Research on the locations of facilities has been very popular in many fields especially for emergency facilities. This study addresses the location of new fire stations based on dynamic fire risks. In order to give consideration to both fairness (well-performed coverage of demand points) and efficiency (rapid response and dealing with each demand point on its own weight) during the process of locating fire stations, a Multi-Level Gradual Coverage Location Model (MLGCLM), based on Maximal Covering Location Problem in the presence of Partial coverage (MCLP-P), is proposed to optimize fire station locations. A computational experiment tested on the data set used in MCLP-P validates that the proposed model has a better performance and demonstrates that the Genetic Algorithm (GA) procedure with adaptive operator designed on Matlab platform is able to effectively solve the problem. Then, this paper uses fire data of Hefei city to estimate the performance of the MLGCLM under different selected fire stations and compares the MLGCLM with the other two classic facility location models: p-median model and maximal covering location model. The results suggest that MLGCLM takes into account both pmedian and maximal covering objectives and can be useful in guiding the future optimization scheme of new fire stations in a city.

International Seminar on Fire and Explosion Hazards (9; 2019; Saint Petersburg, Russia). Proceedings of the Ninth International Seminar on Fire and Explosion Hazards [Электронный ресурс]: 21-26 April 2019, Saint Petersburg, Russia. Vol. 2 / Peter the Great St. Petersburg Polytechnic University, Autonomous Non-Profit Organization "Fire and Explosion Safety", Gefest Holding Ltd ; [edited by A. Snegirev [et al.]. — Электрон. текстовые дан. (1 файл : 98,7 Мб). — Saint Petersburg, 2019. — Загл. с титул. экрана. — Свободный доступ из сети Интернет (чтение, печать, копирование). — Adobe Acrobat Reader 7.0. — <URL:http://elib.spbstu.ru/dl/2/k19-97.pdf>. — <URL:http://doi.org/10.18720/SPBPU/2/k19-97>.