Chapter 4

• Forecasting


  The process of predicting a future event
• Forecasting is the underlying basis of all business decisions: inventory, personnel, production, facilities
• Forecasts time horizon types


  • Short-range forecast (usually < 2 - 3 months)
  • Medium-range forecast (3 months to 2 years)
  • Long-range forecast (> 2 years)
• Short-range forecast


  • Job scheduling, worker assignments, detailed use of system
• Medium-range forecast


  • Sales, production planning, detailed use of system
• Long-range forecast


  • New product or operations planning, design of the system
• Qualitative Forecasting Methods


  • Jury of executive opinion: Pool opinions of high-level experts, sometimes augment by statistical models
  • Delphi method: Panel of experts, queried iteratively
  • Sales force composite: Estimates from individual salespersons are reviewed for reasonableness, then aggregated
  • Market survey: Ask the customer
• Trend Component


  Persistent, overall upward or downward pattern due to changes in population, technology, age, culture, social norms, typically several years in direction
• Seasonal component


  Regular pattern of up and down fluctuations due to weather, customs, social norms, occurs within a single year
• Cyclical component


  Repeating up and down movements, multiple years in duration, affected by a business cycle, political, and economic factors
• Random component


  Erratic and unsystematic fluctuations due to random variation or unforeseen events, short duration and non repeating
• Time-series forecasting models


  • Naive Approach
  • Simple moving average and Weighted moving average
  • Exponential smoothing
  • trend projection
• Trend Projections


  Fitting a trend line to historical data points to project into the medium to long range, linear trends can be found using the least squares technique
• Associative Forecasting


  • Used when changes in one or more independent variables can be used to predict the changes in the dependent variable, most common technique is linear regression analysis
  • Forecasting an outcome based on predictor variables using the least squares technique
• Probability distribution of the estimate

  A forecast is just a point estimate of a future value, this point is actually the mean of a probability distribution
• • Naive Approach
  • Assumes “demand” in the next period is the same as “demand in the most recent period. 
  • Fast, cost effective and efficient
  • Can be a good starting point
• • Simple moving average
  • Assumes an average of the past is a good estimator of future behavior 
  • Used if little or no trend
  • Used for smoothing 
• • Weighted moving average
  • Used when some trend, or special circumstance need to be considered in the forecasting
  • Weights based on experience and intuition
• • Exponential smoothing 
  • Effect of smoothing constants
  • As alpha increases, older values become less significant so, forecast becomes more sensitive 
• • Quantitative models
  • Time series analysis
  • Regression analysis
• The multiplicative seasonal model can adjust trend data for seasonal variations in demand
• Correlation

  • How strong is the linear relationship between the variables?
  • correlation does not necessarily imply causality
  • Coefficient of correlation, r, measures degree of association (r values range from -1 to +1)
• If more than one independent variable is to be used in the model, linear regression can be extended to multiple regression to accommodate several independent variables