A two stage algorithm to design and assess product family

Evrim Yuzgec, Azmi Alazzam, Nagendra Nagarur

Research output: Contribution to conferencePaperpeer-review


This paper presents a two-stage algorithm to design product families and allocate products to assembly lines. Product family design is triggered by the existence of large variety of products in a typical company portfolio. The literature shows that physical similarity and process similarity are the most important factors in product family design. The objective of this paper is to classify products based on their physical appearance such that similar products can be manufactured in the same assembly line. The proposed model starts with determination of similarity ratio among products using Jaccard index. Afterwards, an algorithm is developed for the product allocation to the assembly lines in a way that each assembly line indicates a product family. Additionally, weighted subjective product score is calculated. Cluster analysis is used and products are clustered using weighted subjective product score. Comparison of the cluster analysis and proposed model using Jaccard index are presented and discussed. It is concluded that classification of the products using Jaccard index improves overall efficiency of product lines and gives an efficient preliminary rough cut capacity planning.

Original languageEnglish
Number of pages9
StatePublished - 2012
Event62nd IIE Annual Conference and Expo 2012 - Orlando, FL, United States
Duration: 19 May 201223 May 2012


Conference62nd IIE Annual Conference and Expo 2012
Country/TerritoryUnited States
CityOrlando, FL


  • Cluster analysis
  • Jaccard index
  • Jaccard similarity coefficient
  • Product allocation and production planning
  • Product family design
  • Similarity index


Dive into the research topics of 'A two stage algorithm to design and assess product family'. Together they form a unique fingerprint.

Cite this