- Carllos Watts-Nogueira
- Ranjan Baral
Course: Artificial Intelligence & Machine Learning University of San Diego / Fullstack Academy Section: 2504-FTB-CT-AIM-PT End Date: Jun/2025
BikeEase is a New York-based urban mobility company that provides flexible bike rental options to city residents and tourists. Committed to eco-friendly transportation, the company aims to optimize its operations using data-driven insights powered by AI and machine learning.
This project focuses on building a comprehensive analytics solution to support:
- Demand Forecasting Engine
- Operational Optimization Engine
- User Behavior Analysis
- Visualization Toolkit for decision-making
Developed an end-to-end solution for data import, cleaning, transformation, and visualization using historical bike rental data. The objective is to extract actionable insights and build the foundation for intelligent operations.
Source: FloridaBikeRentals.csv
Features include hourly rental data, weather conditions, holidays, operational status, and seasonality.
Core columns:
- Rented Bike Count
- Hour of the Day
- Temperature, Humidity, Wind Speed
- Visibility, Dew Point, Solar Radiation
- Rainfall, Snowfall
- Seasons, Holiday, Functioning Day
- Python
- Pandas, NumPy
- Seaborn, Matplotlib
- Data Cleaning & Transformation
- Missing value handling
- Duplicate removal
- Standardization & normalization
- Statistical Analysis
- Descriptive statistics
- Pivoting & aggregation
- Visualizations
- Line plots, bar charts, box plots, heatmaps
- Seasonal and hourly rental trends
- Correlation matrix
bike_rental_cleaned.json– Cleaned datasetbike_rental_processed.csv– Processed dataRental_Bike_Data_Dummy.csv– Encoded categorical dataplots/– Folder with saved visualizationsobservations_report.md– Summary of insights and recommendations
To explore and visualize urban bike rental data patterns using AI-friendly preprocessing and statistical techniques — setting the stage for demand forecasting and operational decision-making.