Heuristic synthesis in the design of noncontinuous plants

The design procedures for noncontinuous plants in published literature are limited to batch plants of simple configurations. The goal of this research is to present new methodology for the design of a batch plant in its most general form. The research is based on the premise that the synthesis decisions can be more readily made by incorporating process knowledge in the design process through the use of heuristics. Hence, the focus of this study is on developing design heuristics for various synthesis options. The main design issues are: (i) determination of a suitable design approach; (ii) identification of various design options; (iii) development of heuristics for design and production planning; (iv) combining these heuristics in one procedure.;New heuristics have been developed based on physical understanding of noncontinuous processes. For single product plants, the development of heuristics closely follows mathematical derivation but for multiproduct plants such derivation is very complex. Therefore, the implementation for multiproduct plants is accomplished in two steps. The first step uses a novel technique which represents a multiproduct plant as a single hypothetical product plant. The second step applies the single product heuristics to the hypothetical product plant. A new procedure has been developed to test the performance of these heuristics in an integrated form.;Another design procedure based on simulated annealing (SA) algorithm has been developed because SA can solve problems which cannot be solved by existing methods. It has been used as a tool to develop benchmark problems and test the efficacy of the heuristics. A set of generic annealing parameters has been selected to achieve the best results. Several forms of SA algorithms have been evaluated.;The performance of both approaches have been compared with the existing procedures and with each other. The results for the heuristic approach are within 2% of the global optimum. Inclusion of new heuristics enables the design procedure to solve larger, more general and realistic problems with reasonable computational effort.;The thesis addresses several synthesis options including insertion of intermediate storage and nonidentical parallel units, the use of mixed mode of operation, task merging and splitting and campaign formation.