Cellular manufacturing is a principal application of group technology in which machine cells and part families are generated based on their similarity in the production process to minimize overall movement cost and maximize machine utilization by using complex mathematical programming procedures or computer tools with a lot of computational effort and time to solve problems. In this study, the clustering analysis based on a similarity coefficient is developed to efficiently solve cell formation problems in both single and multiple process routings. A novel similarity coefficient is developed to integrate operation sequence especially adjacent operation, processing time, production volume, machine capacity, and multi-visits to minimize the number of actual inter-cell moves and voids in machine cells. An improved clustering algorithm is proposed for grouping machines into cells and simultaneously determining the machine sequence in cells to reduce intra-cell moves as well as selecting the best process routing for each part. The practical effectiveness of the proposed method is demonstrated through computational experiments involving eighteen test instances, varying in scale from small to large problems. When compared to other complex methods, the proposed approach not only enhances overall group technology efficacy but also significantly reduces computational time, making it a highly promising and practical solution for addressing cellular manufacturing challenges.