A management science approach for developing a decision model for portfolio rationalization of consumer personal computers

Thesis: M.B.A., Massachusetts Institute of Technology, Sloan School of Management, 2015. In conjunction with the Leaders for Global Operations Program at MIT.Thesis: S.M., Massachusetts Institute of Technology, Engineering Systems Division, 2015. In conjunction with the Leaders for Global Operations Program at MIT.Cataloged from PDF version of thesis.Includes bibliographical references (page 47).Commoditization of the PC market has led to a shift in strategy at Dell away from customization towards defining specific consumer configurations ahead of time. This shift creates new planning and decision processes around which configurations to select. Given configurations are defined before true demand can be measured, variability leads to selection of configurations based on imperfect knowledge. The existing variety of configurations has imposed substantial complexity costs on Dell. Modeling this selection process as a Newsvendor "life-time buy" includes this variability while optimizing for profit. Estimated costs of overage, costs of underage, and complexity costs are used as inputs. This thesis seeks to develop a framework informing configuration selection decisions based on demand variability and expected profit. A heuristic model uses configuration volume estimates, profit margin estimates, and historical regional bias in submitted forecasts to score and rank potential configurations. This heuristic determines which configurations to select within a fixed maximum total number of configurations limit, based on the business's market share growth and profit goals. This maximum limit is set by Dell's Vice Chairman Jeff Clarke. The results of this heuristic show a decrease in decision process time, an increase in process transparency, and increase in expected profit. A what-if scenario analysis shows an annual increase of 3.4% in profit which amounts to several million dollars. These results were gathered through implementing the model in November 2014 and comparing the heuristic-based decisions against past process decisions.by Paul Alexander Whigham.M.B.A.S.M