VRP-Optimization

Vehicle Routing Optimization Project

Introduction

This project applies the Capacitated Vehicle Routing Problem (CVRP) to enhance delivery logistics for a local restaurant in Waterloo. Utilizing mathematical modeling and optimization algorithms, we aim to boost operational efficiency, reduce costs, and elevate customer satisfaction.

CVRP Formulation

Decision Variables

• xijk: Binary variable, 1 if vehicle i travels from location j to k, else 0.
• Loadij: Continuous variable for the load of vehicle i after visiting location j.

Objective Function

Minimize the total distance traveled by the fleet:

• n: Number of vehicles
• m: Number of addresses to deliver to
• Distancejk: Distance between address j and k

Constraints

• Visit each location exactly once, excluding the depot.
• Start and end routes at the depot (local restaurant).
• Vehicle load should not exceed capacity.
• Load changes after each delivery.
• No subtours or depot revisits mid-route.

Features

• Mathematical Optimization: Minimizes total travel distance with constraints.
• API Integration: Uses Google’s Distance Matrix API for distance calculations.
• Visualization: Employs Folium for interactive mapping and Matplotlib for plotting.

Requirements

• Python 3.x
• Pandas
• Google Maps API
• OR-Tools
• Folium
• Matplotlib

Report

For a detailed analysis and insights, refer to the complete report: VRP_GITHUB/Report.pdf

Team Member

• Banat Khural - GitHub

Course

CO 370: Deterministic OR Models
Instructor: Martin Pei - Profile

Key Findings

• Efficiency Improvement: Reduction in operational costs and enhancement of customer service through route optimization.
• Practical Application: Real-world application of theoretical models, linking academic concepts with business needs.

Conclusion

• Business Impact: Highlights the significance of route optimization in local business competitiveness and efficiency.
• Future Scope: Discusses potential enhancements, including scalability and real-time traffic data integration.

Visualization Output

For an interactive view of the routes, see the HTML map file:Interactive Map

(Note: The HTML file is best viewed in a web browser.)

##Thank You