Inherently safe and economically optimal design using multi-objective optimization: The case of a refrigeration cycle

Abstract

The economic viability of industrial processes strongly depends on their safe and reliable operation. The method of inherent safe process design enables systematic consideration of safety measures in order to ensure process safe operation at the early stages of process design. The challenge is that the economic measures that are often considered for the design of industrial processes are often incommensurable with the safety measures. In the present research, a novel framework is proposed in which the safety criteria are quantified based on consequence modeling and aggregated with the economic performance using multi-objective optimization programming. The developed methodology was applied to the design of a simple refrigeration cycle. The optimization algorithm was NSGA-II. The results suggested a strong trade-off between the competing economic and safety objectives in terms of Pareto frontiers that clearly quantified the required compromise. It was observed that only with a minor increase in the capital investment, it is possible to significantly improve the safety. While the case of the refrigeration cycle was selected as a demonstrating case, the research methodology is to large extend general and deemed to be acceptable to design and operation of other industrial processes.