Analysis of Issues To address the two major issues, the application will be illustrated using the current problem faced by the company in deciding what to send to the manufactures for the first production run for 10 styles of women’s parkas. The forecast by the committee presents the estimate by 6 members on the potential demand which totals up to 20,000 parkas. The requirement for the first production order is 10,000parkas. First, the quantity for each style needs to be decided upon. The deciding factor given the current situation should be to minimize the risk of overproducing. Overproducing as stated earlier resulted in a salvage value that comes at loss of about 8% of the wholesale price. Essentially minimizing the risk of overproducing is also minimizing the risk of making a loss. Assuming normal distribution in the demand forecast, the approach taken is through the use of the formula: (average demand forecast + standard deviation*k) which essentially aims to reduce overproduction. The average demand forecast is the simple average of the 6 individual member estimates and the standard deviation used is twice of the standard deviation based on 6 individual members past experience. The k value is a variable that will be adjusted for all styles in order to establish the minimum order requirement of 10,000 and basically denotes the no. of standard deviations to the left of demand. The higher the k value means less the risk of overproducing. The model will also favor products that exhibit low coefficient of variation.

Referring to the appendix, Table 1
Illustrates the quantity for each style utilizing the highest k value of 1.0607 that can be achieved without any limitations on where to produce. Assuming no limitations, the chance of overproducing any product is about 15%.
In Table 2, we look at the production solely in Hong Kong where the minimum requirement