كنفرانس بين المللی مهندسی صنايع و سيستم ها

صفحه اصلی / دوازدهمین كنفرانس بين المللی مهندسی صنايع و سيستم ها

A Data-Driven Framework for Agile-Resilient Pharmaceutical Supplier Selection in Industry 4.0

نویسندگان :

Mohammad Rahmani¹ Reza Tavakkoli-Moghaddam²

1- دانشگاه تهران 2- دانشگاه تهران

کلمات کلیدی :

Supplier selection،pharmaceutical supply chain،agility،resilience،Industry 4.0،fuzzy inference،machine learning

چکیده :

Abstract-Pharmaceutical supplier selection should jointly reflect disruption recovery, operational responsiveness, and digital readiness. This paper develops an integrated decision framework in which Step-wise Weight Assessment Ratio Analysis (SWARA) determines the importance of 15 criteria, a weighted fuzzy inference system (WFIS) assigns the interpretable labels Poor, Average, and Good, and machine-learning classifiers learn the WFIS decision surface for rapid reuse. The SWARA assessments obtained from three expert panels give construct weights of 0.3542 for resilience, 0.3636 for agility, and 0.2823 for Industry 4.0. The empirical dataset comprised 450 normalized supplier profiles. The WFIS generated 97 Poor, 258 Average, and 95 Good cases. On a stratified 25% test set, k-nearest neighbors achieved 92.0% accuracy and a macro-F1 of 0.917, while the radial-basis support vector machine obtained the highest five-fold cross-validation macro-F1 of 0.932. Perturbing all SWARA weights by ±10% retained, on average, 99.0% of labels. The framework is transparent and updateable and demonstrates its potential for operational supplier screening, while validation on additional pharmaceutical procurement datasets is recommended before broader deployment.