Introduction to Demand Forecasting:

Forecasting provides an estimate of future demand and the basis for planning and sound business decisions. Since all organizations deal with an unknown future, some error between a forecast and actual demand is to be expected. Thus, the goal of a good forecasting technique is to minimize the deviation between actual demand and forecast. Since a forecast is a prediction of the future, factors that influence demand, the impact of these factors, and whether these factors will continue influence future demands must be considered in developing an accurate forecast. In addition, buyers and sellers should share all relevant information to generate a single consensus forecast so that the correct decision on the supply and demand can be made. The benefits of a better forecast are lower inventories, reduce stock outs, smoother production plans, reduced costs and improved customer service. (Wisner, Tan, & Leong, 2008)

The impact of a poor communication and inaccurate forecast resonates all along the supply chain and results the bull whip effects causing stock outs, lost sales, high cost of inventory and obsolesce, material shortages, poor responsiveness to market dynamics, and poor profitability. (Wisner, Tan, & Leong, 2008)


Project Objective

  • Import and qualify SAP data to determine

    • Material Sales

    • Material stock levels, stock safety levels and lead times

    • Data validity

  • On data processing, the system will forecast sales on a national and plant level and determine

    • Anticipated stock out dates (if no supplier orders made)

    • Proposed order dates based on forecast sales, order quantities and stock safety levels

  • Provide reports and graphs to present detailed data, low stock, stock outs and proposed orders.

  • Provide an alerting system to manage critical events

  • Propose whether to order from supplier or inter-plant transfer based on the various plant stock levels and associated demand forecasting

  • Monitor inter-plant transfers to assess efficiency of material movements (if possible)

  • Forecast console to monitor accuracy of sales forecast, effectiveness of the proposed orders, stock outs, returns, expiring stock etc.

Project Status

The project is in a development stage and active developement was started on the 26th August 2019.

Forecasting

Forecasting is currently based on 5 available methodologies, namely:

  • Holts
  • Dampled Holts
  • Multiplicative Exponential
  • Damped Multiplicative Exponential
  • Holts Winter (Triple Exponential) with seasonal capability.

Currently, the development is focusing on monthly periods and a national overview to provide a demonstration platform within a one month period. Once this is completed, it will also support analysis over weekly periods and include plant views.