Workshop scheduling using practical (inaccurate) data. Part 3: A framework to integrate job releasing, routing and scheduling functions to create a robust predictive schedule

A comprehensive simulation study conducted by the authors investigated the robustness of a predictive scheduling system in a dynamic and stochastic environment. The results revealed that to improve the robustness of a scheduling system, besides using a robust scheduling method with a frequent rescheduling policy, the shop load should be well controlled and kept balanced. Integrating the planning and the scheduling functions has been shown to achieve this objective. This paper discusses the effects of the planning – i.e. job releasing and routing – and the scheduling functions in creating a robust schedule and a framework to integrate the above functions is proposed. This system consists of a planning module that is concerned with job releasing and routing decisions and a scheduling module that provides the detailed scheduling. A mathematical model using the integer programming technique is used to demonstrate a solution for the planning module. In addition, a heuristic algorithm is used to solve the scheduling problem. It is shown that, in terms of shop load balance level and job delivery time, the proposed system performs better than a benchmark loading strategy on the basis of minimum processing costs.