Researched topics, patterns, barriers and enablers of artificial intelligence implementation in supply chain: a Latent-Dirichlet-allocation-based topic-modelling and expert validation

Abstract

The dynamic and ever-evolving landscape of modern technologies, consumer preferences, and competitive forces pose a perpetual challenge to the adaptive capabilities and resilience of supply chains (SC). In response, enterprises are increasingly considering the integration of artificial intelligence (AI) to facilitate strategic metamorphosis, thereby giving rise to a plethora of AI-based functional applications in the realm of supply chain management (SCM). Despite the potential benefits of AI applications in current SCs, very few cases of their successful implementation can be found in the industry, and research into the driving forces and factors impacting the implementation of AI applications in SCs remains scarce. Accordingly, this study explores the literature to discern emerging researched topics, patterns of AI implementation in SC and understand why the enthusiasm around this implementation does not translate into successful action through an investigation around the barriers and enablers of AI implementation in SC. To answer our research questions, we performed a systematic topic modelling-based inductive content analysis to scrutinise the researched topics and patterns in AI implementation in SC and deductively identify the different categories of barriers to and enablers of AI implementation in SCs. To further refine and validate the findings, a group of experts were consulted using semi-structured interviews, which served to both validate and expand upon the identified categories. Finally, we developed a framework for understanding AI implementation in SCs.