AI Evacuation Routing with Dynamic Hazard Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing evacuation systems are inefficient in handling large venues with numerous occupants and exits, as they rely on static data and can direct people towards dangerous or blocked exits, leading to excessive computation time and resource requirements, making them unsuitable for safe evacuation in emergency situations.

Innovation Solution

A system utilizing a digital venue layout, hazardous event detection, indoor location tracking, convolutional neural networks, and AI-based rescue systems to optimize evacuation plans in real-time, identify lost individuals, and dynamically update exit directions, reducing computational resources and improving response speed.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If static data is used for evacuation calculation, then the system is simple to implement, but the system directs people towards exits that may have become dangerous or blocked

Engineering Contradiction:

Improvesystem implementation simplicityVSAvoidevacuation route safety

Core Design Contradiction:

Ease of manufactureVSReliability

Solution Approach 1:

The system transitions from static evacuation plans to dynamic real-time evacuation routing. The hazardous event detection device continuously monitors the venue and updates evacuation routes dynamically based on current threat locations and exit status, ensuring people are directed away from dangerous or blocked exits while maintaining system reliability

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system implements feedback loops where hazardous event detection data and real-time location tracking information continuously feed back into the evacuation calculation module. This allows the system to adapt evacuation routes based on changing conditions, resolving the contradiction between simplicity and safety by automating the update process

Inventive Principle:

Principle #23Feedback

2Reliability

If optimal routes are calculated for thousands of occupants and tens of exits, then evacuation coverage is comprehensive, but computation time becomes too long

Engineering Contradiction:

Improveevacuation plan completenessVSAvoidcomputation time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system pre-calculates and stores evacuation routes for various scenario configurations before an emergency occurs. When an emergency happens, the system quickly retrieves and adapts pre-computed routes rather than calculating from scratch, significantly reducing computation time while maintaining comprehensive evacuation coverage for thousands of occupants

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system divides the large-scale evacuation problem into smaller manageable segments by processing occupants in groups or zones, and calculating routes for subsets of exits simultaneously. This segmentation allows parallel computation, reducing overall computation time while ensuring all occupants are accounted for in the comprehensive evacuation plan

Inventive Principle:

Principle #1Segmentation

Data Source

PatentEP3989136B1Venue evacuation system

Publication Date: 2023.04.05 LUXEMBOURG INSTITUTE OF SCIENCE AND TECHNOLOGY (LIST)

Figure 1
Figure 2

AI summary

The invention relates to an AI-based rescue system (2) for assisting the safe evacuation of a venue upon occurrence of an emergency situation (fire, ship sinking, terrorism, etc.). Real-time simulators (20.2) assist the AI in defining the optimal evacuation plan. Appropriate exit paths are then dynamically communicated to the individuals. The real-time simulators (20.2) base their simulation on pre-simulated scenarios (20.11). The AI-based rescue system (2) also provides detection and assistance to lost individuals.