Effect of confidence interval on bottleneck identification via simulation

This paper presents the application of simulation-based procedure to identify bottleneck in a job-shop environment. Ten jobs ten machines job-shop problem (10×10 JSP) is used to test the simulation-based procedure for identifying the system bottleneck. The result from the case shows that the choice of confidence interval level (CL) used in throughput mean comparison has effect on the bottleneck selection. Two scenarios are presented with two different CL of throughput mean, 75% CL and 90% CL. The result from the experiment shows that when % CL is increased, the judgment of bottleneck identification can be changed. Thus, it is necessary to use appropriate number of replications to match with the required %CL used in the simulation. In general, more replications are necessary when higher percentage of confidence interval level is required.