Parallel ATS Performance Analysis Based on Generalized Stochastic Petri Nets

In order to increase the test efficiency of ATS (Auto Test System), and improve the system performance, we model the parallel test process in GSPN (Generalized Stochastic Petri Nets)-based system performance analysis method, and prove that the isomorphism of the parallel test process is the Markov process. After analyzing the model process performance, we get some results on utilization ratio of all kinds of test resources in the parallel ATS, and average delay data of task process. By means of such results, the bottlenecks which can easily affect the parallel ATS performance show up. With analysis on the test resources assignment in parallel ATS, we regulate the match-relationship between the test resources and test tasks; therefore, the efficiency of ATS can be improved deeply.