Journal of Engineering, Project, and Production Management, 2026, 16(5), 2025-253
Time-Series Prediction of Scenic Area Climate Suitability via DTW-LSTM Model and Tourism Development Correlation
Lecturer, Business School, Xuchang University, Xuchang 461000, China, E-mail: xcuxiaoyue1115@163.com
Project Management
Received November 3, 2025; revised December 21, 2025; accepted March 9, 2026
Available online June 17, 2026
Abstract: Traditional methods face two key limitations. First, they cannot effectively capture complex nonlinear fluctuations driven by tourist’s perception patterns in the climate sequence, which undermines predictive accuracy. Second, few studies have systematically and dynamically linked high-precision climate predictions with multi-dimensional tourism development indicators. To this end, the research applies the dynamic time regularization algorithm to capture the climate fluctuation patterns of “similar days” that conform to tourist’s subjective perceptions and constructs a dual-branch Long Short-Term Memory (LSTM) network to improve the modeling accuracy of complex nonlinear dynamics of climate suitability. Moreover, the grey relational analysis method is adopted to conduct correlation analysis. Results indicated that the proposed model demonstrated outstanding performance in climate suitability prediction. Its mean absolute error was 0.92, root mean squared error was 1.35, mean absolute percentage error was 2.84%, and coefficient of determination reached 0.94, significantly outperforming traditional and single models. Correlation analysis indicated that between 2010 and 2020, the correlation coefficient between the comprehensive comfort index and annual visitor numbers steadily increased from 0.68 to 0.83. This demonstrated that comfortable climate conditions have become a key factor influencing tourists' travel decisions. The model proposed by the research enhances the predictive accuracy of climate suitability in scenic areas and reveals a progressively stronger correlation between climate comfort and tourist flow volume. This provides precise data support and decision-making basis for scenic areas to address climate change and optimize visitor flow management.
Keywords: Climate suitability, grey relational analysis; LSTM neural network, tourism demand analysis, time-series forecasting.
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: Liu, X. (2026). Time-Series Prediction of Scenic Area Climate Suitability via DTW-LSTM Model and Tourism Development Correlation. Journal of Engineering, Project, and Production Management, 16(5), 2025-253.
