Production Forecasting

Production Forecasting

• Forecasting is an estimate of the level of demand for a product over a certain period of time in the future.
• A forecast should cover a time period at least as long as the period of time required to make a decision and to put that decision into effect.

Purposes of Forecasting

• Determine the necessity for and the size of plant expansion (Facility Forecast).
• Determine intermediate planning for existing products to be manufactured with existing facilities.
• Determine short-time scheduling of existing products to be manufactured on existing equipment (Product Forecast).

Forecasting for a New Product

• Difficult task as no past information is available to predict the future.
• Methods:
  • Direct Survey method: approach a representative sample of customers to predict future demand.
  • Indirect Survey method: predict customer attitude and behavior through salesmen, agents, wholesalers, retailers, etc.
  • Comparing with Established Product: compare sales figures with an existing similar product.
  • Limited Market Trial: test the product in a limited market to gauge acceptance.

Forecasting for an Established Product

• Methods:
  • Related Information method: find an index that directly varies with sales volume (e.g., birth rate for baby food sales).
  • Market Research: analyze marketing forces, socio-economic pressures, and changing patterns to predict future demand.
  • Sales Force Composite method: gather opinions from salespeople to estimate future sales.
  • Jury of Executive Opinion method: gather opinions from experts on future sales.
  • Projection Method: project future sales based on historical data using time series analysis or correlation and regression analysis.

Time Series Analysis

• A time series is a chronological data with a dependent variable (e.g., sales, production figures) and an independent variable (unit of time).
• Components of a time series:
  • Long period changes (Trend)
  • Short period changes (Seasonal, Cyclic, Irregular)
• Methods of Estimating Trends:
  • Methods of Inspection or Freehand methods
  • Methods of Averages:
    • Selected point method
    • Semi-avg.method
    • Moving Avg.methods
  • Statistical methods

Statistical Methods

• General Approach to Statistical Forecasting:
  1. Plot demand versus time.
  2. Determine which statistical technique to use.
  3. Evaluate the expected error.
  4. Decide whether to use the technique or find a better one.

Moving Average Forecaster

• A moving average can be used as a forecaster to eliminate periodic fluctuations in a time series.
• Advantages:
  • Gives a good picture of general long-term movement in data.
  • No personal prejudice or bias.
  • Cyclic fluctuations are completely eliminated if the period of moving average is equal to the period of cycles.
  • Simple to use without fitting a curve.
• Disadvantages:
  • Tendency to cut out corners, resulting in loss of data at the ends.
  • No mathematical equation for forecasting.
  • Sharp turns in the graph are reduced to small curvatures.
  • Care must be taken in selecting data and the period of moving average.

Exponential Smoothing Method

• Used to determine the validity of a forecast by comparing it with new observations.
• Conditions to determine the validity of a forecast:
  1. Of three successive points, are two or more in either region A?
  2. Of five successive points, are four or more in either region B?
  3. Are the eight successive points on either side of the center line?