Implementation of Soft Computing-Based Metaheuristic Algorithms in Multiobjective Environmentally-Conscious Machining Operation Sequence Optimization With Carbon Emission Reduction

Abstract The process planning decisions on detailed steps of machining a part including selection of operations, machines, tools and operation sequences are crucial, directly impacting the carbon emissions associated with machining. Operation sequencing generates the optimal order of operations by reducing production time and cost. A need is felt to explore operation sequence optimization by balancing economic and environmental considerations that very few studies have so far addressed. Our study proposes an approach for multiobjective operation sequence optimization by simultaneously minimizing carbon emission and total production cost in machining. Firstly, machining operations are selected and carbon emissions are quantified for individual operations. Secondly, the proposed multiobjective optimization model for operation sequencing is applied. Due to NP-hard nature of this problem due to the exponential growth of potential solution sequences as the number of operations increases, making it computationally challenging to find an optimal sequence, it is often challenging and very time-consuming to solve using exact mathematical models. However, metaheuristic approaches can potentially offer near-optimal solutions in a more reasonable time frame. We implemented Non-dominant Sorting Genetic Algorithm-II(NSGA-II), Multiobjective Particle Swarm Optimization (MOPSO) and Multiobjective Ant colony Optimization (MOACO). Finally, a case study is used to validate and compare their effectiveness in providing low-carbon emissions with cost-effective sequences. NSGA-II is found to have higher computational time but a superior Pareto tradeoff between reducing carbon emissions and production costs.