The development of AI has made it possible to envision “thinking robots” that are capable of learning and taking over human roles. Since the late 1970s, AI has shown considerable promise in improving human decision-making processes and, by extension, productivity across a wide range of business endeavours, thanks to its ability to recognise business patterns, understand business phenomena, seek information, and intelligently analyse data. While artificial intelligence has many practical applications, it is seldom applied in supply chain management (SCM). In order to realise AI's numerous potential benefits, this chapter explores the different areas of AI most suited to tackling real-world SCM difficulties. This chapter of the book achieves precisely that, analysing the history of AI's use in SCM applications and identifying its most potential future uses.
TopIntroduction
For a business to achieve supply chain (SC) excellence in the face of growing demand uncertainty, supply risk, and competitive intensity, it must be able to integrate and orchestrate the full gamut of these end-to-end processes, from sourcing raw materials to delivering finished products to customers (Madhavi, 2021). Several forward-thinking businesses have sought to boost their capacities by enhancing their information sources and exchanging real-time data with their SC partners(Palanivelu & Vasanthi, 2020).Because of this, SCM is beginning to place less value on its physical assets (such as stockpiles, warehouses, and delivery vehicles) and more value on its digital assets (Khan, B. et al., 2021). As the relevance of information to SC success has grown, practitioners in the field have explored many options for improving information management and using this knowledge to better guide business decisions(Jain, 2019).Artificial intelligence (AI) (Pandey, D. et al., 2021a), albeit not new, is one strategy that has the potential to revolutionise the supply chain management industry(Geisel, 2018).“To solve problems in decision-making situations where optimal or exact solutions are either too expensive or difficult to produce, AI (Anand, R. et al., 2023) refers to the use of computers for reasoning, recognising patterns, learning (Saxena, Arpit. et al., 2024) or understanding certain behaviours from experience, acquiring and retaining knowledge, and developing various forms of inference (Pandey, B. K. et al., 2021a). Artificial intelligence (AI) research and development aims, in the broadest sense, to understand what intelligence is and to build computer systems that can mimic human behavioural patterns and produce knowledge that may be used to solve problems in the same way that humans do(Sadiku, Fagbohungbe, & Musa, 2020). Therefore, AI should be able to learn independently, pick up new concepts, reason, draw conclusions, impute meaning, interpret symbols in context, etc.Due to this, AI should be able to learn independently, pick up and understand new concepts, draw their own conclusions, infer meaning, interpret symbols in context, and do so much more on their own(Soni, Sharma, Singh, & Kapoor, 2019).
Understanding complex, linked decision-making processes and developing intelligent knowledge bases important to collaborative problem-solving are just scratching the surface of AI's potential applications, but they are essential to the expanding management philosophy of SCM(Roundy, 2022).For example, Eastman Kodak created a rule-based expert system to incorporate the mental maps of experienced order pickers in order to choose orders from a warehouse in the most expedient way possible. To coordinate the sequential but distinct steps of SC's joint demand planning and forecasting processes, Han et al., (2021) proposed an agent-based forecasting system that can predict end-customer demand (Saxena, A. et al., 2021) via information exchange among multiple SC partners and learn from the past forecasting experience(Bogachov, Kwilinski, Miethlich, Bartosova, & Gurnak, 2020). Expert systems and agent-based systems are two types of artificial intelligence (AI) (Boopathi, S. et al., 2023) that have been shown in these examples to be useful in controlling various parts of the SC (such storage, and stock management).
In light of this example, the goals of this study are to: