A business-intelligence-based concept for the identification of similar parts

A growing demand for individualized products results in an increasing number of product variations. A first step to harmonize these variations is to identify similarities between products based on their technical properties. This paper follows the design science approach: The requirements for the concept are derived from interviews with experts from German industrial companies. Then a concept is developed and prototypically realized and its usefulness is evaluated during a workshop. The approach is based on the fact that each product has a digital representation - the digital product model, which is created during the design of the product and based on feature technology. Each part is described entirely by the total amount of all features and their parameters. By extracting the features from the Computer Aided Design (CAD) system and loading them into a Business Intelligence (BI) infrastructure, it becomes possible to analyze various parts at the feature level. The BI infrastructure enables the application of data mining algorithms, which can be used in order to classify parts based on their feature combinations. Regarding new parts, this concept can support the selection of production technologies by providing information on similar parts that have been produced in the past.