Forecasting

• A forecast is a statement about the future.
• Forecasts are used to help managers plan the system and its use.

Forecast Uses

• Planning the system involves long-range plans related to the types of products and services to offer, facility and equipment levels, and facility location.
• Planning the use of the system involves short- and medium-range plans related to inventory management, workforce levels, purchasing, and budgeting.

Common Features of Forecasts

• Forecasts assume a causal system where the past influences the future.
• Forecasts are rarely perfect because of randomness.
• Forecasts are more accurate for groups versus individuals.
• Forecast accuracy decreases as the time horizon increases.

Elements of a Good Forecast

• Timely
• Reliable
• Accurate
• Meaningful
• Written
• Cost-effective
• Easy to use

Steps in the Forecasting Process

• Determine the purpose of the forecast.
• Establish a time horizon.
• Select a forecasting technique.
• Gather and analyze data.
• Make the forecast.
• Monitor the forecast.

Types of Forecasts

• Judgmental forecasts use subjective inputs (qualitative).
• Time series forecasts use historical data, assuming the future will be like the past (quantitative).
• Associative models use explanatory variables to predict the future.

Judgmental Forecasts (Qualitative)

• Consumer surveys
• Delphi method
• Executive opinions from managers and staff
• Sales force input

Time Series Forecasts (Quantitative)

• Trend: Long-term movement in data
• Seasonality: Short-term regular variations in data
• Irregular variations: Caused by unusual circumstances
• Random variations: Caused by chance
• Cycle: Wave-like variations lasting more than one year

Approaches to Forecasting

• Naïve approach
• Simple Moving Average
• Weighted Moving Average
• Exponential Smoothing
• ES (Exponential Smoothing) with Trend and Seasonality

Naïve Forecast

• Simple to use
• Virtually no cost
• Data analysis is nonexistent
• Easily understandable
• Cannot provide high accuracy

Techniques for Averaging

• Moving average
• Weighted moving average
• Exponential smoothing

Points to Know on Moving Averages

• Pro: Easy to compute and understand.
• Con: All data points are weighted equally.

Disadvantage of simple linear regression

• Applies only to linear relationships with an independent variable.
• Requires a considerable amount of data to establish the relationship (at least 20 data points).
• All observations are weighted equally.

Forecast Accuracy

• Forecast error is the difference between the forecast and actual demand.
• MAD (Mean Absolute Deviation) calculates the mean absolute deviation.
• MAPD (Mean Absolute Percent Deviation) calculates the mean absolute percent deviation.
• Cumulative error
• Average error or bias

Forecast Control

• Tracking signal monitors the forecast to see if it is biased high or low.
• Tracking signal = Σ(Actual - Forecast) / MAD = E / MAD

Sources of Forecast Errors

• The model may be inadequate.
• Irregular variation may occur.
• The forecasting technique may be used incorrectly, or the results misinterpreted.
• There is always random variation in the data.

End Notes

• The two most important factors in choosing a forecasting technique are cost and accuracy.
• Keep it simple.
• Use computer software like =FORECAST(70,{23,34,12},{67,76,56}) to assist with calculations when possible.