Problems for Arena Simulation Exercise

1. Develop a model of a simple serial two-process system. Items arrive at the system with a mean time between arrivals of 10 minutes, with the first arrival at time 0. They are immediately sent to Process 1, which has a single resource with a mean service time of 9 minutes. Upon completion, they are sent to Process 2, which is identical to (but independent of) Process 1. Items depart the system upon completion of Process 2. Performance measures of interest are the average numbers in queue at processes and the total time in system of items. Using replication length of 10,000 minutes and 5 replications, compare the results for the following input distributions: (a) Exponential interarrival times and exponential service times (b) Constant interarrival times and exponential service times (c) Exponential interarrival times and constant service times (d) Constant interarrival times and constant service times Consider first 500 minutes as warm-up period. Animate your model. Show the result graphically.

2. A proposed production system consists of five serial automatic workstations. The processing times at workstations are constant: 11, 10, 11, 11, and 12 (all times given in this problem are in minutes). The part interarrival times are UNIF(13,15). There is an unlimited buffer in front of all workstations, and we will assume that all transfer times are negligible or zero. The unique aspect of this system is that at workstations 2 through 5 there is a chance that the part will need to be reprocessed by the workstations that precedes it. For example, after completion at workstation 2, the part can be sent back to the queue in front of workstation 1. The probability of revisiting a workstation is independent in that the same part could be sent back many times with no change in the probability. At present, it is estimated that this probability, the same for all four workstations, will be between 5 % and 10 %. Develop the simulation model and make six runs of 10,000 minutes each for probabilities of 5, 6, 7, 8, 9, and 10 %. Consider first 500 minutes as warm-up period. Using the results construct a plot of the average cycle time (system time) against the probability of a revisit. Also include the maximum cycle time for each run on your plot. Run the model for 5 replications and compare the results. Animate your model.

3. A production system consists of four serial automatic workstations. Jobs arrive at the first workstation as exponential with mean 8. All transfer times are assumed to be zero and all processing times are constant. There are two types of failures: major and jams. The data for this system are given in the table below (all times are in minutes). Use exponential distributions for the uptimes and uniform distributions for repair times (for instance, repairing jams at workstation 3 is UNIF (2.8, 4.2)). Run your simulation for 10,000 minutes to determine the percent of time each resource spends in the failure state and the ending status of each workstation queue. Consider first 500 minutes as warm-up period. Run the model for 5 replications and show graphically the results for single replication and 5 replications. Animate your model.

Workstation Number 1 2 3 4

Major Failure Means Process Time 8.5 8.3 8.6 8.6 Uptimes 475 570 665 475 Repair 20,30 24,36 28,42 20,30

Jam Means Uptimes 47.5 57.0 66.5 47.5 Repair 2, 3 2.4, 3.6 2.8, 4.2 2, 3

4. An Office that dispenses automotive licenses plates has divided its customer into categories to level the office workload. Customers arrive and enter one of three lines based on their residence location. Model this arrival activity as three independent arrival streams using as exponential interarrival distributiuon with mean 10 minutes for each stream, and an arrival at time 0 for each stream. Each customer type is assigned a single, separate clerk to process the application forms and accept payment, with a separate queue for each. The service time is UNIF (8,10) minutes for all customer types. After completion of this step, all customers are sent to a single, second clerk who checks the forms and issues the plates (this clerk serves all three customer types, who merge into a single first-come, first-served queue for this clerk). The service time for this activity is UNIF (2.66, 3.33) minutes for all customer types. Develop a model of this system and run it for 5,000 minutes, observe the average and maximum time in system for all customer types combined. Consider first 500 minutes as warm-up period. Run the model for 5 replications and show graphically the results for single replication and 5 replications. A consultant has recommended that the office not differentiate between customers at the first stage and use a single line with three clerks who can process any customer type. Develop a model of this system, run it for 5,000 minutes, and compare the results with those from the first system. Animate your model.

5. Customers arrive at an order counter with exponential interarrivals with a mean of 10 minutes; the first customer arrives at time 0. A single clerk accepts and checks the orders and processes payments, taking UNIF(8,10) minutes. Upon completion of this activity, orders are randomly assigned to one of two available stock persons (each stock person has a 50% chance of getting any individual assignment) who retrieve the orders for the customers, taking UNIF(16,20) minutes. These stock persons only retrieve orders for customers who have been assigned specifically to them. Upon receiving their orders the customers depart the system. Develop a model of this system and run the simulation for 5,000 minutes, observing the average and maximum customer time in system. Consider first 500 minutes as warm-up period. Run the model for 5 replications and show graphically the results for single replication and 5 replications. A bright, young engineer has recommended that they eliminate the assignment of an order to a specific stock person and allow both stock persons to select their next activity from a single first-come, first-served order queue. Develop a model of this system for 5,000 minutes, and compare the results to the first system. Animate your model.

6. Items arrive from an inventory-picking system according to an exponential interarrival distribution with mean 1.1 (all times are in minutes), with the first arrival at time 0. Upon arrival, the items are packed by one of four identical packers, with a single queue feeding all four packers; the packing time is TRIA (2.75, 3.3, 4.0). Packed boxes are then separated by type (20%, international and 80% domestic), and sent to shipping. There is a single shipper for international packages and two shippers for domestic packages with a single queue feeding the two domestic shippers. The international shipping time is TRIA (2.3, 3.3, 4.8), and the domestic shipping time is TRIA (1.7, 2.0, 2.7). This packing system works three 8-hour shifts, five days a week. All the packers and shippers are given a 15-minute break two hours into their shift, a 30-minute lunch break four hours into their shift, and a second 15-minute break six hours into their shift; use the wait schedule rule. Run the simulation for four weeks (twenty working days) to determine the average and maximum number of items or boxes in each of the three queues. Consider first day as warm-up period. Run the model for 5 replications and show graphically the results for single replication and 5 replications. Run the simulation for two days to determine the average and maximum number of items or boxes in each of the three queues. Consider first day as warm-up period. Run the model for 5 replications and show graphically the results for single replication and 5 replications and compare this results with the previous simulation. Animate your model, including a change in the appearance of entities after they are packed into box.

7. Parts arrive at a single workstation system according to an exponential interarrival distribution with mean 21 seconds; the first arrival is at time 0. Upon arrival, the parts are initially processed. The processing-time distribution is TRIA (16, 19, 22) seconds. There are several easily identifiable visual characteristics that determine whether a part has a potential quality problem. These parts, about 10% (determined after the initial processing), are sent to a station where they undergo a thorough inspection. The remaining parts are considered good and are sent out of the system. The inspection-time distribution is 95 plus a WEIB (48.5,4.04) random variable, in seconds. About 14% of these parts fail the inspection and are sent to scrap. The parts that pass the inspection are classified as good and are sent out of the system (so these parts didnt need the thorough inspection, but you know what they say about hindsight). Run the simulation for 10,000 minutes to observe the number of good parts that exit the system, the number of scrapped parts, and the number of parts that received the through inspection. Consider first 100 minutes as warm-up period. Run the model for 5 replications and show graphically the results for single replication and 5 replications. Animate your model.

8. Parts arrive in the system exponentially with mean inter arrival time of 10 minuets. 40 % of the parts require processing at a milling center, 30% of them require processing at a drilling center and the remaining parts are to be processed on a lathe. After these machining process parts are sent to an inspection station followed by a dispatch center. There are two inspectors of the same kind. The milling center and drilling center undergoes an exponential failure with mean uptime of 300 minutes and down time of 45 minutes. The time taken at different centers is given below:

Station Milling Drilling Lathe Inspection Dispatch

Time taken in minutes TRIA(12,15,18) UNIF(15,18) TRIA(10,12,16) TRIA(8,10,12) UNIF(5,8)

Develop a simulation model and run for 30 days, each day having working hours of 16. Run for 5 replications. Find the average number in queue at different stations, WIP, utilization. Show the variation in queue length with time using graph.

9. Two different part types arrive at a facility for processing. Parts of type 1 arrive with inter-arrival times following a lognormal distribution with a log mean of 11.5 hours and log standard deviation of 2.0 hours (note that these values are the mean and the standard deviation of this lognormal variable itself); the first arrival is at time zero. These arriving parts wait in the queue designated for the only part types 1s until a (human) operator is available to process them (there is only one such human operator in the facility) and the processing times follow a triangular distribution with parameters 5, 6, and 8 hours. Parts of type two arrive with inter-arrival times following an exponential distribution with the mean of 15 hours; the first arrival is at time 0. These parts wait in a second queue (designated for part 2s only) until the same lonely (human) operator is available to process them; processing time follow a triangular distribution with parameters 3, 7 and 8 hours. After being processed by human operator, all parts are sent for processing of an automatic machine not requiring a human operator, which has processing times distributed as triangular with parameters 4, 6 and 8 hours for both the part types; all parts shares the same firstcome, first-served queue for this automatic machine. Complete parts exit the system. Assume that the times for all the part transfers are negligible. Run the simulation for 5000 hours to determine the average total time in the system (some times called cycle time) for all the parts (lumped together regardless of type), and the average number of items in the queues designated for arriving parts. Animate the model, including use of different pictures for the different part types, and use resources that looked different for busy Vs idle.