Optimizing Multidimensional Decision-Making for Sustainable Agricultural Models: Establishing an Evaluation System Based on Dynamic Entropy Weight Method and Monte Carlo Validation

Jianzhang Li

doi:10.63313/JCSFT.9040

Authors

Jianzhang Li School of Mathematics and Physics, Xi’an Jiaotong-Liverpool University, Suzhou, 215123, China Author

DOI:

https://doi.org/10.63313/JCSFT.9040

Keywords:

Computational Sustainability, Multi-Criteria Decision Analysis, Dynamic Entropy Weight Method, Monte Carlo Simulation, Agricultural Informatics, Decision Support Systems, Optimization, Visualization

Abstract

This study introduces a novel computational framework for evaluating sustainable agricultural practices, leveraging advanced decision-making techniques from computer science. Addressing the limitations of existing assessment methods, which often fail to account for dynamic weight allocation and reliability validation, we propose an integrated evaluation system. This system incorporates 17 indicators across six dimensions (pest control, crop health, biodiversity, etc.), utilizing the dynamic entropy weight method (DEWM) for adaptive index weighting and Monte Carlo simulation for robust uncertainty analysis. Our approach features a three-stage computational workflow: (1) generation of a 3D radar evaluation map to visually compare organic, conventional, and integrated farming practices; (2) application of TOPSIS-based decision optimization enhanced with Pareto frontier analysis to identify optimal agricultural modes; and (3) rigorous cross-validation via 10,000-iteration Monte Carlo sampling to ensure model reliability. Empirical results indicate that integrated farming achieves the highest composite scores (0.82±0.03), significantly outperforming conventional methods in biodiversity enhancement (+38.7%) and long-term sustainability (+25.4%). The model's exceptional reliability is confirmed by a cross-validation mean squared error (MSE) less than 1e-24. Moreover, the proposed framework demonstrates superior performance compared to traditional methods, achieving a 24% higher consistency ratio and 37% reduced computational time. This research marks the first implementation of a cross-validated decision optimization framework that synergistically combines TOPSIS, Pareto frontier analysis, and Monte Carlo validation, offering a scientifically rigorous decision support system for sustainable agriculture. Furthermore, the framework's scalability and adaptability pave the way for integration with emerging technologies such as IoT and machine learning, enabling real-time calibration and enhanced decision-making. This work not only advances computational sustainability and agricultural informatics but also provides policymakers with actionable insights for transitioning to more sustainable agricultural practices.

References

[1] S. Saha, S. Samanta, and J. Chattopadhyay, “Dynamics of a crop, pest and predator model in an agricultural system,” Ecological Complexity, vol. 56, 101082, 2023.

[2] A. K. Pal and S. Mandal, “Dynamical analysis of an integrated pest management predator–prey model with weak Allee effect and state feedback impulsive control,” J. Biol. Dyn., vol. 13, no. 1, pp. 531–550, 2019.

[3] S. Mukhopadhyay and S. Bhattacharyya, “Dynamical analysis, optimum control and pattern formation in the biological pest control model,” Nonlinear Dyn., vol. 104, pp. 1373–1391, 2021.

[4] F. Al Basir and J. Chowdhury, “Modeling and analysis of integrated pest control strategies via impulsive differential equations,” Int. J. Biomathematics, vol. 10, no. 6, 1750082, 2017.

[5] H. L. Smith and P. Waltman, “Models for integrated pest control and their biological implication,” Math. Biosci., vol. 229, no. 1, pp. 1–20, 2011.

[6] [6] M. R. H. Mandal and S. S. Gupta, “Stability and control of a plant epidemic model with pesticide application,” Int. J. Adv. Appl. Sci., vol. 11, no. 2, pp. 80–88, 2024.

[7] F. Al Basir, J. Chowdhury, and D. F. M. Torres, “Dynamics of a double-impulsive control model of integrated pest management using perturbation methods and Floquet theory,” arXiv preprint, arXiv:2304.13548, 2023.

[8] X. Cui and Y. Xia, “Modeling the inhibiting effect on a microbial pesticide model,” arXiv preprint, arXiv:2201.09230, 2022.

[9] J. M. Bull, J. R. White, and A. S. Bell, “Effects of biological control on long-term population dynamics and potential for trophic cascades,” J. Appl. Ecol., vol. 50, no. 2, pp. 469–477, 2013.

[10] A. B. Gumel, C. N. Podder, and K. C. Das, “Integrated pest management models and their dynamical behaviour,” Nonlinear Anal. Real World Appl., vol. 4, no. 1, pp. 1–22, 2003.