Performance modelling and optimisation of the rice processing industrial system using PSO

Abstract

The present work includes the performance modelling and optimisation of the rice processing industry for achieving the maximum availability. The study presents a methodology for availability evaluation of the process industry. For the increased production rate and resource utilisation, the maintenance plan should be developed which can predict the effect of performance parameters with time. The performance modelling for the steady state availability (SSA) has been done using Markov method using historical failure and repair database followed by the system optimisation using particle swarm optimisation (PSO). The performance parameters are assumed to be negative exponential which are independent to each other and repaired units are considered as fine as original. The uncertainties of failure and repair rates (FRR) are removed by selecting these parameters randomly in PSO. The maximum availability of each system for all possible combinations of FRR within fixed minimum and maximum limits for each sub-system has also been computed. The rice processing plant consists of several sub-systems for which optimum availability level for rice processing plant for random combinations of FRR has been found. The computed results are economically beneficial for the plant personnel in improving the production rate and maintenance planning.