Long-term scheduling technique for wastewater minimisation in multipurpose batch processes

Abstract

Most of the methodologies published in literature on wastewater minimisation for batch processes are based on short-term scheduling techniques. When these methods are applied to longer time horizons, the computational time becomes intractable, hence the focus of this thesis. This thesis presents a methodology for simultaneous optimization of production schedule and wastewater minimisation in a multipurpose batch facility. The key feature of the presented methodology is the adaption of cyclic scheduling concepts to wastewater minimisation. The methodology is developed based on continuous-time formulation and the state sequence network (SSN) representation. The methodology is successfully applied to two common literature examples and an industrial case study to demonstrate its effectiveness. None of the currently published wastewater minimisation techniques could solve the case study for a time horizon of 168h. However, through the application of the presented methodology, a time horizon of 168h for the case study was reduced to 8 cycles with the cycle length of 23h, for which the CPU time for the optimum cycle is 64.53s.