An improved approach to scheduling multipurpose batch processes with conditional sequencing

Scheduling of multipurpose batch processes has gained much attention in the past decades. Numerous models using different mathematical modelling approaches have been proposed. The model size and computational performance largely depend on the number of time points, slots or event points required. Most existing models still require a great number of time points, slots or event points mainly because a consumption task always takes place after its related production tasks regardless of whether it consumes materials from the related production tasks. Although two existing models have been developed to overcome this limitation, they can either lead to real time violation or generate suboptimal solutions in some cases. In this work, we develop an improved unit-specific event-based model in which a consumption task takes place after its related production task only if it consumes materials from the production task. A consumption task starts immediately after its related production task completes only if there is no enough storage for materials produced from the producing task. We also allow production and consumption tasks related to the same state to take place at the same event points. The results show that the proposed model generates same or better solutions than existing models with less number of event points and less computational time.