Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Forecasting Chapter 11
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Outline A Forecasting Framework Qualitative Forecasting Methods Time-Series Forecasting Moving Average Exponential Smoothing Forecast Errors Advanced Time-Series Forecasting Causal Forecasting Methods Selecting a Forecasting Method
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc A Forecasting Framework Focus of the chapter Difference between forecasting and planning Forecasting application in various decision areas of operations (capacity planning, inventory management, others) Forecasting uses and methods (See Table 11.1)
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Use of Forecasting (Table 11.1) Operations Decisions
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Use of Forecasting (Table 11.1) Marketing, Finance, HRM
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Qualitative Forecasting Methods Major methods: –Delphi Technique –Market Surveys –Life-cycles Analogy –Informed Judgement Characteristics of the methods (see Table 11.2)
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Time-Series Forecasting Common components in time-series (see Figure 11.1): –Average –Seasonality –Cycle –Trend –Error (random component) “Decomposition” of time-series
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Simple Moving Average: Weighted Moving Average: Moving Average
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Simple Exponential Smoothing: Smoothing Coefficient (alpha) determination Determination of the initial forecast Exponential Smoothing
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Time-Series Data Plot (Figure 11.2)
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Exponential Smoothing Basic logic: The forecast
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Forecast Errors Cumulative Sum of Forecast Error (CFE) Mean Square Error (MSE) Mean Absolute Deviation (MAD) Mean Absolute Percentage Error (MAPE) Tracking Signal (TS)
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Forecast Errors: Formulas Cumulative sum of Forecast Errors Mean Square Error Mean Absolute Deviation Mean Absolute Percentage Error Tracking Signal Mean Error
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Advanced Time-Series Forecasting Adaptive exponential smoothing Comparison of time-series forecasting methods (see Table 11.5) Box-Jenkins method
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Causal Forecasting Models The general model: Other forms of causal model (see Table 11.7): –Econometric –Input-output –Simulation models –Others
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Example of Causal Method Y t = a + b(t) F 7 = (7) =
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Selecting a Forecasting Method User and system sophistication Time and resource available Use or decision characteristics Data availability Data pattern
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Graphical Comparison Moving average method with various n ME
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Graphical Comparison Moving average method with various n MAD
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Graphical Comparison Moving average method with various n MSE
Irwin/McGraw-Hill The McGraw-Hill Companies, Inc Graphical Comparison Moving average method with various n MAPE