An Improved Genetic Algorithm-based Job Shop Scheduling System

Although China is a manufacturing country, but the manufacturing industry withworld-class manufacturing power is still lagging behind compared. Despite some ofthe human cost of factors aside, China’s manufacturing industry has not yet reachedthe level of resource management class, while process management is also a biggap. The addition to the impact on productivity and thus directly affect businessinterests, but also on the quality of the products also have some impact, forenterprises to further enhance the brand has a great influence. Shop scheduling themanufacturing process is an important part of belonging to a typical schedulingproblem, but also a combinatorial optimization problem. Shop scheduling problemsolving with a high degree of complexity, the different production areas may facedifferent shop scheduling problems, scheduling optimization for the actual productionline has practical value, strong practical significance.Firstly, the development and genetic algorithm shop scheduling problem solvingshop scheduling algorithm is applied to a simple description. Next to the shopscheduling problem, and the system shop scheduling problems are introduced, whilethe GA-related content and the presence of genetic algorithm for job shop schedulingproblems are introduced. Binding studies of the existing shop scheduling geneticalgorithm is proposed, and the lack of existing research on the genetic algorithm isapplied to improve shop scheduling problems. After using the improved algorithmdesign and implementation of shop scheduling system, and finally to the actual effectof the improved algorithm to analyze and summarize issues. Improved geneticalgorithm implemented shop scheduling system uses UML Unified ModelingLanguage in the design process to achieve the use of C++language, complete systembuild in VS2013platform. Specific algorithm improvement steps can be divided intothe following steps:1. Cross-probability control parameter settings, the mutation probability is set to improve the way. The introduction of adaptive strategy Srinvas et al., The entirecrossover probability and mutation probability with fitness nonlinear dynamic processof change. Resolve fixed crossover probability, mutation probability may bring partialoptimal solution or the problem of slow convergence.2. The individual coding improvements, using a Gray code encoding. Gray codeis a binary encoding improvements. To strengthen local capacity compared to thebinary Gray code encoding, without loss of the binary-coded operation easiness.3. Increase retention operator avoids crossover and mutation optimal solution inthe case of missing. By increasing retention operator of the population of individualsto retain a great advantage. That is a great advantage of the population, over thefitness of individuals to retain operator can not pass crossover and mutation, directaccess to the next generation.4. Cross-way improvements, combined with multi-cross-cross pattern andconsistent manner, given improved crossover operation. Cross improved methodavoids the multi-point crossover algorithm on the entire chromosome gene crossuneven, there is no consistent method of calculation of cross large problem. It is abetter performance crossover.