OPERATIONS MANAGEMENT for MBAs Fourth Edition Meredith and Shafer I see that you will get an A this semester. John Wiley and Sons, Inc. Topic 2S: Forecasting Chapter 8S : Forecasting
Outline Overview Purpose of forecasting Choosing a forecasting technique Time series methods Causal / associative methods Homework
To plan capacity, we need to estimate demand Trends Facilities Price Advertising Staff Competition Purchasing Quality Location
Purpose of Forecasting Market justification decision Capacity and scheduling decisions Long, Mid, & Short Term
Choosing a Forecasting Method If historical data is available, use it. Plot the data first. If there is a trend, use a technique based on the trend. If there is seasonality, use a technique that accounts for this. If there are known associations, use a causal technique. Choose a technique which gives the smallest mean absolute error. Software and data mining may help.
Some Types of Forecasts Qualitative Time Series Moving Averages Exponential Smoothing Seasonal Causal / Associative Simple Regression Multiple Regression
Time Series Forecasting Time series data is simply a set of values of some variable measured at regular intervals over time. One data set (variable) over time. Based on historical data. The more data the better. Assumption: Past behavior helps us predict future behavior. Time series data can have one or more of the following components / factors / variations. Trend Seasonal Cyclical Random
Time Series Forecasting Quantity | | | | | | | | | | | | Months Quantity Time Year 1 Year 2 Quantity | | | | | | Years
Time Series Forecasting
Time Series Forecasting Techniques Moving Averages Exponential Smoothing Seasonal Methods
Moving Averages Uses an average of past periods actual demands.
Exponential Smoothing
Exponential Smoothing Date Close ES (alpha 0.1) ES (alpha 0.9) ES (alpha 0.2) May-07 21.18 Jun-07 22.67 Jul-07 22.56 21.33 22.52 21.48 Aug-07 24.71 21.45 21.69 Sep-07 24.81 21.78 24.49 22.30 Oct-07 25.81 22.08 24.78 22.80 Nov-07 25.13 22.45 25.71 23.40 Dec-07 25.69 22.72 25.19 23.75 Jan-08 20.33 23.02 25.64 24.14 Feb-08 19.36 22.75 20.86 23.37 Mar-08 20.53 22.41 19.51 22.57 Apr-08 21.58 22.22 20.43 22.16 May-08 22.6 21.46 22.05 Jun-08 20.95 22.20 22.49 Jul-08 21.64 21.10 21.92 Aug-08 22.44 22.03 21.59 21.86 Sep-08 18.34 22.07 22.35 21.98 Oct-08 15.73 21.70 18.74 21.25 Nov-08 13.66 16.03 20.15 Dec-08 14.51 20.36 13.90 18.85 Jan-09 12.77 19.77 14.45 17.98 Feb-09 12.74 19.07 12.94 16.94 Mar-09 15.03 18.44 12.76 16.10 Apr-09 15.62 18.10 14.80 15.89 Forecast 17.85 15.54 15.83
Exponential Smoothing
Multiplicative Method
Multiplicative Method Year1 Year2 Year3 Year4 Yr5Forecast Q1 45 70 100 132.82 Q2 335 370 585 725 843.62 Q3 520 590 830 1160 1300.03 Q4 170 285 215 323.52 Totals 1000 1200 1800 2200 2600 Average 250 300 450 550 650 SFYr1 SFYr2 SFYr3 SFYr4 AvgSF 0.18 0.23 0.22 0.20 1.34 1.23 1.30 1.32 2.08 1.97 1.84 2.11 2.00 0.40 0.57 0.63 0.39 0.50
Simple Linear Regression Multiple Regression causal / associative forecasting
For b, use exponential smoothing with smoothing constants of 0.1 and 0.9. Part c. Compare the 3 methods using mean absolute deviation. Part d. change to: for each method what is your sales forecast for next January (period 13)? Omit e.
Additional Homework 1 D/FW Auto Imports (DAI) sells a variety of very high-end automobiles to customers in the Dallas-Fort Worth metroplex. Weekly demand for DAI’s 2011 BMW M3 has been increasing rapidly. High customer satisfaction is important to DAI and managers are wondering about relationships between customer satisfaction and a variety of service factors in the automobile customer preparation. Once a car is sold, auto preparation is quite extensive. Preparation involves 3 major steps; Performance Checking, Paperwork Preparation, and Detailing. If one resource is utilized (one person and a desk or a dock), the times for each of these processes is as follows: Performance, 12 hours Paperwork, 15 hours Detailing, 8 hours Weekly BMW M3 demand for the past 9 weeks has been; 9, 12, 15, 16, 20, 26, 39, 38, and 43. Weekly customer satisfaction ratings were 7.9, 7.6, 8.3, 8.1, 8.5, 8.6, 9.1, 7.9, and 7.7. Quality goals were met in every week except for weeks 4, 7, and week 9. Develop a spreadsheet and multiple regression analysis that will address the following issues: Assuming that each week 1 additional resource is added to handle demand, determine where resources should be added, determine the preparation operational weekly capacity, efficiency, and utilization. For a preparation operation with an efficiency of 0.90, utilization of 0.80, and weekly quality goal met, predict that week’s customer satisfaction rating. Which factor is most important in customer satisfaction (efficiency, utilization, or quality goal)?