Citi Bike NYC Operations & Rider Insights

In this open-ended project, I explored publicly available Citi Bike trip data to uncover insights that could improve operational logistics and inform marketing strategies. With no set business objective, I guided the analysis by asking: Where and when are docking shortages likely to occur, and who are the primary users of the service?

Using Python (Matplotlib, Seaborn, Folium), I performed data mining, cleaning, time series analysis, and clustering. I applied linear regression and K-means clustering to:
• Identify peak hours and high-traffic stations needing more bikes or docks
• Explore demographic patterns among riders
• Segment usage behavior across time, location, and rider type

Geographic visualizations and Tableau dashboards illustrated key findings for stakeholders, while Excel summaries and GitHub documentation ensured reproducibility and transparency.

Skills used: Data mining, clustering, predictive modeling, time series analysis, data ethics, geographic visualization

Tools: Python (Jupyter Notebook, Matplotlib, Seaborn, Folium), Tableau, Excel, Quandl, GitHub