1. Trang chủ
  2. » Luận Văn - Báo Cáo

Lecture Introduction to operations management - Chapter 11: Forecasting

27 19 0

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

THÔNG TIN TÀI LIỆU

Thông tin cơ bản

Định dạng
Số trang 27
Dung lượng 1,21 MB

Nội dung

In this chapter we will discuss: A forecasting framework, qualitative forecasting methods, time-series forecasting, moving average, exponential smoothing, forecasting errors, advanced time-series forecasting, causal forecasting methods, selecting a forecasting method.

Operations Management Contemporary Concepts and Cases Chapter Eleven Forecasting McGraw­Hill/Irwin Copyright © 2011 by The McGraw­Hill Companies, Inc. All rights reserved Chapter Outline A Forecasting Framework Qualitative Forecasting Methods Time­Series Forecasting Moving Average Exponential Smoothing Forecasting Errors Advanced Time­Series Forecasting Causal Forecasting Methods Selecting a Forecasting Method Collaborative Planning, Forecasting, and Replenishment 11­2 A Forecasting Framework Focus of chapter is on forecasting demand for output  from the operations function – Demand may differ from sales Difference between forecasting and planning – Forecasting:  what we think will happen – Planning:  what we think should happen Forecasting application in various decision areas of  operations (capacity planning, inventory  management, others) Forecasting uses and methods (See Table 11.1) 11­3 Use of Forecasting: Operations Decisions Time Horizon Accuracy Required Number of Forecasts Long Medium Single or few Top Qualitative or causal Long Medium Single or few Top Qualitative and causal Medium High Few Middle Causal and time series Short Highest Many Lower Time series Inventory management Short Highest Many Lower Time series Process design Capacity planning, facilities Aggregate planning Scheduling Management Forecasting Level Method 11­4 Use of Forecasting: Marketing, Finance & HR Time Horizon Accuracy Required Number of Forecasts Long Medium Single or few Top Qualitative Short High Many Middle Time series New product introduction Medium Medium Single Top Qualitative and causal Cost estimating Short High Many Lower Time series Capital budgeting Medium Highest Few Top Causal and time series Long-range marketing programs Pricing decisions Management Forecasting Level Method 11­5 ‘Qualitative’ Forecasting Methods Based on managerial judgment when there  is a lack of data.  No specific model Major methods: – – – – Delphi Technique Market Surveys Life­cycles Analogy Informed Judgment (naïve models) 11­6 Time-Series Forecasting Components of time­series data: – – – – – Average level Trend—general direction (up or down) Seasonality—short term recurring cycles Cycle—long term business cycle Error (random or irregular component) “Decomposition” of time­series – Data are decomposed into four components Moving averages Exponential smoothing 11­7 Moving Average Assumes no trend, seasonal or cyclical  components Simple Moving Average: At Ft Dt Dt N Dt N At Weighted Moving Average: Ft At W1 Dt W2 Dt WN Dt N 11­8 Moving Average Compute three period moving average (number of periods is the decision of the forecaster) Period Actual Demand    1 10    2 18    3 29    4   ­ Forecast      19 (10+18+29)/3 = 19 Period 5 forecast will be (18+29+actual for period 4)/3 11­9 Figure 11.2: Time-Series Data Note: The more periods, the smoother the forecast 11­10 Exponential Smoothingcalculation Facts: – – – – September forecast for sales was 15 September actual sales were 13 Alpha (α) is 0.2 What is the forecast for October? Calculation: – October forecast = September forecast +  α(September actual­September forecast) =15+0.2(13­15)=15+0.2(­2)=15­0.4=14.6 11­13 Forecast Errors In addition to the forecast, one should  compute an estimate of forecast error.  Its  uses include: To monitor erratic demand observations or  “outliers” To determine when the forecasting method is no  longer tracking actual demand To determine the parameter values that provide  the forecast with the least error To set safety stocks or safety capacity 11­14 Forecast Errors Cumulative Sum of Forecast Error (CFE)  Mean Square Error (MSE) Mean Absolute Deviation (MAD)—measure of  deviation in units Mean Absolute Percentage Error (MAPE) Tracking Signal (TS)—relative measure of  bias Mean Error 11­15 Forecast Errors: Formulas Cumulative sum of Forecast Errors n CFE =  i=1 n Mean Square Error Mean Absolute Deviation et MSE =  t e i=1 n n MAD =  i=1 Mean Absolute Percentage Error n MAPE =  i=1 n Tracking Signal | TS  =  et i=1 MAD n | et | n Mean Error et | 100 Dt n et ME =  i=1 n 11­16 Tracking Signal Analogous to control charts in quality  control, viz. if there is no bias, its values  should fluctuate around zero.  Is a relative measure, i.e., the numbers  mean the same for any forecast 11­17 Advanced Time-Series Forecasting Adaptive exponential smoothing – Smoothing coefficient ( ) is varied Box­Jenkins method – Requires about 60 periods of past data 11­18 Time Series vs Causal Models Time series compares data being forecast over  time,  i.e.,  time is the independent variable or  x­axis or x­variable Causal models compare data being forecast  against some other data set which the  forecaster may think is a cause of the  forecasted data, e.g., population size causes  newspaper sales 11­19 Causal Forecasting Models The general regression model: yˆ a bx Other forms of causal model: – – – Econometric Input­output Simulation models 11­20 Example of Time Series Model Yt = a + b(t) t Intercept (a) Slope (b) Dt 120 124 119 124 125 130 Ft 119.52 121.18 122.84 124.5 126.15 127.81 129.47 Dt = actual sales Ft = forecasted sales t = time period (e.g. year) 117.8667 1.657143 F7 = 117.87 + 1.66 (7) = 129.47 = sales forecast for next year 11­21 Example of Causal Model Yt = a + b(t) It 34.6 35.7 36.3 35.2 35.7 36.4 37.6 Intercept (a) Slope (b) Dt 120 124 119 124 125 130 Ft 121.15 123.79 125.22 122.59 123.79 125.46 128.34 Dt = actual sales in year t Ft = forecasted sales It = median family  income (000’s) 38.23094 2.396514 F7 = 38.23 + 2.397 (7) = 128.34 = sales forecast for next year (year 7) 11­22 Selecting a Forecasting Method User and system sophistication – People reluctant to use what they don’t understand Time and resources available – When is forecast needed? – What is value of forecast? Use or decision characteristics, e.g., horizon Data availability and quality Data pattern Don’t force the data to fit the model! 11­23 Forecast Horizons and Forecast Accuracy The longer the forecast horizon, the less  accurate the forecast Long lead times require long forecast horizons Lean, responsive companies have the goal of  decreasing lead times so they are shorter than  the forecast horizon 11­24 Collaborative Planning, Forecasting and Replenishment (CPFR) Aim is to achieve more accurate forecasts Share information in the supply chain with  customers and suppliers Compare forecasts – If discrepancy, look for reason – Agree on consensus forecast Works best in B2B with few customers 11­25 Summary A Forecasting Framework Qualitative Forecasting Methods Time­Series Forecasting Moving Average Exponential Smoothing Forecasting Errors Advanced Time­Series Forecasting Causal Forecasting Methods Selecting a Forecasting Method Collaborative Planning, Forecasting, and Replenishment 11­26 End of Chapter Eleven 11­27 .. .Chapter Outline A? ?Forecasting? ?Framework Qualitative? ?Forecasting? ?Methods Time­Series? ?Forecasting Moving Average Exponential Smoothing Forecasting? ?Errors Advanced Time­Series? ?Forecasting. .. Advanced Time­Series? ?Forecasting Causal? ?Forecasting? ?Methods Selecting a? ?Forecasting? ?Method Collaborative Planning,? ?Forecasting,  and Replenishment 11­2 A Forecasting Framework Focus of? ?chapter? ?is on? ?forecasting? ?demand for output ... Forecasting? ?application in various decision areas of  operations? ?(capacity planning, inventory  management,  others) Forecasting? ?uses and methods (See Table 11.1) 11­3 Use of Forecasting: Operations Decisions Time Horizon

Ngày đăng: 25/05/2021, 21:40