Research On Production Optimization And Simulation Technologies Of Mixed-model Assembly With Real-time Status Information

As the production style of modern discrete manufacturing industry generally adopted, mixed-model assembly lines (MMALs) produce a variety of product models similar in construction and configuration concurrently, facing with a more complex assembly process. The core issue of production management of MMALs such as assembly line sequencing, material handling are, therefore, more complex than the general assembly production. Focusing on the MMALs of JIT (Just in Time) production environments, the assembly line sequencing problem, material handling problem based on the real-time production status information and simulation method of rapid resonse capability to market for MMALs are discussed in this dissertation.Firstly, the thesis provides a systematic literature review on the key technology about MMALs, the influence of assembly line sequencing to MMALs is analyzed. The material distribution pattern at the present is discussed, the current existing problems in automobile assembly line are pointed out and the relevant solutions are given.Then, in order to guarantee the production sequence to satisfy the material usage leveling, minimizing the conveyor stoppage time and setup time incurred by adjustment for model changing, a multi-objectives sequencing model is formulated. An improved genetic algorithm is proposed to solve the multi-criteria model; the computational results show that multi-objective sequencing can meet the objects of material usage, the minimum conveyor stoppage time and minimum setup time.The backward method for getting production status information brings the problem of lacking standardization of information transfer and lower information quantity. To apply the RFID (Radio Frequency IDentification) technology into MMALs is proposed in this dissertation, then the model of production management of MMAL is studied, on this basis, its frame and integration operation mode are established. Through the real-time tracking for product information based on RFID technology, the passive material handling progress is going to turn into positive service and integrates with production process information, which effectively improves the accuracy and agility of material handling progress in the production of MMALs. An algorithm for determining the best time of delivery driver to start the delivery trip based on minimizing of total time penalty function is proposed. This approach makes the material amount of a single delivery trip maximization and improve the efficiency of delivery drivers significantly in the case of does not affect the assembly line normal throughput.In terms of problem of the material requirement is hard to combine with production assembling process, in the basis of the manufacturing bill of material (MBOM), the delivery bill of material (DBOM) considering the material requirement of assembly station is formed. It not only can get material consumption planning according to the takt-time, but also deduce material requirement in different times based on the unit material consumption and production sequence. The material requirement planning under the real-time production status can be used to reasonably arrange the material delivery vehicle routing. The optimization model of VRPTW for material delivery of automobile general assembly line is established, the experimental results of the standard test show the validity and feasibility of the proposed impoved general algorithm, and then the algorithm is used for solving delivery vehicle routing problem.To build simulation model for the manufacturing system rapidly and improve the reusability, based on the organizational forms of resource and the control logic of model, a new modeling and simulation approach based on data-driven and hierarchy control model is proposed. The model generator is combined with data-driven method using the attribute values in the data structure of the simulation model to instantiate those hierarchy model templates and the simulation model for given applied object is built, rapidly realize the adjustment of manufacturing resource and repartition of stations, achieving the simulation analysis for manufacturing system.After that, to validate and demonstrate the application of the results above, the designed corresponding production management system has been successfully applied to a manufacturing enterprise's automobile MMAL. Finally, some conclusions are drawn with some of the further research directions being anticipated.