Ton Nguyen Trong Hien
Faculty of Industrial Technology, Muban Chombueng Rajabhat University, Ratchaburi, 70150, Thailand.

Noppadol Amdee
Faculty of Industrial Technology, Muban Chombueng Rajabhat University, Ratchaburi, 70150, Thailand.

Adisak Sangsongfa
Faculty of Industrial Technology, Muban Chombueng Rajabhat University, Ratchaburi, 70150, Thailand.

DOI https://doi.org/10.33889/IJMEMS.2026.11.4.064

  ;

Abstract

Evaluating barriers in complex decision contexts requires analyzing causal interdependencies. The integration of the Best-Worst Method (BWM), Decision-Making Trial and Evaluation Laboratory (DEMATEL), and Interpretive Structural Modeling (ISM) provides a logically complete paradigm. Yet its reliability is persistently compromised by subjective or statistically flawed threshold determinations used to construct the ISM hierarchy. To address this gap, this study proposes the SI-BDI framework, which mathematically embeds BWM weights to establish a weighted causal network, and couples Swarm Intelligence (SI) metaheuristics with a newly formulated, parameter-free Causal Structure Index (CSI) that jointly maximizes causal clarity and structural depth. Applied to organic rice farming barriers in Vietnam’s Mekong Delta, three independent SI metaheuristics converge to an identical optimal threshold of 0.156, corresponding to a CSI of 0.852. This optimal threshold yields a four-level hierarchy that exhibits exceptional robustness under both leave-one-out and leave-two-out expert exclusions. Furthermore, as evaluated using the proposed Industrial Adaptability-Structural Stability Reliability (IA-SSR) composite metric, the framework demonstrates superior structural resilience under Monte Carlo weight perturbations of up to 10% compared with conventional mean-plus-standard-deviation and Maximum Mean De-Entropy (MMDE) benchmarks. Practically, the framework reveals that institutional deficiencies, specifically the Inadequate Support Policy Framework as the singular Level-4 root cause, propagate through fragmented zoning and absent enterprise leadership to manifest as weak market control. Crucially, the optimal threshold structurally decouples consumer demand from these upstream policy drivers, providing Multi-Criteria Decision-Making-derived empirical grounding for re-sequencing institutional reforms while concurrently deploying independent demand-side interventions.

Keywords- BWM-DEMATEL-ISM, Causal structure index, Organic rice farming, Swarm intelligence, Threshold optimization.

Citation

Trong Hien, T. N., Amdee, N., & Sangsongfa, A. (2026). A Swarm Intelligence-Driven Decision Support Framework for Adaptive Threshold Optimization in Integrated BWM-DEMATEL-ISM Models. International Journal of Mathematical, Engineering and Management Sciences, 11(4), 1560-1589. https://doi.org/10.33889/IJMEMS.2026.11.4.064.