Reebok Case
Group 3: There are several problems faced by Reebok existing system. The first issue is that the firm may not be able to meet demand for some styles of jackets. The second issue is the uncertainty in demand for individual players and teams. Sales of jerseys depend on results of teams and the supply chain may end up having insufficient demand for some styles of jackets while others can have excess supply and become obsolete. The inventory cost can be prominent

To improve the decision making process, demand forecasts should be made. Forecasts of demand are based on newsvendor model. We can calculate the expected overage and expected loss under different order quantity so as to find the optimal quantity for overall production as well individual players.

In this study, we apply the newsvendor model for individual players and we will compute the optimal quantity based on risk-based production planning applied in Obermeyer. The newsvendor model is to maximize the inventory order level with uncertain demand but fixed prices. In the case of Reebok, it is uncertain for the manufacturer to predict the demand for jerseys for certain and certain players. The uncertainty originates from the uncertainty of the performance of teams and players in NFL. With such uncertainty, the firm is concerned in optimizing the ordering level so that the excess inventory will be obsolete whereas some demands cannot be fit

We compute a simulation using Excel for different order quantities for individual players, as well as overall jersey production. We find the optimal quantity based on this data. We use 30-days as simulation period, as this is normal sales cycle for the peak demand.

The optimal order quantity for overall team is based on Newsvendor model. The overage cost of the business is 24-11.9 = 12.1 per jersey, assuming all sales are made to wholesalers. The underage cost of the business is 11.9-7 = 4.9 per jersey. The critical ratio is hence 0.712. The corresponding optimal ordering quantity should be 98410 units. With total ordering quantity at 100000 units, the probability of in-stock is 74%

Other measures of newsvendor model are used in this analysis to help Reebok to predict optimal order quantity and expected profit under optimal order quantity. One important measure is expected loss sales, which is the product between the loss function shown in appendix and the forecasted standard deviation of demand forecast for jerseys. With expected loss sales, expected sales is computed and we can compute the expected profit based on expected sales. Based on our calculated optimal quantity based on newsvender model, it can be seen from the table below that the expected fill rate is 92% overall. For individual players, the expected fill rate ranges from 78% - 84%.

With similar approach, we can compute the corresponding optimal ordering quantity for individual players. The detailed figures are shown below. It should be noted that the probability of overage is 44% for player Smith Antowain, given a minimum order size of 1728 jerseys, as stated in the case.

It should be noted that the estimation of optimal under normal distribution is based on two parameters. The mean is based on expected forecast in the case. The standard deviation is based on doubling the standard deviation used in the model to give a conservative estimate for the expected profits.

Calculations part – Details of the calculation can be found in the Excel spreadsheet
Detailed calculations are shown below

The risk-based production planning is shown as below

Expected profit and loss for different quantity order under Newsvendor model for a 30day period using Brandy Tom as example