A total of 3,400 people lost their lives and 14,670 were injured in fires in the USA in 2017, a 9.6 percent increase over the 2007 casualty rate. Floods meanwhile killed more than 100 Americans last year, a number that has also been increasing. To counter this terrible toll, artificial intelligence researchers are developing systems to improve disaster prediction accuracy and provide timely evacuation guidance for panicked people during emergencies.
Providing real-time evacuation strategies is critical, as research shows that in emergencies many people tend to wait for instructions when they should already be proceeding to an exit. Simulation systems can play a valuable role in identifying and testing evacuation plans to enable individuals to promptly leave a dangerous area via the safest and fastest route.
Smart Evacuation Systems
Aeris – Connectivity Platform
Aeries’ IoT-enabled evacuation system tracks individuals’ movements in real time if a building catches fire. The system can tell the evacuation management team how many people have reached which exit points, and help locate any missing people. The versatile IoT system provides real-time updates and visibility to a rescue team to ensure evacuation proceeds as quickly and efficiently as possible, greatly reducing the risk of death or injury.
CIF Security (ZhongZhi ShengAn) –Intelligent Evacuation Solution
If a fire is detected the CIF Security intelligent system will evaluate and analyze information transmitted by the fire alarm system, then quickly design and provide efficient evacuation plans accordingly. The system can also conduct vision analysis using onsite CCTV data; and contains a simulation engine which can examine the impact of different routing options to further refine evacuation path output. The system provides smart evacuation signs and synthesized voice guidance to help people leave the building in the shortest time. CIF Security has established partnerships with South Rail Hub Chengdu, ZTL hotels, and others.
Fujitsu – Fire and Flood Evacuation
Japanese electronics giant Fujitsu has developed an evacuation app for smartphones designed to help vulnerable individuals make informed and timely evacuation decisions in case of fire or flood. The app provides a real-time map that shows their current location as well as nearby safe and evacuation places. The app also displays the possibilities of flooding in locations near the user as predicted by an AI algorithm, along with the number of people who have already evacuated to nearby shelters — an important feature in high-density, tsunami-prone Japan.
Safety and Security Robots
There is no doubt that robots are being increasingly deployed in public safety, and nowhere will they be more welcome than in the dangerous environments created by fire and flood emergencies. Bots are now used in analyzing fire trends, conducting search and rescue of trapped persons, monitoring hazardous variables and planning evacuation paths. They can function as smart fire alarms capable of detecting abnormalities such as airborne particle count or temperature within buildings. Also, in the event of fire, robust safety and security robots can remain in toxic environments to monitor real-time oxygen content, carbon monoxide concentration, fire trends, etc., to further optimize safest evacuation plans.
IoT and Data-based Prediction
If we can better predict the likelihood of fire or floods, we’ll have more time to develop specific measures to safeguard the public. Data from CCTV and other video cameras within a building can help AI systems detect fire-indicating pixels or patterns by analyzing features such as color and motion. IoT-based sensors meanwhile can provide additional environmental information such as temperature and air particle counts to enable machine learning models to predict emergencies and provide early warning.
Many global institutes are developing predictive models for disaster prevention. Georgia Tech and the Atlanta Fire Department cooperated to improve fire inspection efficiency with the predictive model “Firebird,” a big data tool that predicts buildings’ fire hazard potential. Since implementation, Firebird has demonstrated an impressive 71 percent accuracy across Atlanta.
The development and deployment of intelligent fire and flood prediction and evacuation guidance technologies and methods continues. It’s believed this research area will not only save lives today, but can also be expanded in the future to reduce death rates across additional emergency scenarios.
Author: Leyi Hu | Editor: Michael Sarazen