Questions to: Dr. Jorge Hernandez –
[email protected] ASSIGNMENT The University of Liverpool Management School 2019 – 2020 (Semester 1) EBUS504 (Postgraduates) Operations Modelling and Simulation DEADLINE: January 10th, 2019 Lateness Penalty: Five percentage points shall be deducted from the assessment mark for each working day after the due date up to a maximum of five working days; however, the mark will not be reduced below the pass mark for the assessment (40%). Work assessed at below 40% will not be penalised for late submission of up to five working days. Work received more than five working days after the submission deadline will receive a mark of zero. Cheating: You are encouraged to discuss your general understanding of the exercise with colleagues of other groups, but, you must write up your project report based on the work of your group only. University regulations about cheating – especially COLLUSION and PLAGIARISM (copy from sources without acknowledgement) – apply. Hand-in procedure: Hand your work to the Student Support Office on the ground floor, Chatham Building; these assignments will be date stamped each night. If your work is late for medical or other good cause, attach a copy of your certificate and/or explanation. Notes: You must submit: ● One electronic copy through the TurnItIn on VITAL ● an electronic copy of the witness models developed through TrunItIn on Vital (all in one zip or rar file with your name and ID as filename e.g. EBUS504_SMITH_20091234.zip) Regarding the witness models: It is mandatory that every witness model provided must work and match the provided results. Questions to: Dr. Jorge Hernandez –
[email protected] 1. Practical questions: System Dynamics (30 Marks) Considering the following description for the filling a glass of water process: “The process starts at the Faucet Position. Hence, if the Faucet Position is increased (that is, the faucet is opened further) then the Water Flow increases. Similarly, if the Water Flow increases, then the Current Water Level in the glass will increase. Another element present in the process is the Gap, which is the difference between the Desired Water Level and the Actual Water Level. From this definition, it follows that an increase in Actual Water Level decreases the Gap. In addition, the process considers that a greater value for Gap presumably will lead to an increase in the Faucet Position (as you attempt to fill the glass). Finally, considering the Desired Water Level and Gap relationship, the Gap will influence in the same direction along this link the Desired Water Level”. Answer the following question: a. Draw the Causal-loop diagram (10 marks) b. Establish the sign (positive or negative) for every link (10 marks) c. Establish the sign (positive or negative) for the whole model (10 marks) 2. Practical questions: Discrete-event based car-station model (70 Marks) Process description (see figure 1): Figure 1. The car station service layout. In this car station service, the managers wants to understand the behaviour of their customers, as well as from their own processes, in a period of 1 month. Which considers 30 days, and every day considers 10 hrs of available time. In this service system, there are three main services provided in following stations: (1) Washing car, (2) refuel and (3) mini-market for customers. Questions to: Dr. Jorge Hernandez –
[email protected] The sub-processes are described as follows: Car arrives to each station considering an inter-arrival time uniformly distributed as follows: Station 1, maximum every 5 min and minimum 1 min; Station 2, maximum 10 min and minimum 3 min, Station 3, maximum 20 min and minimum 2 min. All services are capacitated. In the system, there are two sequences. For sequence 1, if a car arrives at Station 1, and the queue is full, it goes to station 3. For sequence 2, if cars arrives to Station 2 and if the queue is full, it goes to station 3. If any car arrives to station 3, and if the queue is full, the car will leave the system. Station 1: This washing station considers three processes, and each process has its own queue. Each process has a capacity for one car, and each queue can have two cars. The processes are: Wash, Dry and Polish. There are two types of services here. The first one, named ultra clean, and the second, named normal clean. For the first one, the cycle times for each of these processes are 5min, 3min and 2min, respectively and has total cost of £30. For the second, the cycle times for each process are 2min, 1min and 1min, respectively and has a total cost of £15. Every day, the average value for cars arrival is that out of the total, the 45% will chose the ultra clean service, and the 55% will chose the normal clean service. Once the car is ready, they leave the system. Station 2: In this station, gas-refuelling service is provided. The operation time, in minutes, is distributed by a triangular distribution as follows: Minimum value 4 min, Maximum value 10 min and most probable value of 7 min. Every car fills the same type of gas. There are six gas stations slots to be used. Each station slot considers a queue of five cars of capacity. Once the refuelling process has finished, the car leaves the system. In every gas station slot, there is one labour to serve the refuelling. In addition every gas station slot requires a cleaning after every 20 cars and the cleaning duration is 2 min. There is one specific labour for cleaning. Once the car is ready, they leave the system. Station 3: This station is named the kiosk, where people do takeaway or buy certain products to satisfy their needs. The Kiosk car parking capacity is 30. Once a car arrives, only the driver will get off the car to buy or collect their products in the kiosk. In this kiosk there are four payment points to receive customer payments. Once in the kiosk, 60% of customers use the take-away system, which means they just attend the kiosk to collect their products, which implies a cycle time uniformly distributed considering a maximum time of 3 min and a minimum time of 1 min. The other 40% customers buy products directly in the kiosk. For this case, the cycle time for selection of products is distributed normally considering an average time of 10 min and a standard distribution of 1 min. The service time at the each payment counter is 1 min, which is attended by one labour. Once the payment is done, customers return to their cars and leave the system. Answer the following question: a. Provide a detailed description on how the model works (you can use flow charts or any other modelling language to support your analysis) and discuss the performance of the current As-Is model by identifying the main information provided in every element (parts, machines, labours, buffers, attributes, histogram). (15 marks maximum) Questions to: Dr. Jorge Hernandez –
[email protected] b. Model this exercise with WITNESS, and provide different scenarios in order to identify how you would improve the performance and validate this with a new model (15 marks maximum). In addition, answers the following questions is required: - Provide a solution where you can establish the best (optimal if possible) number of cars to be attended per day in order to optimise the usage for each station, hence to optimise the queue line capacity usage. It is recommended to use the experimenter function of witness. (10 marks maximum). - Identify the bottlenecks and modify the capacities accordingly in order to increase stations usage? Identify variables on Witness to justify your answer (10 marks maximum). - Without compromising the current available services, suggest a new design of the operations to make the whole system more effective and use witness to try and test your suggestions (10 marks maximum). - Considering your own MSc programme, give some self-reflections and suggest your own contributions from your MSc programme (10 marks maximum).
