IE372 Simulation Spring 2011 Term Project: Waste Transfer Station Part 1 due 18 May 2011, 17:00 Part 2 due 3 June 2011, 17:00
In this project you are expected to simulate a transfer station for waste management, which operates continuously. Small trucks loaded with waste arrive at the station according to an unknown probability distribution. Past data (500 observations) for interarrival times of trucks are given in file “arrivals.dst”. The number of “unit loads” of waste a truck contains has discrete uniform distribution with a minimum of 10 and a maximum of 30 loads. Once a truck arrives at the dumping area, it waits to unload the waste it carries until there is an available port. There are two operational ports in the facility. The time to unload a truck depends on the amount of waste in it, and it is simply 1.2 minutes per load on the truck. The weight of a unit load is 100 kg and it includes five main types of reusables, namely metal, plastic, glass, paper, and organic. There is also some amount of waste that cannot be r ecycled in a unit load. The content of a single unit load is as follows (all in kg). Metal ~ uniform (1, 5) Plastic ~ uniform (30, 40) Glass ~ uniform (10, 20) Paper ~ uniform (5, 10) Organic ~ uniform (5, 20) Remaining: Rest of the load Units of load dumped are gathered in a storage area to be later picked up by a crane. A single crane picks up unit loads one by one and drops them on a conveyor. Available space on the conveyor can account for 9 units of load. Once it is full the crane cannot put more unit loads on the conveyor. Unit loads on the conveyor move one step ahead only when another load is dropped dropped on the conveyor. conveyor. Otherwise Otherwise they wait on the conveyor conveyor for the next units to arrive. arrive. When a unit load arrives at the end of the conveyor, it is fed into a special purpose machine, the metal gatherer, which is used to remove the metal content. Metal gatherer can process one unit load at a time, so the conveyor stops feeding it while it is busy. The distribution of the process time of metal gatherer is dependent on metal content of the unit load and is uniform (2, 2+metal amount in unit load (kg)) minutes. At the end of the metal removal process, the whole metal content of the unit load processed is removed and directed to a small metal storage area. Metal gathered is then sent to a warehouse for recycling when a batch of metal from three unit loads is ready. After the metal content is removed, the remaining of a unit load is directed to one of the two manual process areas. If both areas are in use, the metal gatherer waits for one to be available. When a unit load is in the manual process area, the first worker group, consisting of 10 workers, starts to remove plastic and glass content of the load. After the first group of workers finishes their work, the second group, which also consists of ten workers, starts
working on the same unit load. This second group removes the paper and organic content of the load. The duration of the removal process depends on the number of workers a group has and the amount of content (plastic and glass, or paper and organic) the group is trying to remove; it is determined by the following function. Removal time =
0.12 × content amount (in kg) log(n +1)
minutes, where
n
is the number of workers
in the respective group. The worker groups cannot remove the content with 100% efficiency, so always a portion of the content remains in the unit load processed. The percentage of the content that can be removed depends on the number of workers according to the following function. Percent Removed
k = 1 − ×100 , where n
k
is a constant and
n
is the number of workers in
the respective group. Note that there should be at least two workers in a group. The constant k differs according to the type of content to be removed, and it is 0.6, 1.8, 0.3 and 0.7 for plastic, glass, paper and organic material, respectively. The removed plastic and glass are stored in separate areas. In separate single trucks, these materials are shipped to another warehouse. Trucks are always available and leave as soon as the weight of the respective material gathered exceeds 1000 kg. The removed paper is sold to a paper factory to be used as raw material. There is a power plant that processes organic waste as input to produce biogas. Every 36 hours a truck arrives and collects the organic material gathered in order to transport it to the power plant. However, organic waste older than 3 days is useless for the plant and the truck does not take it. Older organic material is separated and sold as compost. Note that, at the time the waste arrives at the transfer station in small trucks, the age of the organic material is uniformly distributed between 1 and 2 days. The factory has a zipper, which is used to reduce the volume of the waste remaining after the reusables are removed. When the weight of the remaining waste reaches 100 kg after all removal processes are completed, the zipper processes a single load in two minutes. The compressed waste leaves the system to be dumped in a landfill area. The reusable materials are sold at the following unit prices ($/kg): 2.55, 0.05, 0.76, 0.28, 0.17 and 0.10 for metal, plastic, glass, paper, organic and compost, respectively. The cost of disposing the remaining waste to the landfill area is $0.32 per kg. The management of the waste transfer station aims at keeping unit loads of remaining waste at the station for no longer than 180 minutes on the average. Provided that this primary goal is satisfied, they also want to attain a profit that pays back the investment cost in two years. Develop and run a SIMAN model of the current system to identify the problems that prevent
the management from achieving these goals. Then, try various reasonable scenarios to come up with satisfactory alternative solutions that meet the goals. In developing and evaluating your alternatives, consider the estimated costs of resources given below. Investment cost
Metal gatherer Crane Zipper Conveyor Port Manual process area
per unit $700,000 $900,000 $50,000 $100,000 $10,000 $2,000
Annual operation cost per unit
$10,000 $15,000 $3,000 $11,000 -
Monthly wage of a typical worker is $1,500. You may work on any justifiable modifications in the system provided that you state your assumptions. You should conduct statistical output analysis to justify your conclusions and suggestions.
Project Report Content Part 1: 1. Statement of the Problem: One paragraph explanation of the purpose and objectives of the project, and design approach of your model. 2. Structure of the model: a) Table of all attributes used and their purposes. b) Table of all variables used and their purposes. c) A brief explanation of modules (distinguishable parts) of your model. 3. The Model: The efficiency of your model will be considered in evaluating your project. a) Listings of the model frame and the experiment frame. b) Model files to be uploaded via METU-Online. 4. Pilot Run Results: a) An explanation of how you verified your model and why you feel the model works correctly. Using end-of-run figures, develop some balance equations to show that none of your entities are lost. Also, show that your statistics are compatible with each other. b) SIMAN Summary Report for a single replication. c) Initial thoughts on alternative scenarios you intend to try. Part 2: 5. Output Analysis: a) Performance measures used. (Classify these as primary and secondary. Define primary measures to evaluate the system goals, and use these in deciding on statistical output analysis parameters. Use secondary measures for diagnostic purposes, i.e. to determine what prevents you from achieving the goals.) b) Output analysis to determine truncation point, replication length and number of replications (if replication/deletion method is used), batch size and number of batches (if batch means method is used). c) Generation of alternative scenarios. (Based on initial estimates found for your performance measures, explain how you came up with these scenarios.) d) Estimation of performance measures for each alternative scenario analyzed. e) Statistical comparison of alternative scenarios generated. 6. Conclusions: Your main conclusions regarding the estimates you found, suggestions to improve the system’s performance, and difficulties you experienced in conducting the simulation study.
The project report (Part 2) is due on Friday, 3 June 2011 by 17:00. An intermediate report (Part 1) is due on Wednesday, 18 May 2011 by 17:00. Part 1 should include the first four sections given in the report format above. After you submit the intermediate report, you will be given our SIMAN model of the system and may use this model in the second part of the project. You should also upload your .doe file and report file via Metu-Online. Create a ".zip" file that contains both the soft copy of your report and your ".doe" file. Do not forget to name your zip file as "YourGroupName.zip".
