The Manufacturing Logistic Management Information System And Optimized Model Based On The Combination Of MRP & JIT

With the rapid development of the economic globalization and information technology, the modem logistics is rising in the whole world, which is generally thought as "the third profit source" except for reducing material consume and raising labor productivity. Material Requirements Planning (MRP) and Just-In-Time (JIT) are two production management methods, which have widely been applied in the dispered manufacture at present. They play an important role in reducing middle inventory and decreasing occupation of floating capital and increasing reliability of supply in enterprise to some extent, but as far as the whole production logistic process organized and controlled by enterprise is concerned, they are only partial-approaches to logistic control, and they have their own advantages and disadvantages. Only through integrating them into an organic whole, can the benefit and adaptation of the whole production process be remarkably improved. Thus, integrating MRP with JIT represents the trend of model of production logistic system in modem manufacture.This paper, based on CYYC Management Information System, does research on the theories of MRP and JIT, which direct the system design. In order to solve the problem how to combine MRP with JIT, this thesis proposes a Markov Decision Process (MDP) model to optimize the hybrid Push/Pull control strategies. The integral goals, the systematic holistic structure, and functional model tree of the system are raised in this dissertation. Meanwhile based on the analysis of requirement for logistics system, combining with model-based design methodology, the paper presents the analysis and design of all sub-system for logistics information, which includes process model. Furthermore, to counter the uncertainty of the processing time in the practical production process, this paper builds a single machine earliness/tardiness with fuzzy processing time. The simulation result proved the convergence speed of the algorithm is quick. And the algorithm has beneficial influences to solve the production scheduling.