Greeshma05042005/Transport-demand-prediction-using-regression-models-and-ML-techniques.

This Data Science Capstone Project focuses on predicting passenger demand for Mobiticket bus routes in Kenya. Using regression models, exploratory data analysis, and machine learning techniques, the project helps optimize seat allocation, scheduling, and resource planning. Transport Demand Prediction (Data Science Capstone Project) Project Statement This project aims to build a **regression model** that predicts the number of seats Mobiticket can expect to sell for each ride, based on route, date, and time. The dataset covers 14 routes starting from towns in western Kenya and ending in Nairobi. Business Context Mobiticket operates bus/shuttle rides from multiple towns (e.g., Kisii, Migori, Keroka, Rongo, etc.) into Nairobi. The journey takes 8–9 hours, followed by 2–3 hours of traffic into the city. Traffic, travel times, and socio-economic events directly influence passenger demand. 👉 A reliable demand prediction model can help optimize **seat allocation, scheduling, and operational efficiency**. 📊 Dataset - **14 routes** ending in Nairobi - Routes originate from towns like: Awendo, Homa Bay, Kisii, Migori, Sirare, Sori, etc. - Key Nairobi stops: **Kawangware, Westlands, Afya Centre** - Data includes travel timings, traffic factors, and passenger booking info - 🛠️ Tools & Libraries - **Pandas** → Data manipulation & aggregation - **NumPy** → Efficient numerical operations - **Matplotlib & Seaborn** → Data visualization & EDA - **Scikit-learn** → Model training, optimization, and evaluation