The project aims to develop a predictive model for estimating the demand for shared bikes operated by BoomBikes, a US-based bike-sharing provider. With the ongoing challenges posed by the COVID-19 pandemic, BoomBikes seeks to understand the factors influencing bike demand post-pandemic to develop a robust business strategy. The dataset includes various features such as weather conditions, seasonality, and user types, which will be utilized to build a multiple linear regression model. Through data exploration, feature engineering, model building, and evaluation, the project aims to identify significant variables affecting bike demand and provide insights for BoomBikes to optimize their operations and meet customer expectations effectively.
-
This project aims to develop a predictive model for estimating the demand for shared bikes operated by BoomBikes, a US-based bike-sharing provider. The model will utilize various features such as weather conditions, seasonality, and user types to forecast bike demand post-COVID-19 pandemic. By understanding the factors influencing bike demand, BoomBikes can develop a strategic plan to optimize their operations and increase revenue.
-
BoomBikes has experienced significant revenue declines due to the COVID-19 pandemic and seeks to understand the demand for shared bikes post-pandemic to revitalize its business. By analyzing historical data and building a predictive model, BoomBikes aims to prepare for the market rebound and differentiate itself from competitors.
-
The primary business problem this project addresses is the need to forecast bike demand accurately post-COVID-19 pandemic. BoomBikes wants to identify key factors influencing bike demand to optimize inventory management, marketing strategies, and operational decisions. By doing so, BoomBikes aims to meet customer expectations, increase customer satisfaction, and ultimately boost revenue.
-
The dataset provided contains historical data on daily bike demands across the American market, including features such as weather conditions, seasonality, and user types. Additionally, it includes variables like 'casual,' 'registered,' and 'cnt,' where 'cnt' represents the total number of bike rentals, including both casual and registered users. This dataset will be used to train and evaluate the predictive model for bike demand forecasting.
-
Top 3 predictor variables that influences the bike booking are:
- Temperature
- Year
- Weather Situation
-
Another top 2 predictor variables that influences the bike booking are:
- Season
- Windspeed
-
Temperature has highest positive correlation with target variable.
- Python - 3.11.6
- Pandas - version 2.2.0
- Numpy - version 1.26.4
- Scikit-learn - version 1.4.1.post1
- Seaborn - version 0.13.2
- Matplotlib - version 3.8.3
- This Bike Demand Prediction Project is a part of my assignment for Post Graduate Diploma Degree in AI & ML at IIIT-Bangalore
Created by [https://github.com/imkushwaha/] - feel free to contact me!