A New Algorithm for Cyclic Scheduling and Design of Multipurpose Batch Plants

Abstract

Many researchers have studied the scheduling, planning, and design of multipurpose batch processes. However, not so many studies have treated design and scheduling or design and planning simultaneously. The complexity of these systems stems from the fact that plant configuration must be determined for the purpose of process scheduling, yet scheduling must be done to devise the plant configuration. In the present study, a new algorithm for determining the best multipurpose scheduling and plant configuration is suggested. Since the objective function of the problem is nonlinear, it is linearized using a separable programming method. The proposed method consists of a number of procedures. First, a feasible configuration is obtained. Next, both the optimum equipment size and cyclic scheduling are determined for the plant configuration obtained in the first procedure. Last, the evolutionary design method proposed by Fuchino et al.(J. Chem. Eng. Jpn. 1994, 27, 57−64) is used to find the solution that minimizes the total cost. The efficacy of the proposed approach is demonstrated in three examples. Usually, equipment sizes are considered by a continuous variable in mixed integer linear programming and task processing time is assumed to be constant. However, most types of equipment are manufactured only in discrete volume classes. In addition, processing times are dependent on batch sizes. Hence, to apply the optimization methods developed here to real industries, the method was modified such that the volumes of equipment are considered as discrete variables and the processing time is a function of batch size.

Multimodal processes scheduling in mesh-like network environment

“…The existing approach to solving the SCCPs scheduling problem is based upon the simulation models, e.g. the Petri nets (Song and Lee 1998), the algebraic models (Kampmeyer 2006) upon the (max,+) algebra or the artificial intelligent methods (Heo et al 2003). The SCCP driven models, assuming a unique process execution along each cyclic route while allowing to take into account the stream-like flow of local cyclic processes, e.g.…”

“…It is evident that the composition of schedules (i) X ′ of all the substructures of the structure SC does not lead to interferences in the execution of the operationthe schedules (i) X ′ on the resources (i) R 1 , (i) R 3 and (i) R 9 . On the basis of the obtained schedules it is also possible to determine (according to (15) the dispatching rules for all the resources of the structure SC; the rules are presented in Tab. 1.…”

Declarative approach to cyclic steady state space refinement: periodic process scheduling

“…Among them, the most frequently applied are the mathematical programming approaches (usually integer programming or mixed integer programming [24]), max-plus algebra [16], constraint logic programming [2][3][4][5]22], swarm and evolutionary algorithms [6,10,20], and Petri nets [19]. The majority of these are oriented at finding a minimum cycle or maximum throughput while assuming a deadlock-free flow of processes.…”

Multicriteria Optimization of Multiproduct Batch Chemical Process Using Genetic Algorithm

“…It should be emphasized that the batch plant's design has long been identified as a key problem in chemical engineering as reported in the literature (Montagna et al. 2000; Cao and Yuan 2002; Chunfeng and Xin 2002; Heo et al. 2003; Cavin et al.…”

“…The number, the required volume and size of parallel equipment units in each stage are then determined to minimize the investment. It should be emphasized that the batch plant's design has long been identified as a key problem in chemical engineering as reported in the literature (Montagna et al 2000;Cao and Yuan 2002;Chunfeng and Xin 2002;Heo et al 2003;Cavin et al 2004;Chakraborty et al 2004;Goel et al 2004;Pinto et al 2005). Formulation of a batch plant design generally involves mathematical programming methods, such as linear programming (LP), nonlinear programming (NLP), mixed-integer LP or mixed-integer NLP (MINLP).…”