Case study
BIXI Analysis
An analysis of Montreal BIXI bike-share ridership against date, weather, time of day, location and pricing, to model the optimal fleet size by season.
Overview
Analysis of bike share ridership against date, weather, time of day, location and pricing, to find which of those actually move the number of rides and by how much.
One of several analyses produced under a contract for data-driven written content, where the output had to be both correct and readable by people who do not read charts for a living.
What I own
The analysis and the visualisations.
Complexities tackled
Weather and season are the same variable wearing two hats. Ridership rises in summer and rises in good weather, and summer is when the weather is good. Reporting either one without accounting for the other produces a confident number that means nothing. Separating them is most of the analytical work in a dataset like this.
Time of day is not one signal. A commuter peak and a leisure peak behave differently and respond to different things. Treating all rides as one population averages the two into a shape that describes neither.
The deliverable was an argument, not a notebook. The client wanted content, which means every chart had to carry one clear claim rather than showing everything the data contained.
Stack
Python with Pandas for the analysis, Matplotlib for the charts.
Working on something similar?
Tell me what you are building and what is in the way. I will tell you honestly whether I am the right person for it.