Modify Model 7-1 to account for acquiring a new customer, in addition to the one supplying the existing three part types. This new customer will supply two new types of parts—call them Type 4 and Type 5. The arrival process is in addition to and independent of that for the original three part types and has exponential interarrival times with mean 15 minutes; the f rst arrival is at time 0. When the parts arrive, assign 40% of the new parts to be Type 4 and the rest to be Type 5. Here are the process plans and mean processing times (in minutes) for the new part types: Part Type Cell/ Mean Proc. Time Cell/ Mean Proc. Time Cell/ Mean Proc. Time Cell/ Mean Proc. Time 4 1 6.1 3 5.2 2 1.3 4 2.4 5 2 3.5 3 4.1 4 3.1 1 2.0 While people feel comfortable with these mean processing times, there’s not very good information on their distributions, so you’re asked just to assume that the distributions are uniform with the indicated mean but plus or minus 0.2 minute in each case. For example, if a mean of 6.1 is shown, assume that the distribution is uniform between 5.9 and 6.3. Make all necessary changes to the model, including the modules, animation kel01315_ch07_311-344.indd 342 17/12/13 10:10 AM Intermediate Modeling and Steady-State Statistical Analysis 343 pictures (create new entity pictures for the new part types), and anything else required. Be sure that the animation works properly, including the clockwise-only movement of all entities and the new part types.

(a) Clearly, adding this new customer is going to clog the system compared to what it was. Using the time-average total number of parts in all four queues combined, how bad does it get compared to the original system? Just make one run of each alternative (you’ll do a better job on statistical analysis in the next part of this exercise). Include a text box in your model with your time-average total number of parts in all four queues combined, both for this model and for the original Model 7-1.

(b) In an effort to alleviate the added congestion introduced by the new customer, you’ve been given a budget to buy one new machine to be placed either in Cell 1, 2, or 4 (not in Cell 3). Where would you recommend placing it in terms of the time-average total number of parts in all four queues combined? Assume that the new machine will work at the same rate as the machine it joins in whatever cell. You might want to refer to Section 6.5 and specify your own output statistic to use PAN. View this as a terminating simulation and make 100 replications with the goal to select the best placement of the new machine; include for comparison a base-case scenario in which no machines are added anywhere.