READ THESE REMINDERS THOROUGLY BEFORE STARTING
• Make sure to create individual folders for each problem that you will be working on and put all 3 folders in a main folder and add your name to the main folder
• Make sure to use your initials in Arena modules and add your problem Project#, title and your name to both Arena flow panel and Run Setup Parameters.
• Make sure to upload ALL Arena files (including model, run files and pdf versions of your results)
• You HAVE TO type/add your results and interpretations in THIS word file and make sure to type your name on to both the 1st page and to word file name.
• Proper use of American English language is necessary to receive full grade. Feel free to continue writing on a new page in this document as needed.
Problem 1 – Output Analysis of CSS Model (30pts):
Focus on given simple call center system (CSS) model. The default model includes 26 trunk lines and you are asked to study the impact of different trunk line capacity scenarios on the “# of rejected calls” (rejected calls are counted with a record module) over 100 replications with the given run setup in the model. Your manager wants you to test 3 resource scenarios and see what works best to have the least average rejected calls. Consider the following scenarios:
– Scenario 1: Working with the 20 Trunk Lines
– Scenario 2: Working with the 26 Trunk Lines (Default case)
– Scenario 3: Working with the 32 Trunk Lines
Conduct the following analyses by using output analyzer.
– Provide histograms of rejected calls for each scenario
– Confidence intervals for rejected calls (classical C.I. on the mean)
– Look at confidence intervals and conduct a One Way ANOVA test (if required).
– Conclude your analysis with suggestion of a scenario that will provide the least average rejected calls with statistical proof and interpretations.
– Do not try to model the problem; ARENA model is already given on BB
– You need to use the ARENA model to create the statistics for the scenarios and conduct the required analyses with OUTPUT ANALYZER.
– This problem is also explained in the textbook.
Problem 2: Acute-care Facility (40 points)
An acute-care facility treats non-emergency patients (cuts, colds, etc.) Patients arrive according to an exponential inter-arrival time distribution with a mean of 11 (all times are in minutes). Upon arrival, they check in at a registration desk staffed by a single nurse. Registration times follow a triangular distribution with parameters 6, 10 and 19. After completing the registration, they wait for an available examination room. There are three identical rooms and three doctors. Each exam requires a doctor and a room. Data show that patients can be divided into two groups with regard to different examination times. The first group (%65 of patients) has exam times that follow triangular distribution with parameters 13, 22 and 39 mins. The second group has triangular exam times with parameters 24, 36 and 59 mins. Upon completion of examination, patients are sent home. The facility is open 16 hours each day. Make 200 replications for one day, use base time units as minutes and use 2 hours of warm up time and minutes as base time unit. Collect and interpret results about the following system performance metrics: Average & half width of 1) Number of patients in system and queues, 2) Waiting times, 3) Resource utilization levels. Show your results as tabular format in the world file.
Average Half Width
Patients in System (WIP)
Patients in Queue
Group 1 Queue
Group 2 Queue
Group 1 Queue
Group 2 Queue
• What is the bottleneck resource in the system? Why ?
• What are the areas that need improvement from system performance viewpoint? Provide a detailed discussion and feedback.
Problem 3 – Ceiling Fan Production System (30pts):
Kits of ceiling fans arrive at an assembly system with TRIA (2, 5, 10) interarrival times (all times are in minutes). There are three assembly operators and the kits are automatically sent to the first available operator for assembly. The fan assembly time is operator dependent as given below:
Operator Assembly Time
1 TRIA (15,18,21)
2 TRIA (16,19,22)
3 TRIA (17,20,23)
Upon completing of the assembly process, the fans are inspected with approximately7% being found defective. A defective fan is sent back for repair process, which will be performed at the same assembly process. The defective fans have a higher priority than the new arriving fans for assembly. Build and run your model for 24,000 mins with 2000 mins warm-up time and 10 replications. The system working time is 8 hours a day. Report the followings with sufficient interpretations.
– # WIP
– Total time per entity
– Average waiting time and number waiting
– Average resource utilizations
– Proportion of defectives to total output
Note: If you can’t model the operator dependent assembly process or defective units re-process at the same assembly fully, make sure to submit a working model and results to receive partial credits.