{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# 3 - 4 Interactive Models for Operations and Supply Chain...

This preview shows page 1. Sign up to view the full content.

This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 4 Interactive Models for Operations and Supply Chain Management The Simple Exponential Smoothing Model Simple exponential smoothing is a popular forecasting technique, particularly when trying to smooth out the effects of random ﬂuctuation. Simple exponential smoothing is actually a sophisticated weighted average technique that allows the user to specify the relative weight given the immediate past demand and the demands from the more distant past. The greater the weight on past demands, and the less the weight on the most recent demand, the less the impact of recent changes on the forecast. The lower the weight on the most recent demand, the greater the smoothing effect, and the less the forecast responds to demand changes. The responsweness 0f the forecast is controlled by the size of 0L (alpha). A larger or results in a greater weight being placed on the most recent demand and less on the previous forecasts. Exhibit 1.2 shows the Simple Exponential Smoothing Model. —— Demand “- Forecast EXHIBIT 1.2 Screen View of the Simple Exponential Smoothing Model Demand Forecasting 5 Interactive Case Reducing Coffee Waste by Forecasting Accurately A local branch ofﬁce of a large consulting ﬁrm has gourmet coffee delivered twice a day from a local coffee shop. Coffee is delivered in large urns at 7:30 am. and again at noon. Connie Andrews, the office manager, wants to continue to provide the service as a “perk” to the employees, but she does not want to pay for coffee that is ultimately thrown away. Analysis of the coffee demand shows that over the long term there appears to be no seasonal pattern or trend. In the short term, how— ever, there are often periods of time when demand seems to be increasing, and then it shifts and decreases for a period of weeks. The most recent 12 weeks of demand are presented below. Analysis 1. Make sure that the starting default values match the demands provided above. Examine the graph of the demand for coffee. Examine how the forecast follows the demand when the alpha value is set to 0.1. a. Change the alpha value to 0.3. What happens to the forecast? Pay particular attention to the impact that the demand increase in period 2 and the demand decrease in period 7 have on the subsequent forecasts. What happens to the “response” of the forecast in periods 3 and 8 when you change alpha? b. In order to further investigate the link between the value of alpha and the responsive— ness of the forecast, with the alpha at 0.1, use your mouse to drag the demand for period 9 to 70 units. What happens to the demand forecast for period 10? Experiment by doing this for each value of alpha. How would you quantify the relationship between alpha and the impact a change in alpha has on the responsiveness of the forecast? c. Repeat your experiment, but examine the impact that a reduction in the demand at period 7 has on the forecast for period 8. How does the forecast respond to the drop in demand at various levels of alpha? 2. Make sure that all demand values are at their starting defaults. If they are not, bring them to their defaults by clicking on the “Reset” button. a. With the alpha value at 0.1, what are the current values for MFE and MAD? What do these numbers mean? ...
View Full Document

{[ snackBarMessage ]}