Monday, June 17, 2019

FMS Optimisation using Discrete Event Simulation and Genetic Algorithm Essay

FMS Optimisation using Discrete Event Simulation and Genetic Algorithm - experiment ExampleThe production planning in the manufacturing systems argon forecasted using Enterprise Resource Planning package recently. Since the market demand varies every now and so the process has to be driven based on that. Flexible manufacturing system functions by utilising these advancements and deliver multiple products of sufficient quantity as per the demand.Genetic Algorithms are found to provide solutions for real-time problems in various operations. It has been used conveniently for researchers for various search and optimization problems. Owing to the problems associated with FMS optimization using Genetic algorithmic rule and discrete simulation system this present project is initiated. Kazuhiro Saitou et al. (2002) presented a robust design of FMS using colored Petri nets and genetic algorithm. In their work it was found that the imaging allocation and operation schedule were modelled as colored Petri nets. Their robust model designed minimized the production cost under multiple operation plan. It as able to handle large data sets conveniently as well as operates flexibly by using an genetic algorithm merged with shortest close operation time dispatching rule and automatically finds the optimal resource. These kinds of simulation can be more applicable in situation where there is varied affair specification. Discrete event simulationThe discrete event simulation works powerfully in optimization and decision-making process in manufacturing systems. Merchawl and Elmaraghy (1998) developed an analytical memory access to customize the discrete event simulation for decision-making in flexible manufacturing systems. Planning horizon, the overall system middling interarrival time and the average number of workstation influences the simulation bombardment time. In their approach they reduced the simulation run time by aggregating the number of workstations. They also validated their methods with sample and control measures by running the applications with and without aggregation of the workstation. The results showed a 400% time reduction with fewer errors.Mostly the Genetic Algorithms (GA) is coupled with other techniques or processes to handle complex situation. Studies carried out revealed that increasing the alteration rates above optimum level cannot solve the problems associated. The study was focused on finding methods to improve the performance of GA by improving the average fitness of the initial population, P. Fenton and P. Walsh (2005)Review and AnalysisThe project impart focus on initial aspects of reviewing the present complications and problems associated with the utilities of GA and Discrete event simulation methods. The archetypical phase of review will focus on identifying the present application of these algorithms in various domains and its recent advancements. The next phase of the review analysis will be focused towards identifying all the limitations of these systems at the implementations

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.