A Constraint Programming Approach to the Multi-stage Batch Scheduling Problem

A hybrid Constraint Programming (CP) formulation able to address the multi-stage batch scheduling problem is presented. This approach handles different features found in industrial environments such as finite unit ready-times, dissimilar parallel equipment in each stage, sequence-dependent changeover times, topology constraints, forbidden job-equipment assignments, etc. Moreover, it renders good computational results with a variety of objective functions. However, the most important point about the proposed approach is its ability to handle constrained situations regarding both renewable and nonrenewable resources. When dealing with non-renewable resources, critical raw materials having more than one delivery date can be modeled. The consumption of raw materials can be described by two distinct patterns: (i) the whole amount is needed at beginning of the job demanding the raw material and (ii) the consumption occurs at a constant rate along the processing of the demanding task.