We propose a general method for the solution of chemical production scheduling problems in network environments. The method consists of three stages. In the first stage, a discrete-time mixed-integer programming (MIP) model is solved to quickly obtain an approximate solution. In the second stage, the solution is mapped onto newly introduced unit- and material-specific continuous-time grids, using a mapping algorithm. In the third stage, a continuous-time linear programming (LP) model is solved to improve the accuracy of the mapped discrete-time solution by refining the timing of events and batch sizes. The proposed method takes advantage of the complementary strengths of discrete- and continuous-time formulations, which enables us to not only handle various processing features (e.g., intermediate deliveries and orders, time-varying resource availability and cost, variable processing times), but also obtain order of magnitude speedups in the solution of large-scale instances.

我们提出了一种解决网络环境中化学生产调度问题的通用方法。该方法包括三个阶段。在第一阶段，求解离散时间混合整数规划（MIP）模型以快速获得近似解。在第二阶段，使用映射算法将解决方案映射到新引入的单位特定和材料特定的连续时间网格上。在第三阶段中，解决了连续时间线性规划（LP）模型，通过改进事件的时间和批量大小来提高映射的离散时间解决方案的准确性。所提出的方法利用了离散时间和连续时间配方的互补优势，这使我们不仅能够处理各种处理功能（例如，中间交货和定单，