Advanced Materials Research Vols. 655-657

Abstract: Bottlenecks heavily affect the material flow and workloads of assembly lines, and eventually constrain the productivity. The most economic way to eliminate or mitigate bottlenecks of a specific assembly line is to smooth the workloads at different stations towards a balanced system. This paper formulates a smoothness index, by which to measure the uniformity of workloads at serially located stations. To minimize the smoothness index, a Memetic algorithm is adopted and coding/decoding methods of the algorithm are devised. The effectiveness of the proposed algorithm is validated by a comparison with computing results of Package Lingo 11.0.

Abstract: In this paper, a hybrid population-based incremental learning algorithm (HPBIL) is proposed for solving the m-machine reentrant permutation flow-shop scheduling problem (MRPFSSP). The objective function is to minimize the maximum completion time (i.e., makespan). In HPBIL, the PBIL with a proposed Insert-based mutation is used to perform global exploration, and an Interchange-based neighborhood search with first move strategy is designed to enhance the local exploitation ability. Computational experiments and comparisons demonstrate the effectiveness of the proposed HPBIL.

Abstract: Cooperation between schedulers plays a key role in supply chain manufacturing. We studied the assembly scheduling system where the non-dominator suppliers cooperate in a bargaining way. Regarding different combinations of the bargaining power factors, the efficiency of cooperation was investigated by numerical experiments. Study shows the relation between the bargaining power and the efficiency.

Abstract: Scheduling is an effective optimization methodology which has been widely used for production planning. This paper presents a scheduling model to optimize the output of an assembly line in F Semiconductor Company in Tianjin, China. The authors formulate the optimization problem as linear programming. The model and its implementation are described in detail in this article. The optimum production allocations have been founded by the scheduling model and the output has been increased.

Abstract: Resource conflicts have new characteristics under the condition of virtual manufacturing cell. The existing model tools have a shortage in solving the resource conflict detection of virtual cell. So this paper puts forward a new method of resource conflicts detection on virtual manufacturing cell. According to the resource plan of productive task, Bayesian network model of virtual manufacturing cell was established. We raised the model of resource conflicts occurring on the basis of the Bayesian network. We detected the position and quantity of the resource conflicts in the Bayesian network in accordance with the model of resource conflicts occurring. Then we used an example to verify this method. The results show that this method is feasible and effective. The method lays the foundation of the specific resolution of the resource conflicts on virtual cell.

Abstract: Assembly accuracy has a crucial impact on the movement, load and fatigue of gear drivetrain. In this paper, a methodology on the accuracy prediction and control of complex drivetrain is presented. The proposed approach tries to simplify the assembly tolerance computation but grasping main assembly error factors in the assembly process of gear drivetrain. Jacobian -Torsor method is used in the analysis process, and results show a proper performance which indicates a potential opportunity for further deep and wide research on the 3D assembly analysis and simplification method.

Abstract: How to quickly improve the production level and market supply capacity, how to decrease the production time to prepare are the key issues to Automobile manufacturing enterprises. Digital factory provides a good solution to this critical problem. This article describes the use of the benefits of digital factory technology and put forward how to use digital factory technology in the automobile manufacturing process planning and production line teaching.

Abstract: In order to buffer the randomness failure of automated production car body shop is growing tendency to set buffer between lines to solve this problem. Then how to design the buffer capacity is the problem. This article did analysis of the basic planning data of body shop, then provides how to build simulation model and how to design the buffer capacity base on different planning phases through simulation analysis.

Abstract: Open shop scheduling problem is a typical NP problem with wide engineering background. It is of importance with respect of theory and application. In this paper, a mixed integer programming model was established with the objective to minimize the makespan based on the characteristics of the open shop, and a evolution genetic algorithm(EGA) was proposed. The representation of chromosome used in this paper was composed of two layers: operation layer and machine layer, which can be encoded and decoded easily. In generating initial population, DS/LTRP heuristic was used in order to improve the quality of the population. And particular crossover operation was proposed, which generated multiple offspring at a time, so that the efficiency of the algorithm can be well improved. At last, the proposed algorithm was tested on taillard benchmark, and numerical example showed good result.