Kanban optimization by simulation and evolution
研究了多阶段看板控制系统的优化设计问题，将批量大小作为决策变量，采用仿真结合遗传算法求解，并通过示例验证了方法的可行性。
The problem of the optimal design of multistage systems with Kanban control mechanism is investigated. The optimization problem generalizes those from literature by considering a general criterion function and including the lot sizes as decision variables. Since no analytical solutions can be expected simulation combined with a genetic algorithm is used. The simulator KaSimIR as well as the optimization tool LEO are briefly described. Some examples demonstrate the usability of the approach.