Car sharing is one of the earliest implementations of a shared economy and promises to be a very useful and eco-friendly way for individual transportation. There are, however, certain challenges that impede the acceptance of this service. One such challenge is an incorrect distribution of cars in a city, in which the geolocational demand for cars does not coincide with their actual locations.
Certainly, this is just one problem out of many. In order to account for these problems and simulate the dynamics of car sharing in a city, it is handy to simulate it in a computer programme, where one could apply interesting interventions to change people's behaviour, guiding them to leave/re-distribute cars to more optimal locations in a city. For instance, one could think of providing small nudges (e.g. monetary incentives) for leaving a shared car in a sub-optimal location for an individual's purpose, yet in a more optimal location for matching public demand.
cd car-sharing-simulator
python3 agent.py
All parameters of the simulation can be found in ./agent.py. I will be improving the presentation of these in future versions.
- Green dots: available cars
- Red dots: in-use cars
- Black dots: reserved cars
- Black circle: destination demand hotspots (people tend to go there more often during certain parts of the day, e.g. shopping mall)
- Purple circle: individual car demand (radius of the circle determines how far a person is willing to walk to get to a car)
In this repo, I provide some code for simulating car sharing within a city. For any questions about the functions of the code, feel free to contact me or raise suggestions in this repo.