🚑 Intelligent Emergency Dispatch System

📘 Overview

The Intelligent Emergency Dispatch System simulates real-time ambulance assignment to emergencies in a city, aiming to minimize response time while efficiently handling multiple emergencies and ambulance distributions.
The city is represented as a weighted graph, where:

• Nodes represent locations (e.g., hospitals, intersections, residential areas)
• Edges represent roads with associated travel times.

The system implements three algorithmic versions to analyze space–time trade-offs:

• Version A: Uses Heaps (Priority Queues) + Adjacency Lists with Dijkstra/BFS for nearest-ambulance selection.
• Version B: Uses Adjacency List + HashMap and Multi-Source Dijkstra for optimized global shortest-path updates.
• Version C: Uses k-d Tree / Range Tree for nearest-neighbor queries and local BFS/Dijkstra for regional pathfinding.

Each version is tested on synthetic city data and compared based on runtime, memory usage, and average response time, demonstrating how data structure design impacts real-world dispatch efficiency.

---

📂 Dataset Structure

This project uses synthetic datasets representing different aspects of the city and emergency dispatch system.
Each dataset is stored as a .csv file for modular design and easy experimentation.

1️⃣ city_nodes.csv

Represents all important locations in the simulated city.

Attribute	Description
node_id	Unique identifier for each location (integer).
name	Human-readable name of the location (e.g., “Hospital A”, “Sector 5”).
latitude	Latitude coordinate of the location.
longitude	Longitude coordinate of the location.
type	Type of location — e.g., hospital, residential, commercial, intersection.

---

2️⃣ city_edges.csv

Defines the road network connections between different city nodes.

Attribute	Description
edge_id	Unique identifier for each road segment.
source_node	Starting node of the road (refers to node_id from city_nodes.csv).
destination_node	Ending node of the road.
distance_km	Physical distance between nodes (in kilometers).
travel_time_min	Average travel time between the nodes under normal conditions (in minutes).
road_condition	Road status (e.g., smooth, moderate, congested).

---

3️⃣ ambulance_locations.csv

Represents the initial locations and capacities of ambulances available for dispatch.

Attribute	Description
ambulance_id	Unique identifier for each ambulance.
current_node	Current location of the ambulance (node_id from city_nodes.csv).
status	Current status — available, on-duty, maintenance.
capacity	Maximum number of patients that can be transported (usually 1–2).
avg_speed_kmph	Average speed of the ambulance (used to estimate response time).

---

4️⃣ emergency_requests.csv

Contains simulated emergency call data to test dispatch algorithms.

Attribute	Description
emergency_id	Unique identifier for each emergency event.
timestamp	Time of emergency occurrence.
location_node	Node ID of the emergency’s location.
urgency_level	Severity level of the emergency (Low, Medium, High, Critical).
num_patients	Number of patients involved.
type_of_emergency	Type of incident (e.g., accident, cardiac arrest, fire injury).

---

5️⃣ hospital_data.csv

Details about hospitals in the city where ambulances deliver patients.

Attribute	Description
hospital_id	Unique identifier for each hospital.
node_id	Node ID corresponding to the hospital’s location (city_nodes.csv).
name	Hospital name.
available_beds	Number of beds currently available for patients.
emergency_capacity	Maximum capacity for handling simultaneous emergencies.
avg_treatment_time_min	Average time taken for patient stabilization.

---

6️⃣ traffic_conditions.csv

Represents real-time or simulated dynamic traffic variations in the city.

Attribute	Description
timestamp	Time when traffic data was recorded.
edge_id	Road segment (city_edges.csv) affected by traffic.
traffic_level	Traffic intensity (Low, Medium, High).
delay_factor	Additional delay multiplier applied to travel_time_min based on traffic conditions.

---

⚙️ Workflow Overview

1. Data Generation: Synthetic datasets (CSV files) are generated using Python.
2. City Graph Construction: Load city_nodes.csv and city_edges.csv to build a weighted graph.
3. Algorithm Implementation:
   • Version A: Dijkstra/BFS + Priority Queue.
   • Version B: Multi-Source Dijkstra with HashMaps.
   • Version C: k-d Tree for nearest-neighbor + local Dijkstra.
4. Emergency Simulation: Load emergency_requests.csv and assign ambulances dynamically.
5. Performance Evaluation: Compare algorithms using metrics:
   • Average response time
   • Memory usage
   • Computation time
6. Visualization: Use network plots and real-time dispatch animations (optional).

---

📊 Example Metrics

Metric	Description
Avg Response Time	Average time taken for an ambulance to reach the emergency.
Memory Utilization	Amount of memory consumed by each version.
Computation Time	Total runtime for processing all emergency requests.

---

🧠 Key Learning Outcomes

• Understanding of graph-based optimization for real-world systems.
• Analysis of data structure efficiency under varying workloads.
• Insights into space–time trade-offs in algorithm design.
• Simulation of real-time decision-making in emergency services.

---

🧩 Technologies Used

• Python 3.x
• Pandas, NumPy, NetworkX, Matplotlib
• Heapq / PriorityQueue
• Scikit-learn (for k-d Tree implementation)

---

📁 Repository Structure

Intelligent-Emergency-Dispatch-System/
│
├── data/
│   ├── city_nodes.csv
│   ├── city_edges.csv
│   ├── ambulance_locations.csv
│   ├── emergency_requests.csv
│   ├── hospital_data.csv
│   └── traffic_conditions.csv
│
├── src/
│   ├── version_A_dijkstra.py
│   ├── version_B_multi_source.py
│   ├── version_C_kd_tree.py
│   └── utils.py
│
├── results/
│   ├── runtime_comparison.csv
│   └── response_time_analysis.png
│
├── README.md
└── requirements.txt

---

🧾 License

This project is open-source under the MIT License — free to use, modify, and distribute with attribution.

---

✨ Team Members

Sanjana Nathani
🎓 M.Sc. Data Science | DAU, Gandhinagar
📍 Vadodara, Gujarat, India
📧 sanjananathani06@gmailcom

Urvi Kava
🎓 M.Sc. Data Science | DAU, Gandhinagar
📍 Vadodara, Gujarat, India
📧 urvikawa2004@gmailcom

Aditya Jana
🎓 M.Sc. Data Science | DAU, Gandhinagar
📍 Vadodara, Gujarat, India
📧 adityajana20.6@gmailcom

---

💡 Acknowledgments

Special thanks to open-source libraries and contributors that make simulation and data analysis possible — NetworkX, Pandas, NumPy, and Matplotlib.