Decisions about how much to order, when to order, and how to inventory effectively are also complicated by the rapidly changing environment within which order, inventory and supply planning is carried out.  Forecasting is very much a part of the supply management process and directly affects both quantity and delivery.


The need for raw materials, services, parts and subassemblies is usually derived from a sales forecast, which is driven by marketing.  Despite this situation, missed forecasts are often blamed for overages or shortages no matter who made the original forecast or how bad the forecast was.  The real problem with forecasts is their lack of reliability.  Forecasts will usually be wrong, but will they exceed or fall short of actual requirements, and by how much?  Continuous improvement methods can be applied to forecasting by tracking forecast accuracy and taking steps to eliminate root causes of forecast error.


  • Forecasting Techniques
    • Quantitative Forecasting – This approach uses past data to predict the future.  One class of quantitative forecasting techniques called causal models tries to identify leading indicators, from which linear or multiple regression models are developed.  A second quantitative method assumes that sales (or other items to be forecast) follow a repetitive pattern over time.  The analyst’s job is to identify the pattern and develop a forecast.


    • Qualitative Forecasting – involves gathering opinions from a number of people and using these opinions with a degree of judgment to give a forecast.  The Delphi Technique is an example.



  • Collaborative Planning Forecasting and Replenishment (CPFR) – links sales and marketing processes to supply chain planning and execution processes.  When changes in demand, promotions, or policy occur, jointly managed forecasts and plans can be adjusted immediately, minimizing or eliminating costly after-the fact corrections for both parties.