Route-Optimization-And-Visualization

Project Overview

This project delivers a two-day food-truck tour around the IUPUI campus in Indianapolis. It combines Google Maps APIs with Python-based route-optimization algorithms to generate a data-driven itinerary across 14 food-truck locations, minimizing travel time while maximizing coverage.

Where's the Code?

Recommended: Use this link to launch the Google Colab environment of the project for an interactive experience:

Background and Motivation

In October 2023, during my junior year at Indiana University–Purdue University Indianapolis (IUPUI), I developed this project as part of my application to the Xtern Talent Program, a selective initiative that connects emerging technology talent with Indiana-based companies.

Applicants were given a deliberately open-ended prompt:

“Create a weekend foodie plan around Indianapolis utilizing Google Maps APIs.”

Rather than submitting a simple itinerary, I treated the prompt as a technical challenge: how could a casual planning task be turned into a problem that meaningfully demonstrated data science and software engineering skills?

With only one weekend available between schoolwork and extracurriculars, I completed the entire project—from initial concept to final submission—under significant time pressure, requiring disciplined scope management and strategic prioritization.

Technical Approach

I reframed the weekend plan as a route-optimization problem. The goal was to determine the most efficient way to visit a large number of food trucks near campus within a limited time window.

To accomplish this, I:

Pulled real-world location and distance data using Google Maps APIs
Modeled the itinerary as an optimization problem
Used Python-based algorithms to compute efficient travel sequences
Produced a structured two-day schedule that minimized travel overhead while preserving practical usability

This turned a vague prompt into a reproducible, data-driven solution grounded in real constraints.

Results and Impact

The project was a major factor in my selection for the Xtern Program, which evaluated more than 1,000 applicants through four competitive elimination rounds. Submissions were reviewed by industry professionals for both technical depth and execution quality.

Advancing through the program led to interviews with multiple companies, including Sallie Mae and Franciscan Health. I accepted an internship with Sallie Mae for Summer 2024, which continued part-time through my senior year and culminated in a full-time offer in May 2025.

This project directly initiated a long-term professional relationship and demonstrated my ability to:

Identify real-world problems
Integrate multiple APIs
Apply algorithmic thinking
Deliver a polished technical product under time constraints

Key Takeaway

This project exemplifies curiosity-driven engineering: starting with a loose requirement and elevating it into a technically rigorous, real-world solution that stands out through thoughtful use of data, APIs, and optimization techniques.

Project Architecture

Data Directory Structure

data/
├── Distance Matrix API (Data Tables)/
│   ├── baseline_distance_travel_data.csv      # Template structure
│   └── updated_distance_travel_data.csv       # Populated with API data
├── Initial Data Cleaning/
│   ├── baseline_food_trucks.csv               # Raw API output
│   └── cleaned_food_trucks.csv                # Manually cleaned dataset
└── Weekend Plan (Route and Location Information)/
    ├── foodtruck_information.csv              # Final food truck details
    ├── route_information.csv                  # Daily route information
    ├── Day1GoogleRoute.PNG                    # Day 1 visual route map
    └── Day2GoogleRoute.PNG                    # Day 2 visual route map

1. Data Collection Phase

Google Places API: Queries for food truck locations within 5 miles of IUPUI campus center (coordinates: 39.7743, -86.1764)
Output: baseline_food_trucks.csv containing name, address, rating, price level, URL, hours, location ID, coordinates, and initial distance data
Key Function: get_data(api_key) - retrieves food truck information from Google Maps Places API

2. Data Cleaning & Preparation

Input: baseline_food_trucks.csv
Output: cleaned_food_trucks.csv
Cleanup Actions:
- Removed locations with incompatible operating hours
- Removed low-rating locations (rating < 1.0)
- Manually added cuisine type data
- Standardized URLs with manual verification
- Added day-specific hours (Friday/Saturday)
- Cleaned coordinate formatting
- Assigned unique location IDs for easier sorting
Final Dataset: 14 food truck locations with complete operational information

3. Distance Matrix API Integration

Process:
1. Generated baseline distance matrix from cleaned locations to all destinations
2. Queried Distance Matrix API for precise distance and travel time data
3. Populated missing distance and duration metrics
4. Created distance matrix in kilometers for optimization algorithm
Data Files:
- baseline_distance_travel_data.csv - template structure
- updated_distance_travel_data.csv - completed with API data

4. Route Optimization

Algorithm: Python TSP (Traveling Salesman Problem) with simulated annealing heuristic
Input: Distance matrix (14 x 14 locations)
Output: Optimized route permutation minimizing total travel distance
Result: Efficient two-day itinerary with 7 locations per day

5. Weekend Foodie Plan

Deliverables:
- foodtruck_information.csv - All food truck details with route position assignments
- route_information.csv - Daily route information including travel times and cumulative distances
- Day1GoogleRoute.PNG - Visual route map for Day 1
- Day2GoogleRoute.PNG - Visual route map for Day 2
Features:
- Route positions for each location
- Operating hours (opening/closing times)
- Estimated 30-40 minute stops per location
- Total trip distance and duration per day
- All data organized by importance and relevance

6. Visualization

Generated Google Maps route images for both days

Day 1 Route Visualization:

Day 2 Route Visualization:

Technology Stack

Platforms

Google Colab: Cloud-based Jupyter notebook environment for code execution and documentation

Libraries & APIs

pandas: Data manipulation and CSV processing
googlemaps: Google Maps Places API and Distance Matrix API integration
python_tsp: TSP solving with dynamic programming and simulated annealing algorithms

Core Dependencies

pandas
googlemaps
python_tsp

Project Execution Flow

Required Order of Operations:

Initialize Packages - Install and import all dependencies
Google Maps API Integration - Set up API functions for data collection
Main Script - Execute data gathering with valid API key
Data Cleaning - Review and manually clean baseline dataset
Distance Matrix API Integration - Populate distance and travel time metrics
Route Optimization - Apply TSP algorithm to calculate optimal route
Weekend Foodie Plan - Generate final itinerary and visualizations

Key Functions:

get_input() - User prompt for data collection initiation
get_key() - API key input validation
get_data(api_key) - Retrieves food truck data from Google Places API
format_data(data) - Cleans and exports initial dataset to CSV
solve_tsp_simulated_annealing() - Optimizes route using simulated annealing heuristic

Design Decisions & Implementation Notes