Multi-robot Assembly Cell Scheduling Problem With Limited Buffer Constraints

The multi-robot assembly cell has been widely used in practical production due to its high degree of automation,high flexibility and the ability to adapt to the changing market environment of modern society.However,when constructing the multi robot assembly cell scheduling problem model,the buffer capacity in the cell is usually assumed to be infinite,and the effect of buffer capacity constraint on multi robot assembly cell scheduling is ignored.In addition,the scheduling optimization of assembly and handling operations at the same time can effectively improve the efficiency of multi-robot assembly cell.Therefore,multi-robot assembly cell scheduling problem with limited buffer constraints is proposed.The scheduling objective is to minimize the makespan,select the cache capacity reasonably,and optimize the scheduling of multi-robot assembly cell.The main contents of this paper are as follows:(1)The static and dynamic scheduling models of multi-robot assembly cell scheduling problem with limited buffer constraints are established.According to the characteristics of multi-robot assembly cell,the layout structure within the cell is constructed,and the principle of robot handling between processes is proposed.According to the move J motion mode,the handling time is calculated,and the static scheduling model of multi-robot assembly cell scheduling problem with limited buffer constraints is established.Based on the static scheduling model of multi-robot assembly cell,the dynamic scheduling model of multi-robot assembly cell is established by considering the processing methods of three common dynamic events:workpiece delay,robot fault and new order insertion.(2)Aiming at the multi-robot assembly cell scheduling problem with limited buffer constraints,an improved genetic algorithm based on critical path is proposed to solve it.Firstly,according to the characteristics of the problem,a three-layer coding method and an insertion decoding method are proposed.Secondly,aiming at the deficiency of genetic algorithm in population diversity and local search ability,two initialization methods are defined in the initialization stage,which are positive process and negative process,and the positive process initialization method is used to initialize the population.In the selection stage,the population diversity threshold is set.When the ratio of population diversity to population size is lower than the threshold,a supplementary population is generated to improve the population diversity.In the crossover stage,a crossover method based on critical path is proposed to improve the efficiency of population gene exchange;In the mutation stage,a mutation method based on critical path is proposed to improve the local search ability of the algorithm.(3)The improved genetic algorithm is used to solve the multi-robot assembly cell static scheduling problem with imited buffer constraints.Firstly,the static scheduling algorithm of multi robot assembly unit considering cache constraint based on improved genetic algorithm is proposed.Secondly,through the LA01 standard test algorithm and the orthogonal test method,the influence trend of each parameter on the performance of improved genetic algorithm is analyzed,and a group of optimal parameters is selected as the most follow-up experimental parameters.Then,one of the improved operations is deleted step by step,and the deleted algorithm is compared with the improved genetic algorithm to verify the effectiveness of each improved operation.Finally,the cellular particle swarm optimization algorithm,variable neighborhood search algorithm and standard genetic algorithm are used as the comparison algorithms of the improved genetic algorithm.Through 20 groups of standard test examples of classic flexible job shop,the experimental results show that the improved genetic algorithm has stronger solving ability.In the case of multi robot assembly cell scheduling problem with limited cache constraints,the improved genetic algorithm also achieves better results by comparing the above four algorithms.(4)The dynamic events of multi-robot assembly cell in three kinds of dynamic events,including workpiece delay,robot fault and new order insertion,are studied,and the rescheduling strategies corresponding to the dynamic events are proposed respectively.The continuous dynamic scheduling problem is divided into a series of discrete static scheduling problems.Then,combining the improved genetic algorithm with the dynamic event rescheduling strategy,a dynamic scheduling algorithm based on the improved genetic algorithm is proposed,and the above algorithm is used to optimize the solution of three dynamic problem cases,which verifies the effectiveness of the rescheduling strategy in solving the dynamic scheduling problem of multi robot assembly unit with limited cache constraints.