THR-DMPOM model for reliable high-precision evaluation to enhance real-time performance management and optimize...

Area of Science:

• * Management Science
• * Artificial Intelligence
• * Tourism Management

Background:

• * Human Resource Performance Evaluation (HRPE) in tourism faces challenges with data volatility and local optimality.
• * Existing methods struggle with dynamic, multidimensional performance data and nonlinear interdependencies.
• * Need for a robust model to handle real-time performance fluctuations and optimize resource allocation.

Purpose of the Study:

• * To propose the Tourism Human Resource-Deep Momentum Performance Optimization Model (THR-DMPOM) for HRPE in the tourism sector.
• * To address limitations of existing models in dynamic and complex evaluation environments.
• * To enhance the stability, accuracy, and adaptability of performance evaluation systems.

Main Methods:

• * Development of a 24-item evaluation index system based on literature review and industry expert interviews.
• * Implementation of recursive momentum accumulation, cross-layer weight smoothing, and adaptive gradient correction mechanisms.
• * Utilizing multilayer networks for continuous dynamic weight updating and historical gradient propagation.

Main Results:

• * THR-DMPOM demonstrated high stability and accuracy in experiments with tourism enterprise datasets, with systemic deviation within ±1.0.
• * Distribution consistency validation showed strong agreement between model scores and historical records (D-values near 1, p-values > 0.7).
• * Outperformed FAHP, F-TOPSIS, traditional AHP, TOPSIS, and static machine learning methods in accuracy, efficiency, and adaptive weight learning.

Conclusions:

• * THR-DMPOM provides a robust and high-precision framework for real-time performance management in dynamic tourism environments.
• * The model effectively mitigates local optima and weight oscillation issues through adaptive weight learning and historical information sharing.
• * Significant improvements in handling complex, dynamic performance data and optimizing resource allocation within tourism enterprises.