Project Title: Predicting Demand for Shared Bikes

Overview

This project aims to build a multiple linear regression model to predict the demand for shared bikes based on various factors. The prediction model will help BoomBikes, a US-based bike-sharing provider, understand the factors influencing bike demand and prepare a strategic plan to meet customer needs post-COVID-19.

Table of Contents

• General Info
• Project Structure
• Setup and Requirements
• Technologies Used
• Conclusions
• Acknowledgements

General Information

• Background: The rising popularity of shared bikes as a sustainable and convenient mode of transportation requires effective demand forecasting to optimize bike availability.
• Business Problem: The key challenge is to accurately predict the number of bike rentals, which helps in efficient resource allocation and better customer service.
• Dataset: The dataset includes features such as weather conditions, temperature, humidity, wind speed, time of the day, and holiday status. This data is utilized to forecast the bike rental demand.

Project Structure

• day.csv: The dataset used for building the prediction model.
• bike_sharing_demand_prediction.ipynb: Jupyter notebook containing the code for data preprocessing, model building, and evaluation.
• README.md: This README file.

Setup and Requirements

Ensure you have the following Python packages installed:

• pandas
• numpy
• scikit-learn
• statsmodels

You can install the required packages using pip: pip install pandas numpy scikit-learn statsmodels

Technologies Used

• python - version 3.11.9
  • Pandas - version 1.3.3
  • NumPy - version 1.21.2
  • Scikit-learn - version 0.24.2
  • statsmodels - version 0.14.2

Conclusions

• Key Factors Influencing Demand: Weather conditions, particularly temperature and humidity, significantly affect bike rental rates. Peak Demand Times: Demand peaks during certain hours of the day, highlighting the importance of time in predicting rentals.
• Holiday Effect: There is a noticeable difference in demand on holidays compared to regular days, indicating that holiday status is a crucial feature.
• Model Performance: The multiple linear regression model provides a good fit, explaining a significant portion of the variance in bike demand.

Acknowledgements

• This project was inspired by real-world business case from United States.

Contact

Created by [@ManisCodeBase] - feel free to contact me @ [email protected]!