SIMULATION BASED PERFORMANCE EVALUATION OF AUTOMATED MANUFACTURING- LINE

Abstract

The increasing demand for productivity, varying product life-cycle demands, and optimal allocation of resources, has motivated continuous research in the last years in modeling and performance analysis of manufacturing systems. Many real-world systems consist of different workstations having random or deterministic processing, failure and repair times. For such systems discrete event simulation seems to be the best analysis tool. To respond to varying product life-cycle demands and internal disturbances, an automated manufacturing line designers must design manufacturing lines having capacity flexibility to meet changing market demands without earliness/tardiness and error recovery flexibility to deal with internal disturbances. Because conventional manufacturing lines don't support these features, re-configurable manufacturing lines are needed that can handle varying product life-cycle demands and internal disturbances. Since, an automated assembly line is a type of an automated manufacturing line. In this work these two terms will be used interchangeably. In the present work, model for the re-configurable assembly line has been proposed. The taken model involves a set of movable robots, which move on a guided path. Each robot can serve more than one workstation. If a robot break down, it can be pulled out of the line and the remaining robots will perform work with out much delay. The failed robot can be reintroduced into assembly line after its repair. At the same time if there is a change in product demand, changing the number of robots in the system can vary production rate of. the assembly line. The proposed model has plug and play capability. Thus the model can respond to w both external as well as internal disturbances. Simulation software using discrete event simulation technique has been developed to analyze the performance of the assembly line. Several assumptions are made such as part arrival follows exponential distribution; robot failure follows constant distribution and constant robot repair time. Performance is evaluated in terms of resource utilization, flexibility, WIP, buffer allocation problem (BAP).. .etc. iii