Supermarket Simulation

Modeling and visualization of the movement of supermarket visitors based on real customer data.

This project uses

• Monte-Carlo Simulation
• Markov-Chain Modeling and
• the Pygame library

to predict movement of customers based on previously captured data. By creating two transition matrices, we can predict the probability of customers to move to a different section of the supermarket while also taking into account the time of the day.

It was written in week 8 of the Spiced Data Science bootcamp in Berlin.

Visualization of simulated customer-movement:

EDA:

Files

• eda_and_transition_matrix.ipynb: EDA and creating of transposition matrices
• customer_simulation.ipynb: Simulation and prediction of customer behavior
• supermarket_visual.py: Visualization of simulated customer-movement