Article: Intelligent decision system for urban emergency management based on combined deep learning and optimisation algorithm Journal: International Journal of Environmental Technology and Management (IJETM) 2026 Vol.29 No.7 pp.59 - 85 Abstract: Environmental emergencies bring drastic fluctuations in urban public resource demands, and traditional scheduling methods suffer from low accuracy, slow response and unbalanced resource allocation. This paper proposes a hybrid model combining long short-term memory (LSTM) and grasshopper optimisation algorithm (GOA) for intelligent public service scheduling. The LSTM is adopted to capture time-series features and predict demands of medical, transportation and energy resources, while GOA optimises model hyperparameters and scheduling strategies to avoid local optima. Multiple comparative experiments are conducted on real emergency datasets. The prediction accuracy for medical resources is improved by 67.57%, and resource allocation fairness and comprehensive performance are also significantly enhanced. Robustness tests verify its stable performance under noisy and incomplete data conditions. This hybrid approach provides an effective intelligent decision tool for urban environmental emergency management and public resource allocation, and can be widely applied to various urban emergency scenarios. Inderscience Publishers - linking academia, business and industry through research

Title: Intelligent decision system for urban emergency management based on combined deep learning and optimisation algorithm

Authors: Xiaodan Cai; Jianlin Qiu; Haiyan Cao; Fang Wu; Xiangxiang Mei

Addresses: School of Yonyou Digital and Intelligence, Nantong Institute of Technology, Nantong, 226001, China ' School of Yonyou Digital and Intelligence, Nantong Institute of Technology, Nantong, 226001, China ' School of Yonyou Digital and Intelligence, Nantong Institute of Technology, Nantong, 226001, China ' School of Yonyou Digital and Intelligence, Nantong Institute of Technology, Nantong, 226001, China ' School of Yonyou Digital and Intelligence, Nantong Institute of Technology, Nantong, 226001, China

Abstract: Environmental emergencies bring drastic fluctuations in urban public resource demands, and traditional scheduling methods suffer from low accuracy, slow response and unbalanced resource allocation. This paper proposes a hybrid model combining long short-term memory (LSTM) and grasshopper optimisation algorithm (GOA) for intelligent public service scheduling. The LSTM is adopted to capture time-series features and predict demands of medical, transportation and energy resources, while GOA optimises model hyperparameters and scheduling strategies to avoid local optima. Multiple comparative experiments are conducted on real emergency datasets. The prediction accuracy for medical resources is improved by 67.57%, and resource allocation fairness and comprehensive performance are also significantly enhanced. Robustness tests verify its stable performance under noisy and incomplete data conditions. This hybrid approach provides an effective intelligent decision tool for urban environmental emergency management and public resource allocation, and can be widely applied to various urban emergency scenarios.

Keywords: long short-term memory; LSTM; grasshopper optimisation algorithm; GOA; intelligent scheduling; urban public service; environmental emergency management; resource allocation.

DOI: 10.1504/IJETM.2026.155736

International Journal of Environmental Technology and Management, 2026 Vol.29 No.7, pp.59 - 85

Received: 22 Apr 2026
Accepted: 29 Jun 2026
Published online: 11 Aug 2026 *

Title: Intelligent decision system for urban emergency management based on combined deep learning and optimisation algorithm

Keep up-to-date