Consider a single station assembly and test system, orders are released to the system with a rate of 3 units per hour (Poisson process). We assume that there is only a single station which can perform both the assembly and testing functions. The processing time of the station (when performing both functions together) follows exponential distribution with a rate of 8 units per hour. There is a probability of 0.5 that finished products are of bad quality. These poor quality products need to be disassembled by an operator before re-channeling back to the same station for assembly again. The time spent for disassembling also follows exponential distribution at a rate of 6 units per hour and there are more than enough operators who can perform disassembly tasks (we can assume that the number of operators is infinite). It is possible that an order will visit the station several times before finally passing the quality check. A simulation is developed for this system. When we run the simulation, we terminate it when 100 units of orders have been completed. The mean cycle time for these 100 orders is found to be 80 minutes and the sample variance is 12,000 minute2. Compute the confidence interval for the mean cycle time. Comment on the above computation. Are there any better ways to collect data and to compute the confidence interval for the mean cycle time?