Exploring the economic and environmental effects of e-commerce implementation of green delivery schemes

ECON0052 2019-20 Delivery, Group 1

Alexander Craggs, Allysia Dee, Assel Issayeva, Wei Thong, Justin Wong

A repository for the agent-based model developed to explore how agents interact in an economy for e-commerce goods, and how

---

Statistics of the model output

Outputs of the model are saved in .csv files under the ./SavedStats directory. They are prefixed by their simulation model (benchmark, normal/ simple, and social simulations), and suffixed by whether the output is on the economy level or on the agent level (_simulation and _agent, respectively).

Visualisations of the model output

Visaulisations are done in Jupyter Notebook. They can be found within the ./Visualisations directory, which is accompanied with Python modules containing visualisation helper functions.

Running the simulations

Install Python and dependencies

The application was developed in Python 3.

To install Python code dependencies (outside of Pycharm), the easiest way is using PIP.

pip install -r requirements.txt

Configurations

The Engine model parameters are inputted in main.py, under Engine inputs. You can tweak the parameters of the Engine, such as number of agents, emissions level of green and normal delivery, and so on here.

After, you can start the agent-based model simulations by running main.py

python3 main.py

If everything is working properly you should see

Initialising engine
Initialising agent 0
Initialising agent 1
...

This means that the Engine and Agent classes are being initialised. Upon initialisation, you should see that the simulations begin to run, followed by dataframes containing the statistics of the model output.