TY - JOUR
T1 - Suitability evaluation of emergency shelter site selection based on improved regret theory for toxic gas leakage accidents
AU - Wang, Jinghong
AU - Ge, Congcong
AU - Liu, Yuqing
AU - Wu, Jialin
AU - Wang, Yan
AU - Jiang, Juncheng
N1 - Publisher Copyright:
© 2025 Elsevier Ltd
PY - 2025/8
Y1 - 2025/8
N2 - During a disaster, shelters can effectively reduce the threat to human life safety and facilitate the rescue operations conducted by emergency responders. Hence, the rationality of emergency shelter site selection directly affects evacuation efficiency and rescue effectiveness. This study evaluates the suitability of emergency shelter site selection in the context of accidents in chemical industry parks. To address the limitations of existing evaluation models in accounting for toxic gas impacts, a risk function was introduced to refine the regret theory, enabling a more accurate reflection of toxic gas effects on shelter layout. Using an improved CTM model with the Tianjin Port area in China as the study site, evacuation route, time, and speed data were obtained and used as input values to calculate the utility and regret values for selecting different shelters. Given the significant impact of wind direction on gas dispersion, simulations, and evaluations were conducted with various wind direction parameters. The evaluation results indicate that under the influence of toxic gases, with ENE and NE wind directions, Shelter S4, located northwest of the Tianjin Port area, has the lowest regret value, with a regret value difference of 42.42 %–82.35 % compared to other location options. Under SE wind direction, the shelter with the lowest regret value is Shelter S2, located southeast of the Tianjin Port area. The regret value differences among the other nearby shelters are insignificant, with the maximum difference within 31 %. Overall, setting the shelter at location S4 is a more rational choice. The results provide a more scientific and reasonable reference for emergency shelter selection under the influence of toxic gases.
AB - During a disaster, shelters can effectively reduce the threat to human life safety and facilitate the rescue operations conducted by emergency responders. Hence, the rationality of emergency shelter site selection directly affects evacuation efficiency and rescue effectiveness. This study evaluates the suitability of emergency shelter site selection in the context of accidents in chemical industry parks. To address the limitations of existing evaluation models in accounting for toxic gas impacts, a risk function was introduced to refine the regret theory, enabling a more accurate reflection of toxic gas effects on shelter layout. Using an improved CTM model with the Tianjin Port area in China as the study site, evacuation route, time, and speed data were obtained and used as input values to calculate the utility and regret values for selecting different shelters. Given the significant impact of wind direction on gas dispersion, simulations, and evaluations were conducted with various wind direction parameters. The evaluation results indicate that under the influence of toxic gases, with ENE and NE wind directions, Shelter S4, located northwest of the Tianjin Port area, has the lowest regret value, with a regret value difference of 42.42 %–82.35 % compared to other location options. Under SE wind direction, the shelter with the lowest regret value is Shelter S2, located southeast of the Tianjin Port area. The regret value differences among the other nearby shelters are insignificant, with the maximum difference within 31 %. Overall, setting the shelter at location S4 is a more rational choice. The results provide a more scientific and reasonable reference for emergency shelter selection under the influence of toxic gases.
KW - CTM model
KW - Emergency shelter
KW - Regret theory
KW - Risk function
KW - Toxic gas leakage
UR - http://www.scopus.com/inward/record.url?scp=105000660000&partnerID=8YFLogxK
U2 - 10.1016/j.jlp.2025.105641
DO - 10.1016/j.jlp.2025.105641
M3 - 文章
AN - SCOPUS:105000660000
SN - 0950-4230
VL - 96
JO - Journal of Loss Prevention in the Process Industries
JF - Journal of Loss Prevention in the Process Industries
M1 - 105641
ER -