Pioneering Heat Exchanger Network Synthesis: A TOPSIS Driven Paradigm for Optimal Solutions

Abstract

Multi-Criteria Decision Making (MCDM) is a significant challenge across various domains, requiring adept resolution of conflicts arising from diverse objectives and criteria. This study proposes an innovative approach aimed at optimizing controllability, minimizing irreversibility, and maximizing overall effectiveness in control system design to address this challenge. The primary objectives of this study are to introduce a novel methodology for selecting Heat Exchanger Networks (HEN) using the well-established Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method. Additionally, a closeness coefficient is introduced to rank alternatives networks based on their proximity to the ideal solution. Two illustrative case studies are presented to showcase the methodology's effectiveness, adaptability, and robustness in discrete multi-criteria decision-making problems, particularly in the context of HEN selection. Consistently identifying HEN configurations that fulfill controllability objectives, the methodology demonstrates its effectiveness and potential for broader applications beyond HEN optimization. The case study results affirm the adaptability and robustness of the proposed approach. In summary, this paper introduces an original and versatile approach to address the complexities of multi-criteria decision-making, specifically in the context of HEN selection. Rooted in the TOPSIS method and fortified by the closeness coefficient, the methodology holds promise for intricate decision-making processes and offers transformative possibilities for control system design. The study concludes by inviting further exploration of the proposed methodology, emphasizing its significant contribution to the field and its potential for widespread impact. Researchers and practitioners are encouraged to investigate and apply this innovative approach in diverse decision-making scenarios. The ranking results reveal that alternatives M and K is the optimum one among all the alternatives for both cases with a closeness coefficient equal to 0.651 and 0.971.