Modified particle swarm algorithm for scheduling agricultural products

Industries tend to manufacture higher quality products with a low cost due to the competitive environmental loop and dynamic customers’ demand. The integration of intelligent manufacturing planning to agricultural operations and products will permit a spike in efficiency of yield, especially in the regions of agriculturally based economies. Though, to gain a vivid exploitation, the old-fashioned agriculture products manufacturing planning and scheduling techniques needs to be revised, especially the methodology in job shop planning must be augmented with efficient operational sequential scheduling. Cellular Manufacturing Systems (CMS) inclined to possess a higher complexity than the traditional manufacturing systems. Three major problems are coping in CMS domain: cell formation, product family selection, and product scheduling. This work deals with the problem of product scheduling in the CMS environment. A mixed integer linear programming mathematical model is introduced for the conflicting performance measures i.e. minimization of work in process (WIP) and maximization of average machine cell utilization. Since the current problem is considered as NP-hard problem, so a modified particle swarm optimization (MPSO) algorithm is proposed to find the optimum scheduling under the given constraint model of conflicting objectives. In the proposed MPSO, basic PSO is integrated with NEH heuristic to achieve better optimal sequence in less computation time. The obtained results are compared with other hybrid PSO algorithms with seed solutions from Gupta and Palmer heuristics. Furthermore, results are also compared against meta-heuristics such as genetic algorithm (GA), standard PSO, and artificial bee colony (ABC) algorithm. It shows that the proposed MPSO algorithm performs better than the existing compared algorithms. A real time case scenario of agriculture-based manufacturing industry is solved to validate the proposed planning algorithm.