This paper advances the use of computers for industrial engineering applications by developing an approach to integrating manufacturing-based group technology (GT) concepts with material requirements planning (MRP) systems. Component parts having similar processing requirements are collected into sets, or “part families”. A case study of gear manufacturing at Black· Decker shows that this approach allows for substantial reductions in setup costs when the elements of the part family are combined into a single batch for processing. The problem is complicated by the fact that the setups between components in the part family were found to be sequence dependent. An optimal dynamic programming formulation for this problem is developed to minimize the setup and carrying costs. When compared to an optimal policy for scheduling the part requirements independently of one another, as is normally done in MRP systems, the optimal group scheduling algorithm saves 23% (on average) of the setup and carrying costs. A lot sizing heuristic for part families was developed as an extension of the material requirements planning technique of part-period-balancing. The heuristic was found to be within 5% of the optimal solution (on average). The results of this research suggest that many of the production economies of group technology can be realized in the job shop by better production planning, without uprooting machines to form machine cells.