Determining Optimal Product Mix of Semiconductor Products Using DEA-super Efficiency Model and Apriori Algorithm

Product mix, which means the list of products provided to the market, is an important factor affecting a company's profitability. A company should identify the customer's preferred product attributes, reflect the attributes in the production and arrange the optimal product mix. The semiconductor product is a product which has complex manufacturing process. In order to efficiently utilize resources and improve the overall profitability of investment for semiconductor products, the optimal product mix should be set. This study proposes a method using DEA-super efficiency model and Apriori algorithm to determine the optimal product mix by calculating the efficiencies and deriving the customers' purchasing preference. The proposed method uses the DEA-super efficiency model to calculate the efficiencies of the product attribute types and the efficiencies of the customer groups. In addition, the Apriori algorithm is used to reflect the customers' preferred attribute type for determining the product mix. In this study, we assessed the practical application's feasibility of the proposed method by applying it to a Korean semiconductor manufacturing company. The result shows that the proposed method can effectively determine the optimal mixing ratio of semiconductor products.