Over the past three decades, Cellular Manufacturing Systems (CMS) have attracted a lot of attention from manufacturers because of its positive impacts on analysis of batch-type production and also a wide range of potential application areas. Machine cell formation and part family creation are two important tasks of cellular manufacturing systems. Most of the current CMS design methods have been developed for a static production environment. This paper addresses the problem of machine cell formation and part family formation for a dynamic production requirement with the objective of minimizing the material handling cost, penalty for cell load variation and the machine relocation cost. The parameters considered include demand of parts in different period, routing sequences, processing time and machine capacities. In this work a new heuristic approach named PRABHA is proposed for machine cell formation and the part family formation. The computational results of the proposed heuristics approach were obtained and compared with the Genetic Algorithm approach and it was found that the proposed heuristics PRABHA outperforms the Genetic Algorithm.