A new methodology to cluster derivative product modules: an application

Abstract

Companies are trying several ways to offer competitive and highly differentiated products. The goal for the product platform is to share elements for common functions and to differentiate each product in the family by satisfying different requirements as much as possible. This study focuses on the product variety and short product life cycles that result from the increase and diversification in consumer needs and expectations. Proposed methodology aims to maximise the use of common product modules by considering platform-based derivative products and modular product design approaches to minimise the planning complexity in supply chain, manufacturing and service for derivative products. Functional and technical features of the products are determined in the first step. Then, design structure matrix is formed. After defining product components, similarity matrix for derivative products is formed. A clustering algorithm based on Clonal Selection is used to generate critical product modules. Data from a home appliance manufacturer are used to assess three versions of a product by also considering the production process. The grouping enabled to shorten the release time of a new derivative product to the market.

Keywords:

• clonal selection algorithm
• clustering algorithm
• commonality index
• design structure matrix
• modular design