Artificial Intelligence model for social sustainability risk management in the apparel supply chain

Abstract

The social dimension of sustainability has received special attention from researchers and managers in recent years. However, in supply chain management, many unsolved problems prevent reaching a satisfactory level of sustainability. The purpose of this study is to use an Artificial Intelligence (AI) model, to identify risk probability in social dimension of sustainability when contracting suppliers in the apparel supply chain (SC). A Bayesian (belief) Network (BN) model is applied based on supply chain information of six global companies from developed countries and five focal companies from Brazil, a developing country. The originality of this work is the Bayesian Network model, which uses the Human Development Index (HDI) and Global Slavery Index (GSI) to determine the risk probability of non-compliance in social aspects in the SC, fostering the monitoring and control of social burdens in the SC. The interrelation between companies and different external source data can provide important information to help managers and practitioners for decision making to mitigate risk and improve transparency along the chain. The model has been applied to eleven companies of the apparel sector and the main results are the identification of the social risk probability levels. From these results, it is possible to identify where the main fragilities are, being an important tool for practitioners and decision-makers to better manage the social dimension of sustainability in the supply chain.