Algorithm Research For Flexible Job-shop Scheduling Problem

With the rapid development of science and technology and economic globalization, theproduction scale of enterprises are also increasingly larger, the complexity of the productionprocess is increasingly high, while facing increasingly fierce market competition environment.In recent decades, the production process has also undergone significant change, mainly in thecontinuity of the complexity of the production process and production processes. In anincreasingly competitive circumstances, to ensure access to maximize the benefits ofproduction must be reasonable arrangements and use of resources, reduce the duration andproduction costs. Thus, the job-shop scheduling problems attract more and more attention.The job-shop scheduling problem is a simplified model of the actual production system, as alinear programming problem, more and more intelligent optimization algorithms are applied tosuch problems. For example: the genetic algorithm, simulated annealing algorithm, heuristicalgorithm, particle swarm optimization, etc. Among these intelligent optimization algorithm,genetic algorithm, its search for a wide range of parallel capability, while the dependence ofthe problem model is not high, so genetic algorithm in solving such problems, has a moreobvious advantage than the other algorithms.In this paper, using genetic algorithm to solve the job-shop scheduling problems, firstly,discusses the importance of the job shop scheduling, research methods, and proposed a specialclass with the processing path of flexible job-shop scheduling problem.Secondly establish the mathematical model respectively for the traditional job-shopscheduling problem and flexible job-shop scheduling problem. Process coding is used for thejob-shop problems. For the flexible job-shop problems, based on genetic algorithms, ascheduling approach is presented, which can be used to solve the problem. A new chromosomerepresentation with two-dimensional matrix is designed, which can enlarge traditional codingmode based on operators in expression of the machines. A new operation is designed to changethe extra information to expand the search range during the evolution process.Finally，the effectiveness of the algorithm is verified by computing results with a schedulingproblem, which meets the minimum processing time.