.png){kind=icon}
.png){kind=icon}
Return
A balanced evacuation algorithm for facilities with multiple exits
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ejor.2020.07.012.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Over the last few years, there has been an increase in the number of disasters caused by human crowding due to evacuees attempting to exit during emergencies. This has led to a rise in optimization studies on emergency evacuation plans that mitigate the loss of life and injury. Although a substantial amount of this research focuses on guiding evacuees toward the nearest emergency exits, they do not always consider the problem of congestion. This paper will present an algorithm that will ensure that pedestrians safely evacuate facilities with multiple exits by introducing a balanced evacuation algorithm: BEME. This approach will help reduce overcrowding and congestion surrounding the exits by overcoming the limitations of traditional strategies such as approaching the nearest exit and ensuring optimal evacuation. BEME's performance was compared with two established artificial intelligence techniques: simulated annealing and depth-first search. The evacuation model evaluation considered a number of variations in the spatial placement of the exits, number of exits, and number of pedestrians. The results showed that the proposed algorithm could significantly reduce the number of pedestrians for every exit. BEME differs from the benchmarked techniques, as it is consistent regarding the various exit placements and hence supports the algorithm's use with existing infrastructures that utilize inefficiently placed exits. Moreover, this approach resolves the problem of overcrowding and congestion around exits using a balanced evacuation that helps maximize safety and avoid life-threatening hazards. (C) 2020 Elsevier B.V. All rights reserved.
Keywords:
Simulation
Emergency evacuation
Multiple exits
Simulated annealing (SA)
Depth-first search (DFS)
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6
Papers:
2.2W
Citations:
6.4W