GitHub repository

What is this?

Final project for the "Simulation and Performance Evaluation" course at UniTN. The project is a simulation engine that can take a input file describing a matro system and produces a serties of log files describing various metrics.

Goal

Develop a discrete simulation engine capable of synthesising csv logs about various metrics. This engine was not supposed to be a 1:1 representation of the world, in fact it makes a few simplications. The logs are used by some python scripts in ordert to estimate the whole system performances.

What did we do?

We developed a simulation engine capable of taking as input a json file describing a metro system and simulate an arbitrary amount of time with various paramters such as: train speed, train crash rate, etc.
We developed a series of python script to analyse the key performace indicator such as the person served per minute or the average time waited before entering a train; we also inferred the fairness of the entire system using the Gini's index and the Lorenz gap.