Emergency routing system for buildings using Dijkstra’s algorithm

Kee, Chin Yew (2024) Emergency routing system for buildings using Dijkstra’s algorithm. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full Text) Emergency routing system for buildings using Dijkstra’s algorithm.pdf - Submitted Version Download (2MB)

Abstract

This project aims to develop an Emergency Routing System for buildings using Dijkstra's Algorithm to enhance the effectiveness of emergency evacuations. The primary objectives include assessing existing systems, developing an optimized routing algorithm, and evaluating its performance statistically. The system calculates the shortest and safest evacuation routes, minimizing evacuation time and congestion. Utilizing the Spiral Model, the project undergoes iterative cycles of planning, risk analysis, engineering, and evaluation for continuous refinement and risk management. Building plans are analysed using AnyLogic for simulation, providing accurate data for the routing algorithm. Potential risks like data accuracy and system failures are mitigated during the risk analysis phase. The engineering phase involves iterative design, prototyping, coding, and testing, ensuring system functionality and performance. The evaluation phase reviews progress, gathers stakeholder feedback, and assesses key metrics such as response time, route accuracy, system uptime, and user satisfaction. The proposed solution includes a human detection model using a pretrained TensorFlow API and an evacuation time estimation model based on Convolutional Neural Networks. Dijkstra's Algorithm calculates the best evacuation route considering predicted times rather than distances. Testing against algorithms like Bellman-Ford and Floyd-Warshall shows that the system improves evacuation times and accuracy while ensuring reliability. The system's advantages include real-time updates, adaptability to dynamic scenarios, and higher user satisfaction compared to traditional solutions. Future developments may include integration with smart building technologies and adaptation for larger-scale environments, enhancing its applicability in diverse emergency scenarios. This project aims to provide a reliable and efficient emergency routing system that improves evacuation success rates under varied conditions.

Actions (login required)

View Item