A comprehensive guide in R on how to use clustering to make Bike distribution more efficient for Madrid's bike rental system. Check the PDF in English as a way to digest all the information in a simpler way. You can also check the files and the analyses one by one.
In this repository, you will find common and advanced data analyses for bike rental systems. In particular, this repository contains:
-
Analysis of trip data from the bike rental system of Madrid, combined with coordinates of each station and weather data - clustering and assymetry
-
ARIMA/SARIMA Model development for Demand forecasting of bike rental systems
-
A novel approach on rebalancing more efficiently the system (correcting assymetry) using Time Series Clustering and Dynamic Time Warping distances.