Journal of Engineering, Project, and Production Management, 2026, 16(4), 2025-194
A Hybrid I-LSTM and GA-AP Clustering Framework for Urban Building Health Monitoring
1
Instructor, School of Architectural Engineering, Hebei Vocational
University of Industry and Technology, Shijiazhuang, 050091, China.
2
Instructor, Department of Basic Courses, Hebei Vocational University of
Industry and Technology, Shijiazhuang, 050091, China, E-mail: ljunhuajh@outlook.com
(corresponding author).
Project Management
Received September 5, 2025; revised October 27, 2025; accepted November 8, 2025
Available online May 29, 2026
Abstract: At present, current methods suffer from low monitoring accuracy and poor real-time performance in urban building health monitoring methods. To address these issues, this study proposes an urban building health monitoring method based on an Improved Long Short-Term Memory (I-LSTM) and a Genetic Algorithm-Affinity Propagation Clustering (GA-AP) algorithm. First, this study uses an Improved Long Short-Term Memory (I-LSTM) network to predict urban building settlement. These predictions are then used as input for a clustering model to classify health levels. By using the random forest algorithm to screen key features and inputting the selected features into a clustering model, the classification of health status levels has been achieved. This study utilizes genetic algorithms to optimize the parameters of clustering models and improve the accuracy of health status assessment. The experiment showed that the warning accuracy and response time of the research method were 94.58% and 0.18 seconds. In practical applications, the average number of errors in risk level classification was only 1 per week, and the number of missed detections was only 0.5 per week. The contour coefficient and average percentage error in the clustering process of health levels were 0.90 and 0.23. In addition, this method exhibited strong robustness in complex environments. The proposed method can effectively improve the intelligence level of the monitoring system, significantly enhance timeliness and reliability of the warning mechanism, and provide solid guarantees for urban building safety.
Keywords: LSTM, city building, health monitoring, GA, clustering, sparrow search algorithm.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Zhang, W. and Li, J. (2026). A Hybrid I-LSTM and GA-AP Clustering Framework for Urban Building Health Monitoring. Journal of Engineering, Project, and Production Management, 16(4), 2025-194.
