A novel unit-specific event-based formulation for short-term scheduling of multitasking processes in scientific service facilities

Abstract

Scientific service facilities examine a number of samples from different customers for several physical and chemical properties using processing units with large capacities. A processing unit can process a great number of samples simultaneously. The process in such scientific service facility can be treated as a multi-tasking multipurpose batch process. Despite the great interest in developing models for scheduling of process industry during the past three decades, scheduling of multi-tasking multipurpose batch processes in a scientific service facility has not been considered adequately. In this work, we develop three novel mathematical models using the well-established unit-specific event-based modelling approach. The computational results demonstrate that the proposed mathematical models are able to reduce the number of event points required, which leads to a significant reduction in the model size and computational time. One of the proposed models in which the timing variables are defined based on processing units is the most efficient in most cases especially when minimization of makespan is used as the objective, where at least one order of magnitude less computational time than all other models is required to generate the optimum solution compared to other existing models.