Product mix determination under uncertainty within a framework proposed for effective product management

Thesis Type: Doctorate

Institution Of The Thesis: Orta Doğu Teknik Üniversitesi, Faculty of Engineering, Department of Industrial Engineering, Turkey

Approval Date: 2013




In many real life problems, uncertainty is a major complexity for decision makers. A typical example to such a case is product mix problem. In this study, we develop a methodology to aid the decision makers in product mix determination at the strategic level of product management under uncertainty. The methodology is based on a simulation optimization approach by which scenarios are generated using a statistical design of experiment approach. To the best of our knowledge, this methodology developed to aid the decision maker in product mix determination is a novel and original approach. Our product mix model determines “how many” to produce from each product for each market where it will be sold. The decision maker questions the financial performance (profit) of the company by the results of the model. The product level is considered as product line and/or family since the product mix problem is handled at the strategic level of the product management framework. Depending on the best product mix and expected financial performance (profit) brought by the mix, the decision makers may choose to change their candidate product set and re-use our approach to find a new optimal product mix and its expected profit. In that sense the method developed in this study aids the decision maker by answering several “what-if” questions such as what profit level is obtained if the set of candidate products is changed, what happens to the profit level if a new market entrance is considered with the existing products, or if market conditions are volatile, what is the effect of these conditions to the level of profit, and so on. The model can also be used for budgeting purposes considering product breakdown and market disaggregation if and when necessary. The variants of the model are presented to serve these purposes. This information can be used as an input for aggregate production planning (APP) in which deterministic forecasts of demand for the aggregate products are used traditionally. In that sense, our method improves the traditional production management information system in APP. Further research directions involving extensions of the model and the solution approach are provided. Key words: product-mix, uncertainty, product management, simulation