Development of an Efficient Prediction Model for Optimal Design of Serial Production Lines

One of the problems encountered in design and implementation a serial production line (SPL) is buffer size between machine tools. The SPL has an important impact on productivity whole system. tools’ characteristics including their uptimes downtimes process parameters are main factors that affect decision regarding size, thus SPL. Due to dynamic nature this problem, it complex find optimal Thus, paper, Efficient Prediction Model (EPM) developed using Artificial Neural Network (ANN). purpose EPM each succeeding pair tools at any given An optimization model based genetic algorithms (GA) used generate learning data for prediction or near bay tool proposed approach integrates methodologies evaluate, predict sizes maximum productivity. Including uptime downtime enable method be improve running as well new Numerical examples five fifteen were conducted independently research results show ability determine reasonable amount time. In particular, case studies accurately especially machines even higher number yet with acceptable but less accuracy. Finally, performance was compared some state art methods reported literature. comparison shows superiority present identify throughput under same downtimes.