A COMPUTATIONAL APPROACH OPTIMIZING PUSH/PULL FLOW IN AN AEROSPACE TRANSMISSION OVERHAUL SHOP

摘要

This paper investigates the possibility of applying hybrid push/pull production control strategies in a transmission overhaul shop. The shop currently uses MRP/push system. The high variability in process times and limited resources available make it difficult to deliver products on time. A simulation model of a horizontally integrated hybrid system which is "half push and half pull" representing the shop processes is developed and used to evaluate system performance. The system can be optimized by determining points of integration, optimal values of safety stock for the push part and number of Kanbans for the pull part. The proposed method employs genetic algorithm driving the parameterized simulation model. Two research questions are answered in this paper. First, what is the best way to find good solutions to this type of design problem? Second, what programming elements can be combined to create some kind of tunable simulation model of this class of complex system of mixed manufacturing philosophies. A new production control strategy that will reduce the inventory holding cost and cost due to missed due dates is recommended to this shop.