Quality improvement of multistage and multi-response grinding processes: an insight into two different methodologies for parameter optimisation

Process quality improvement using appropriate optimisation methodology has been a continual research endeavour. However, search for optimal path conditions for multi-stage and multi-response grinding in mass-scale manufacturing still remains a critical and difficult task for researchers. In this context, two different methodologies may be adopted to determine optimal process setting conditions. The first methodology (Methodology-1) is to assume each stage as independent, and thereby determine optimal setting conditions for the individual stages. Based on individual stage optimal process conditions, overall optimal path conditions are selected. Another possible methodology (Methodology-2) for optimisation is to consider all the stages as a single system, with their interdependency, and thereby determine the overall optimal path conditions. In this paper, an attempt has been made to compare and contrast the solution quality, as determined by genetic algorithm, and tabu search for both the methodology. The computational results show the relative superiority of tabu search.