Question: Go to a laundromat, and "relive" the authors; data-collection experience discussed in Example. (Make sure that the management gives permission to perform this study.)
Example: The Laundromat
As budding simulation students, the first two authors bad assignments to simulate the operation of an ongoing system. One of these systems, which seemed to be a rather simple operation, was a self-service laundromat with I 0 washing machines and six dryers. However, the data-collection aspect of the problem rapidly became rather enormous. The interarrivaltime distribution was not homogeneous; it changed by time of day and by day of week. The laundromat was open 7 days a week for 16 hours per day, or 1 12 hours per week. It would bave been impossible to cover the operation of the lanndromat with the limited resources available (two students who were also taking four other courses) and with a tight time constraint (the simulation was to be completed in a 4-week peri6d). Additionally, the distribution of time between arrivals during one week might not have been followed during the next week. As a compromise, a sample of times was selected, and the interarrival-time distributions were classified according to arrival rate (perhaps inappropriately) as "medium," and·tow." Service-time distributions also presented a difficult problem from many perspectives. The proportion of customers demanding the vanous service combinations had to,be.observed and recorded. The simplest case was the customer desiring one washer followed by one dryer. However, a customer might choose two washing machines followed by one dryer, one dryer only, and so on. The customers used numbered machines, and it was possible to follow the customers via-that reference, rather than remembering them by personal characteristics. Because of the dependence between washer demand and dryer demand for an individual customer, it would have been inappropriate to treat the service times for washers and dryers separately as independent variables.