Optimization of Mobile Emergency Team Deployment in Water Crises Using Neural Networks (Case Study: Pakdasht, Varamin, Pishva and Qarchak Counties)

Moazezifarhadifar, Malihe; Tavakoli, Iman

doi:10.22093/wwj.2025.544414.3513

Optimization of Mobile Emergency Team Deployment in Water Crises Using Neural Networks (Case Study: Pakdasht, Varamin, Pishva and Qarchak Counties)

Articles in Press

Document Type : Research Paper

Authors

Malihe Moazezifarhadifar ¹

Iman Tavakoli ²

¹ Master Graduated Expert of Energy, Wastewater Development and Operation Part, Southeastern Tehran Province Water and Wastewater Company, Varamin, Iran

² Master Graduated Expert of Geographic Information System, Water Development and Operation Part, Southeastern Tehran Province Water and Wastewater Company, Varamin, Iran

10.22093/wwj.2025.544414.3513

Abstract

The water scarcity crisis in Tehran Province and the Varamin Plain, coupled with frequent water and power outages, has led to air entrainment in pipelines, increased pressure, and pipe bursts, highlighting the need for intelligent water management in the deployment of emergency response teams. The innovation of this study lies in applying artificial intelligence–based methods for spatial analysis of incidents and identifying optimal locations for the deployment of mobile emergency teams. The main focus is on reducing response time and improving service coverage through the determination of optimal points using the Self-Organizing Map algorithm. This innovative approach can contribute to the development of integrated software for incident management and faster decision-making under emergency conditions. The study area covers four counties: Varamin, Pishva, Qarchak, and Pakdasht. Using the Self-Organizing Feature Mapneural network algorithm, 16 optimal points for the deployment of mobile emergency teams were identified to enhance coverage and reduce response time, thereby minimizing losses caused by water and power disruptions. A total of 3,603 incident points recorded in the GIS system with UTM coordinates (Zone 39N) were used for spatial analysis and clustering in MATLAB. The research process included spatial data collection and preprocessing, SOM execution, and output map generation in the GIS environment. Among the 16 optimal locations, 9 are situated in high-incident-density areas (more than 8 incidents per square kilometer), and 6 are located in zones with high customer density (over 1,666 customers per square kilometer). The distribution of other points across lower-density zones ensures adequate coverage of rural and sparsely populated areas. The results indicate that the SOM algorithm successfully identified spatial patterns of incidents and population density, achieving balanced and efficient site selection for mobile emergency teams. The main advantage of SOM lies in its ability to analyze two-dimensional spatial data precisely, preserve topological structure, and adapt to data variability-making it superior to other clustering and metaheuristic methods. The findings confirm that the SOM algorithm is an effective approach for urban crisis management and optimal deployment of emergency resources, with potential for further development using more complex datasets to enhance rapid response systems.

Keywords

Optimal Management, Water Resources, Neural Network, SOFM, Mobile Emergency Services, Crisis Management

Subjects

Monitoring

Ahmed, B. and Forte, R., 2016. Landslide risk zoning applying Kohonen’s self-organizing map neural network technique. Paper Presented at the 1^st Bangladesh Planning Research Conference (BPRC), Department of Urban and Regional Planning, Jahangirnagar University, Dhaka, Bangladesh. https://doi.org/10.13140/RG.2.1.2748.7766/1.

Bozorgmehr, M., 2021. Location of emergency service centers of provincial gas distribution companies (case study: North Khorasan Gas Company – Bojnord City). MSc. Thesis, Eshragh Institute of Higher Education, Bojnord, Iran. (In Persian). [Link]

Cottrell, M. and Verleysen, M., 2006. Advances in self-organizing maps. Neural Networks, 19(6-7), 721-722. https://doi.org/10.1016/j.neunet.2006.05.011.

Davoudi, R., Dehghanian, F. and Pirayesh, M., 2014. Location of emergency service vehicles for maximizing expected coverage. Proceedings of the 10^th International Conference on Industrial Engineering, Tehran, Iran. (In Persian). [Link]

Gong, X., Liang, J., Zeng, Y., Meng, F., Fong, S. and Yang, L., 2022. A Hierarchical Multi-Objective Programming Approach to Planning Locations for Macro and Micro Fire Stations, In Neri, F., Du, K. L., Varadarajan, V. K., Angel-Antonio, S. B. and Jiang, Z. eds. Computer and Communication Engineering (CCCE 2022). Cham: Springer, Communications in Computer and Information Science, pp.1630. https://doi.org/10.1007/978-3-031-17422-3_16.

Huang, F., Yin, K., Huang, J., Gui, L. and Wang, P., 2017. Landslide susceptibility mapping based on self-organizing-map network and extreme learning machine. Engineering Geology, 223, 11–22. https://doi.org/10.1016/j.enggeo.2017.04.013.

Kohonen, T., 2001. Self-Organizing Maps. 3^rd Ed. Springer Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-56927-2.

Negnevitsky, M., Voropai, N., Kurbatsky, V. and Tomin, N., 2013. Development of an intelligent system for preventing large-scale emergencies in power systems. In: Power and Energy Society General Meeting (PES), IEEE, Vancouver, BC, Canada. https://doi.org/10.1109/PESMG.2013.6672099.

Nikoo, M. R. and Mahjouri, N., 2013. Water quality zoning using probabilistic support vector machines and self-organizing maps. Water Resources Management, 27(7), 2577-2594. https://doi.org/10.1007/s11269-013-0304-5.

Panasetsky, D. and Tomin, N., 2010. Using of neural network technology and multi-agent systems to preventing large-scale emergencies in electric power systems. In: IEEE Conference Proceedings. https://doi.org/10.1109/IYCE.2013.6604142.

Smirnov, A. V., Levashova, T., Krizhanovsky, A., Shilov, N. and Kashevnik, A., 2009. Self-organizing resource network for traffic accident response. In: IEEE Conference Proceedings, Sweden. [Link]

Sokhansefat, G., Delavar, M. R., Nadi, S. and Khamespanah, F., 2011. Risk assessment in urban planning for disaster management using Kohonen self‑organizing feature map neural network, In Proceedings of the ISPRS International Workshop on Geospatial Information for Disaster Management (Gi4DM), Tehran, Iran. [Link]

Wu, Z., Xue, W., Xu, H., Yan, D., Wang, H. and Qi, W., 2022. Urban flood risk assessment in Zhengzhou, China, based on a D-number-improved analytic hierarchy process and a self-organizing map algorithm, Remote Sensing, 14(19), 4777. https://doi.org/10.3390/rs14194777.

Yu, Z., Sohail, A., Nofal, T. A. and Tavares, J. M. R. S., 2022. Explainability of neural network clustering in interpreting the COVID-19 emergency data. Journal of Biological Systems, 30(5), 2240122. https://doi.org/10.1142/S0218348X22401223.

Zare, M., 2021. Water crisis in Varamin plain, Shargh Daily. (In Persian). [Link]