Beyond Import Scale: An Entropy-TOPSIS Framework for Prioritizing Import-Related Supply-Chain Vulnerability in Bangladesh Manufacturing

Abstract This study proposes and implements a multi-criteria decision-making framework to evaluate and prioritize import-related supply chain vulnerabilities among eight strategically important manufacturing input categories in Bangladesh. The analysis utilizes a source-verified panel dataset consisting of 40 monthly observations for each category from January 2023 to April 2026, including HS27 Mineral Fuels & Oils, HS29 Organic Chemicals, HS39 Plastics & Articles, HS40 Rubber & Articles, HS52 Cotton, HS72 Iron & Steel, HS84 Machinery & Parts, and HS85 Electrical Machinery.This eight group are significant. Vulnerability is assessed using five dimensions: Exposure, Volatility, Growth pressure, Shock intensity, and Persistence (EVGSP). Objective criterion weights are determined through the Entropy method, while the TOPSIS approach is applied to generate a comparative vulnerability ranking. The results show that Shock intensity has the highest contribution (0.3074), followed by Volatility (0.2887), Exposure (0.1606), Persistence (0.1362), and Growth pressure (0.1071). HS27 Mineral Fuels & Oils emerges as the most vulnerable category with the highest TOPSIS closeness coefficient (0.9724), followed by HS84 Machinery & Parts (0.3336) and HS72 Iron & Steel (0.2914), whereas HS39 Plastics & Articles demonstrates the lowest vulnerability level (0.1307). Statistical validation by one-way ANOVA confirms significant differences in average monthly import values among the selected categories (F(7,312)=212.67, p<0.001, η²=0.827). Furthermore, an exploratory regression analysis reveals that Exposure and Shock collectively explain 95.2% of the variation in TOPSIS scores; however, the limited number of alternatives restricts causal interpretation. The Sensitivity analysis using equal weighting, CRITIC weighting, and criterion elimination demonstrates strong consistency with the original ranking, with CRITIC based results achieving a Spearman correlation coefficient of 0.929. The findings demonstrate that import dependency alone is insufficient for vulnerability assessment, as instability and sudden shocks significantly influence supply chain risk prioritization. The proposed framework provides a systematic early-warning mechanism to support procurement planning, inventory management, and strategic import monitoring while avoiding unsupported conclusions regarding actual shortages or production disruptions.