Research On The Job Scheduling Problem Of The Automobile Engine Under The Mixed Model Assembly Environment

With the rapid development of science and technological progress and social economy, people’s demand for personalized and diversified products also continued to improve. For the automotive industry, people urgently want to have a distinct personality characteristic of the product, while the core of the car is the engine manufacturer. To respond to these needs of personalized and diverse,now, the domestic automobile engine manufacturing most have adopted mixed mode manufacturing assembly lines to produce more varieties of products instead of manufacturing a single product. For mixed assembly line applications focused on how to get a reasonable sort of production makes the production more balanced, reduced inventory, and to ensure product delivery, so research products mixed assembly line production sequencing issues, can play mixed assembly line advantage of the mode of production better, thereby increasing production efficiency.According to the production of the actual situation and comparing the previous results of single-stage production process, such as product assembly stage or research parts processing stage, we research the product’s job scheduling problem in the mixed assembly environment throughout the production process from the work order to the parts machining. This paper adopts the delivery period, parts supply leveling and minimum switching time of three objective functions, and established the corresponding mathematical model.This paper proposes an improved genetic algorithm to solve mixed assembly process problems, which has already been proved to be NP-hard combinatorial optimization problems. In the first stage of the algorithm applicates delivery dates and parts consumption levelized solving multi-objective optimization product production order, and using its results as the ordering constraints for the second stage of machining parts. The proposed algorithm, on the one hand, uses of non-dominated solutions complement iteration offspring solution set to ensure the superiority of offspring individuals; on the other hand, joins the local search algorithm to avoid falling into local optimal algorithm, and full use of the genetic algorithm randomness and global convergence properties of itself, to ensure that the algorithm is effective and efficient. Finally, an example of the algorithm by simulation experiments, through the process of problem solving simulation analysis proved the effectiveness and practicality.