Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.783481
Title: Intelligent egress
Author: Ingram, Liam
ISNI:       0000 0004 7969 0664
Awarding Body: University of Edinburgh
Current Institution: University of Edinburgh
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Abstract:
Intelligent egress proposes a novel approach to enhancing the safety of evacuations from fire emergencies by means of way-finding systems that exploit real time information gathered from building sensor data. In standard approaches to fire evacuations in the built environment, occupants are rarely provided with any information that could aid egress path selection. It is well known that occupants unfamiliar with a building will often re-trace their original route of ingress, or simply follow others. These factors can lead to inefficiencies that can violate design assumptions on egress route utilisation, resulting in a greater possibility of increased evacuation time and unnecessary queuing, creating a higher risk of occupants being exposed to hazardous conditions. This project has demonstrated the potential benefit of installing an intelligent egress system, across a range of building complexities, by use of simulated evacuations. BRE has developed the Monte-Carlo risk assessment tool, CRISP, which has been used throughout the project. A novel dynamic route planning system has been developed to utilise live sensor data from these CRISP simulations to produce effective evacuation plans in real time. The sensor data is constantly reviewed, with the selected paths being altered where appropriate. By directing occupants along safer paths it was possible to reduce overall exposure to danger with steered and un-steered evacuations being compared, using fractional equivalent dose (FED) as the means of discrimination. To represent the high probability that occupants will not always follow instructions during a real life event, a variety of obedience levels were also considered. Results indicated that the more complex the building layout and the more available egress routes the greater the potential benefits of increased system sophistication. The importance of the dynamic aspect of the system, updating route instructions according to the evolving environment has been demonstrated for all but the most benign fire events. Tests with instruction obedience of 50% have also been shown to result in lower FED levels than for un-steered evacuations. The benefits of modifying the system to the particulars of a building layout, by implementing specifically designed heuristics is also discussed. The potential benefits of a sensor driven dynamic route planning system have been conclusively demonstrated, which should encourage further investigation.
Supervisor: Welch, Stephen ; Hadden, Rory Sponsor: Engineering and Physical Sciences Research Council (EPSRC)
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.783481  DOI: Not available
Keywords: intelligent egress ; safety evacuation ; real time information ; building sensor data ; CRSIP ; dynamic route planning
Share: