This project focuses on Discrete Event Simulation of Simple Queues.
This study investigates the performance of queuing systems by comparing M/M/n, M/D/n and hyper-exponential queues through discrete-event simulations. It investigates the role of service rate distribution and scheduling policies - FIFO and SJF - on the average waiting time, especially under high system load. The results show that the multi-server model significantly reduces the waiting time, with SJF outperforming FIFO in terms of efficiency and fairness. The research also finds that the deterministic service model provides predictability, whereas the hyper-exponential model shows greater variability, posing a challenge for system management. These insights have practical implications for optimising real-world service systems.
Run all the cells in the Jupyter notebook file Li_14823551_Liu_14821672_Cao_15056910_assignment2.ipynb and get the visual results.
- Python
- Numpy
- Simpy
- Matplotlib
This project is licensed under the MIT License - see the LICENSE file for details.
- Mingjie Li 14823551
- Lingyuan Liu 14821672
- Qiziyi Cao 15056910
Processing: 100%|███████████████████████████████| 3/3 [00:00<00:00, 26.31file/s]
Total commits: 13
Total ctimes: 181
Total files: 5
Total loc: 1160
| Author | loc | coms | fils | distribution |
|:-----------|------:|-------:|-------:|:----------------|
| jazminli | 654 | 6 | 1 | 56.4/46.2/20.0 |
| LYLiu | 473 | 4 | 1 | 40.8/30.8/20.0 |
| Mingjie li | 22 | 1 | 2 | 1.9/ 7.7/40.0 |
| Qiziyi Cao | 11 | 2 | 1 | 0.9/15.4/20.0 |