A Method for Performance Evaluation of Automated Flow Lines

In order to find the optimal configuration of an automated flow line it is necessary to compare the performance of several alternative configurations. In this situation the use of simulation is normally not viable given the high modelling and computational effort it would require; on the other hand, the application of classical analytical tools would often entail the adoption of unrealistic assumptions. To solve this problem, the paper presents an approximate analytical method that overcomes many of the limitations of analytical tools while preserving good accuracy and low computational effort. In particular the method can deal with deterministic processing times, finite buffer capacity and multiple failure modes of the machines. This last feature represents an important improvement over existing techniques and is very useful when dealing with automated assembly lines where each station is composed by different devices that have different reliability and different ways to get repaired. The validation of the method has been carried out by means of extensive testing and comparison with simulation. An application to a real case is also reported.