Case Questions 1 Is seasonal exponential smoothing the best

Case Questions

1. Is seasonal exponential smoothing the best model for forecasting Urban Run athletic wear? Why?

2. Explain what has happened to the data for Urban Run. What are the consequences of continuing to use seasonal exponential smoothing? What model would you use? Generate a forecast for the four quarters of the fourth year using your model. Determine your forecast error and the inventory consequences.

3. Is exponential smoothing with trend the best model for forecasting five-pocket cargo jeans? Why?

4. What method would you use to forecast monthly cargo jean demand for the second year given the previous year’s monthly demand? Explain why you selected your approach. Generate the forecasts for each month of the second year with your method. Determine your forecast error and the inventory consequences.

D operations-Manageme r ×\\ 353 / 826 CASE: Bram-Wear Lenny Bram, owner and manager of Bram-Wear, was analyzing to be with the quantities ordered by the buyers. Specifically, the performance data for the men\'s clothing retailer. He was con- problem centered on two items: an athletic shoe called Urban cerned that inventories were high for certain clothing items, Run and the five-pocket cargo jeans. meaning that the company the need for significant markdowns. At the same time, it had run ried by Bram-Wear for the past four years. Quarterly data for the out of stock for other items early in the season. Some customers past four years are shown in the table. The company seemed to appeared frustrated by not finding the items they were looking always be out of stock of this athletic shoc. The model used by for and needed to go elsewhere. Lenny knew that the problem, buyers to forecast sales for this item had been seasonal exponen- though not yet serious, needed to be addressed immediately would potentially incur losses due to Urban Run was a popular athletic shoe that had been car- tial smoothing. Looking at the data, Lenny wondered if this was the best method to use. It seemed to work well in the beginning Background Bram-Wear was a retailer that sold clothing catering to young urban, professional men. It primarily carried upscale, casual to a forecasting problem. When the product was introduced last attire, as well as a small quantity of outerwear and footwear. Its year, it was expected to have a large upward trend. The buy- success did not come from carrying a large product variety, but es believed the trend would continue and used an exponential from a very focused style with an abundance of sizes and colors. thing model with trend to forecast sales. However, they The data for the five-pocket cargo jean seemed also to point seemed to have too much inventory of this product. As with the Bram-Wear had extremely good financial performance over the past five years. Lenny had attributed the company\'s success to a group of excellent buyers. The buyers seemed able to accurately target the style preferences of their customers and correctly forecast product quantities. One challenge was keeping up with customer buying patterns and trends. Urban Run athletic shoe, Lenny wondered if the right forecasting model was being applied to the data. It seemed he would have to dig out his old operations management text to solve this problem. Demand for 5-Pocket Cargo Jeans Year 1 Year2 Demand Demand Month To determine the source of the problem, Lenny had requested forecast and sales data by product category. Looking at the sheets of data, it appeared that the problem was not with the specific styles or items carried in stock; rather, the problem appeared Demand for Urban Run Athletic Shoe Year 2 Year 3 Year 4 Year 1 Demand Demand Demand Demand Quarter

Solution

1. Seasonal exponential smoothing is not the best model for forecasting urban run athletic wear because it is only used for non-seasonal time series forecasting and resulted in out of stock. This means that the method attributes exponentially reducing weights.

2. Using the method, the data for urban run athletic has been smoothened and gives the wrong forecast. As the method shows the significant failure, the continuing use of this method would lead to forecast more erroneous and gives lower stock level.

I would use the simple moving average method.

In this method forecast error would not be determined as no actual data exists to compare it. Therefore, the inventory consequences cannot be identified.

3. Exponential smoothing with trend is not the right method to use forecasting five- pocket cargo jean because the forecast level of stock are too high. This mislead the users of the information and result in having too much inventory of the product.

4. I would use the Simple moving average method to forecast monthly cargo jean.

Demand for 5-pocket cargo jeans

In this method, the forecast error would not be determined as no actual data exists to compare it. Therefore, the inventory consequences cannot be identified.

	Year 1	Year 2	Year 3	Year 4
Quarter	Demand	Demand	Demand	Demand
1	10	14	20	22
2	29	31	26	35
3	26	29	28	28
4	15	18	30	30