Improved Leapfrog Algorithm For Solving Optimization Problem Of Variable Production Scheduling

The scheduling optimization problem in the Flexible Flow Shop Scheduling Problem(FFSP)is a classic problem in the field of scheduling optimization problems.The FFSP scheduling problem involves parallel machine allocation and workpiece on-line sorting.It is difficult to solve and is a classic NP-hard problem[1-4].Flow Shop Scheduling Problem With Variable Processing Time(FFSP-VPT)is a research hotspot in the field of scheduling optimization problems in recent years.In the classical scheduling problem,the processing time of the workpiece is generally regarded as It is a "rigid" parameter that cannot be changed.In the flexible flow shop scheduling optimization problem during the work,the processing time of the workpiece is a variable parameter with "flexibility" characteristics,that is,according to the requirements of the user and the order,The machining time of the workpiece is selected within the range.As the processing speed increases,the quality of the product will decrease,resulting in an increase in the rework rate.The burden on the production of the workshop will increase,which will increase the completion time of the process.The balance between the processing time and the rework rate of the unqualified product is The research focus of flexible flow shop in variable working time.The main research contents of this paper are as follows:(1)In the process of codec,a two-stage coding method based on ROV coding is proposed,which decodes the individual into the workpiece sequence and the work order selection sequence,according to the first in first out(FIFO)principle and the most First free machine priority(first available machine,FAM)principle for scheduling.The standard leapfrog algorithm is used to solve the continuous problem.The two-stage coding method makes the leapfrog algorithm more suitable for solving the discrete scheduling optimization problem.(2)In the global optimization algorithm,the Shuffled Frog Leaping Algorithm(SFLA)is used as the global optimization algorithm.The leapfrog algorithm has the characteristics of strong global optimization and high accuracy.The maximum update distance in the standard leapfrog algorithm is set to a fixed value,which causes the algorithm to fall into local extremum.This paper introduces the update distance into the adaptive factor to prevent the algorithm from falling into Locally optimal..(3)By analyzing and improving the shortcomings of the leapfrog algorithm,an adaptive leapfrog algorithm based on elite individual sets is proposed for the FFSP-VPT problem.Finding the optimal solution can also result in fewer iterations.In order to make ASFLA play a better effect,the chaotic mapping method produces a relatively uniform initial population,improves the quality of the initial population,designs a crossover test,and obtains the optimal value of the algorithm parameters,and optimizes the SFSP-VPT scheduling problem for the ASFLA algorithm.The performance test verifies the effectiveness of the leapfrog algorithm in solving the problem of flexible flow shop scheduling optimization during variable working hours.(4)Establish a mathematical model of flexible flow shop in variable time,define parameters,assume several variables,determine basic constraints,study the relationship between processing time and rework rate of unqualified products,establish quality inspection links,establish rework rate and processing time The calculation model determines the rework workpiece processing rules.(5)An initial population establishment method based on minimizing the maximum completion time optimization target is proposed to improve the quality of the initial solution in the initial population generation process of the leapfrog algorithm,and the improved leapfrog algorithm based on the optimization target(I-ASFLA)is validated.Solve the effectiveness of the variable-flow flexible flow shop problem(FFSP-VPT).