"What could go wrong?". Unlocking AI Potential in Supply Chain Management: Design Principles and Best Practices

The fourth industrial revolution, widely known as Industry 4.0, is profoundly reshaping the industrial landscape through the integration of cutting-edge technologies such as cyber-physical systems, the Internet of Things (IoT), cloud computing, and Artificial Intelligence (AI). Artificial Intelligence is a technology that aims to replicate human abilities in learning, reasoning, and self-correction. In recent years, AI has gained increasing popularity and attracted the attention of researchers and organizations due to advancements in the underlying technology, its growing applicability, and its potential to deliver significant results. The interest in this technology is so substantial that experts predict the AI market will grow tenfold within the next decade as stated in a recent article published on Forbes Advisor (Maheschwari, 2024). Today’s industrial context is characterized by a highly competitive marketplace and increasing complexity. Additionally, the Covid-19 pandemic in 2020 highlighted the weaknesses and vulnerabilities of traditional supply chain and production processes. Together with growing stakeholder demands for resilience, these challenges have accelerated the need for organizations to shift toward technology-driven paradigms. The ability of AI to enhance decision-making, reduce cycle times, and improve operational efficiency positions it as one of the most promising technologies to address operations and supply chain challenges. Despite its transformative potential, AI adoption on an industrial scale remains limited (Armonk, 2024). Currently, the development and implementation of such technologies are primarily the domain of large organizations. Early adopters have explored the technology’s potential and plan to undertake more AI projects, demonstrating its disruptive impact on organizational processes. However, implementing such projects requires significant investment in financial and human resources, as well as a strong commitment to innovation and digital transformation capabilities. Many organizations attempting to adopt AI encounter significant challenges during implementation, leading to project failures and to limited results compared to expectations (Merhi, 2023). This creates a paradox: while AI can offer substantial benefits, its successful integration requires overcoming numerous obstacles, which often became clear and evident only when implementation fails. For this reason, some researchers have begun investigating the reasons behind AI project failures and identifying 68 critical success factors that practitioners should consider during the implementation process. (Merhi, 2021, 2023; Meyer & Henke, 2023). Although some studies have tackled the topic of AI implementation effectiveness, there are still limited contributions that provide comprehensive and empirically tested guidelines to support managers and companies during this complex process. This research seeks to bridge this gap by developing a generalized framework of design principles to guide practitioners in AI system implementation in the field of operations and supply chain management. The framework is derived based on a mixed method approach, combining a systematic literature review, which identifies the current state of knowledge on factors and guidelines for AI implementation, and the development of case studies to evaluate its potential effectiveness. In particular, the systematic literature review identified 15 works that allowed to develop a set of 10 design principles aimed at supporting practitioners in deploying AI in supply chain processes. The framework stands out from previous works, and it not only provides practitioners with critical factors to consider, but also suggests approaches and activities, highlighting their importance, that organizations should adopt to enhance the likelihood of a successful implementation. To validate the framework, an empirical investigation has been conducted. Specifically, three case studies have been developed in large industrial organizations. In particular, the case studies investigated implementation projects of AI solutions to support supply chain management processes, analyzing in detail the processes followed during these projects. The findings from the case studies have been compared to the proposed framework to assess its applicability and establish a hierarchy of importance for the identified design principles. In particular, we identified some implementation principles that represent necessary conditions for a successful implementation project, and others that contribute to making the implementation more effective and can reduce the duration of the implementation project. Based on this analysis, a final framework for AI implementation is proposed, along with guidelines for its use in practical applications. In particular, the identified design principles refer to the following areas: • Employees management • Stakeholders’ management • Top management • Data and infrastructure • AI knowledge • Post deployment • Financial capability • Governance and communication • Organization’s strategy • Organization’s structure For each principle a set of proper practices have been developed, based on both the systematic literature review and the analyzed case studies, thus identifying 32 best practices. 69 The contribution of the work is twofold. First of all, we summarize the current knowledge on AI implementation in SCM in a structured framework that provide a reference point to any work willing to investigate this topic in further detail. Second, we provide a reference guideline for practitioners on how to tackle AI implementation, supporting innovation and anticipating relevant constraints to the first steps of the development projects.