A Hybrid Computational Framework for Regional Logistics Optimization: From Factor Discovery to Strategic Performance Ranking
Abstract
Efficient regional logistics optimization requires computational frameworks capable of handling high-dimensional criteria, uncertain decision environments, and complex interactions among influencing factors. This study proposes a hybrid algorithmic framework integrating Exploratory Factor Analysis (EFA), Structural Equation Modeling (SEM), and SEM-weighted Fuzzy Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) for regional logistics performance evaluation and optimization. The proposed approach combines statistical learning, causal modelling, and fuzzy multi-criteria decision analysis to transform multidimensional stakeholder information into an objective decision-support system. The framework was validated using 670 survey responses collected from logistics stakeholders in Southeast Vietnam. Initially, EFA was employed for dimensionality reduction and latent factor extraction, with sampling adequacy confirmed by a KMO value of 0.9544 and a significant Bartlett’s test (χ² = 33,683.3, p < 0.001). The SEM module validated the measurement and structural relationships, achieving satisfactory model fit (CMIN/DF = 1.82, CFI = 0.948, TLI = 0.941, RMSEA = 0.046, and SRMR = 0.041). The optimized weighting mechanism identified Logistics Infrastructure (CSHT) as the dominant criterion with a contribution weight of 34.7%, followed by Human Resources (NLNV, 21.7%), Science and Technology Capability (KHCN, 21.6%), Regional Linkage (LKV, 15.3%), and State Management and Policy Support (QLNN, 6.8%). Finally, the Fuzzy TOPSIS ranking algorithm successfully differentiated regional logistics performance based on closeness coefficients to ideal solutions. The proposed hybrid algorithm provides an interpretable and robust computational framework for logistics optimization, enabling evidence-based prioritization of infrastructure development, digital transformation, and regional coordination strategies.