Layout Reconstruction And Dynamic Scheduling Optimization Of Dual Robot Assembly Cell

With the rapid improvement of productivity levels and the increasingly diversified needs of users,traditional flow job shops are replaced by flexible job shops and robot manufacturing cells,in order to meet the changeable mixed production mode of multi-variety and small batch.The dual robot assembly cell,belonging to a robot manufacturing cell that performs assembly production tasks,includes two robots which can use different parts to assemble a variety of products on a small number of facilities with different functions by changing the end clamping tools.This paper studied the layout of the dual robot assembly cell and explored the impact of layout reconstruction on assembly scheduling.the order-oriented dynamic scheduling problem,based on the reconfigurable dual-robot assembly unit,was studied to bring better solutions for actual production.The main research contents of this paper are as follows:(1)The layout reconstruction model of the dual robot assembly cell was established,in which the facilities were simplified and the facility access points were set.And the simplified physical model was used to calculate the time for the robot to move.Combined with the physical model,a mathematical model of a dual robot assembly cell under fixed production tasks was constructed.(2)In the dynamic scheduling problem,a coding method considering the due date was formulated.The due date was considered while coding,so that all products in each order should be completed before the due date of the order as much as possible,and production delays were avoided effectively by the method.(3)An M-EDA,retains the probability model of EDA,was proposed for the scheduling problem of the reconfigurable dual robot assembly cell.And in order to overcome the shortcomings of the algorithm in the local search ability,M-EDA enters sub-population search when it falls into a local optimum,and adds specific cross and mutation operations.Through algorithm comparison experiments,the distribution of optimal solutions obtained by M-EDA were significantly better than that of EDA.(4)The multi-objective optimization algorithm based on M-EDA was used to solve the reconfigurable dual robot assembly cell scheduling problem.With two objectives of the stock cost and the completion time,multi-objective M-EDA was used to optimize the production tasks of a single production cycle.Finally,the TOPSIS method was used to select a optimal solution in the Pareto frontier solution set as a scheduling and layout plan.