Dynamic Pricing Strategy for Ride-Sharing Service

Project Overview

This project implements a dynamic pricing model for a ride-sharing service. It adjusts ride costs based on current demand (number of riders) and supply (number of drivers) to optimize pricing and improve profitability.

Features

• Dynamic price calculation based on supply and demand
• Data preprocessing pipeline
• Machine learning model for price prediction
• Profitability analysis
• Visualization of results

Components

1. Data Preprocessing

The data_preprocessing_pipeline function handles:

• Identification of numeric and categorical features
• Handling of missing values
• Outlier detection and treatment using IQR method

2. Dynamic Pricing Model

Calculates demand_multiplier and supply_multiplier based on current market conditions to adjust ride prices.

3. Profitability Analysis

Compares the adjusted ride cost with historical costs to determine the profitability of each ride under the new pricing model.

4. Price Prediction Model

A machine learning model that predicts ride prices based on:

• Number of riders
• Number of drivers
• Vehicle type (Economy/Premium)
• Expected ride duration

5. Visualization

Includes several visualizations:

• Donut chart showing the distribution of profitable vs. loss-making rides
• Scatter plot comparing actual vs. predicted prices

How to Use

1. Preprocess your data:

   preprocessed_data = data_preprocessing_pipeline(raw_data)

2. Apply the dynamic pricing model:

   data['adjusted_ride_cost'] = calculate_dynamic_price(data)

3. Analyze profitability:

   data['profit_percentage'] = calculate_profit_percentage(data)

4. Predict prices for new rides:

   predicted_price = predict_price(number_of_riders, number_of_drivers, vehicle_type, Expected_Ride_Duration)

5. Visualize results:

   plot_profitability_distribution(data)
   plot_actual_vs_predicted(y_test, y_pred)

Requirements

• Python 3.7+
• pandas
• numpy
• scikit-learn
• plotly

Installation

1. Clone this repository
2. Install required packages:

   pip install pandas numpy scikit-learn plotly
   

Future Improvements

• Incorporate more features into the pricing model (e.g., time of day, weather conditions)
• Implement A/B testing to compare different pricing strategies
• Develop a real-time pricing update system

Contributor

Vishal Bokhare