Research On Single Piece And Small Batch Mixed Shop Scheduling Algorithm Based On MES

In large discrete manufacturing industry, single piece and small batchproduction mode occupies a large proportion, such as heavy mechanicalequipment manufacturing, ship manufacturing, power station complete sets ofequipment manufacturing, aircraft manufacturing and so on. In the enterprisewith the production mode, many varieties of products which batch is small orsingle piece are produced during given periods of time according to orders.There exist some characteristics such as product variety, unstable productionconditions, various processes, large machine constraints, long product movingroute in process, low production continuity in production process. So solving theproblem of single piece and small batch production scheduling has become animportant research direction of the scheduling field.At present, the manufacturing execution system (Manufacturing ExecutionSystem, MES) applied in large discrete manufacturing industry is more and morepopular, and provides real-time information environment for job shop schedulingby the real-time production data collection and analysis.Based on the analysis of the characteristics of small batch MES shopscheduling, the mathematical model is established for single and small batchmixed shop scheduling problem by considering the scheduling cycle and jobprocessing preparation time, and a scheduling genetic algorithm is proposed toeffectively shorten the scheduling period, improve job processing continuity andreduce the frequency of tooling replacement. For solving the problem that theexisting ant colony algorithm using disjunctive graph to describe the relationshipbetween job processing increases the complexity of the algorithm, a generalizedant colony algorithm is also proposed to effectively solve the single piece andsmall batch mixed shop scheduling problem. The algorithm reduces thecomplexity of the algorithm and enhances the robustness of the algorithm. The experiments show that the algorithms proposed outperform other existingalgorithms.