about the hamburger that goes into the Happy Meal?  What type of demand do the french fries represent?  What about the toilet tissue, oil, and mustard?

You are probably thinking… “Who Cares?”

Well, actually I am sure the supply planner at McDonalds cares.  She / he might like to have a forecast for happy meals in order to determine daily requirements for ground beef, pre-frozen hamburgers or whatever.  She / he may want to know how many printed, paper drink cups to order.   What items would McDonalds forecast?

They would forecast anything that is sold direct to their customers.  This means that some items are only considered independent demand, others are considered strictly dependent demand and still others can be considered both depending on usage.  This is summarized in the following slide.



There are seven basic steps involved in any Forecasting Process:

  • Determine the use of the forecast
  • Select the items to be forecasted
  • Determine the time horizon of the forecast
  • Select the forecasting model(s)
  • Gather the data needed to make the forecast
  • Make the forecast
  • Validate and implement results