This preview shows pages 1–5. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 1 1 Demand Forecasting Chapter 4 (Part 4) 2 Time Series Forecasting Models Require historical data forecasts extrapolate past data into the future Some of the most commonly used methods simple moving average weighted moving average exponential smoothing Time series forecasts have been shown to typically yield more accurate shortterm forecasts 2 3 Developing a Time Series Forecast 1. Identify time series data components 2. Choose appropriate time series methods depends on the data components present 3. Evaluate different methods calculate forecasts using historical (past) data evaluate forecasting errors for each method choose method which performs best assumption is if it works well on past data, then it will work well in the future 4. Implement method of choice 5. Monitor forecast performance 4 Identify Time Series Components Horizontal Trend T i m e Demand Seasonal 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 T i m e ( y e a r s ) Demand Cyclical 1 2 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 1 9 2 0 2 1 2 2 2 3 2 4 T i m e ( m o n t h s ) Demand T i m e Demand 3 5 Choose Appropriate Time Series Method Guide to Selecting Appropriate Time Series Methods Method Time Series Components Simple Moving Average Weighted Moving Average Exponential Smoothing Trend Adjusted Exponential Smoothing Linear Regression Seasonal Decomposition Trend Seasonal Cyclical Horizontal 6 Na ve Method The nave forecast for the next period equals the demand for the current period. This is expressed as: F t+1 = A t Example Period Demand Forecast 1 15 2 28 3 25 4 28 25 15 F t+1 ~ forecast for period t+1 A t ~ actual demand in period t 4 7 Simple Moving Average Simple moving average places the same weight (or emphasis) on each time period. Method works well when the demand is fairly stable over time. Method does not do a good job of forecasting when a trend is present. The forecast lags the actual demand because of averaging effect....
View
Full
Document
 Spring '07
 billthompson

Click to edit the document details