The findings arrive at a moment when supply chains have moved from the operational back office to the front pages. The pandemic, Brexit-related border friction, the Suez Canal blockage and ongoing geopolitical instability have exposed how fragile modern logistics networks can be, and vendors of AI-powered planning and visibility platforms have been quick to position their products as the remedy. The new research, however, suggests that the relationship between technology investment and supply chain performance is far more conditional than much of the marketing implies, and that organisations frequently underestimate the organisational work required before analytics can pay off.

To reach its conclusions, the team, drawn from the Royal Docks School of Business and Law at the University of East London and the University of Hertfordshire, conducted a large-scale systematic review of the academic literature. They examined more than 500 studies published over a decade, ultimately selecting 145 papers from the period between 2015 and 2026 for detailed analysis. The reviewed research spanned analytics, supply chain visibility, Industry 4.0, resilience, sustainability and governance, giving the authors an unusually broad evidence base from which to assess when and why big data analytics actually creates value in operations and supply chain management.

The central pattern that emerged from the evidence is that technology is a necessary but insufficient ingredient. Analytics platforms can crunch enormous volumes of transactional, sensor and logistics data to detect demand signals, flag supplier risk and shorten reaction times when disruption strikes. Yet the value of those outputs depends entirely on whether the information they produce is trusted, understood and acted upon by human decision-makers. A forecasting model that is statistically impressive but poorly explained, or a visibility dashboard that managers regard with suspicion, contributes almost nothing to operational performance, however advanced its underlying algorithms may be.

Dr Godfried Adaba, Lecturer in Supply Chain Management at the University of East London, summarised the message for business in blunt terms. ‘The message for businesses is that buying the latest AI or data technology is not enough. These tools can help businesses but only if the foundations are already in place,’ he said. He added that organisations need reliable data, people with the right skills, clear responsibility for decisions and managers who understand and trust the information they are being given. ‘If those things are missing, advanced technology can become an expensive system that delivers little real value,’ he warned.

The skills dimension is one of the most persistent constraints identified in the literature. Big data analytics in a supply chain context requires people who can combine statistical and computational competence with deep operational knowledge of procurement, logistics and production planning. Data scientists who do not understand how warehouses, transport networks or supplier relationships actually work can build models that are technically elegant but practically useless, while experienced operations managers who lack analytical literacy may dismiss or misinterpret valid analytical findings. The researchers point to the importance of hybrid teams that bring together technical and operational expertise, allowing analytical outputs to be shaped by practical realities and operational judgement to be informed by evidence.

Data quality and data governance form the second pillar of the argument. Machine learning models and AI systems are only as good as the data they are trained and run on, and supply chain data is notoriously messy, fragmented across enterprise systems, suppliers, carriers and customers, and often recorded to inconsistent standards. Poor master data, gaps in historical records and inconsistent definitions of key measures can silently corrupt even advanced models, producing forecasts and risk scores that look authoritative but are built on sand. Strong data governance, encompassing clear ownership, quality controls, standardised definitions and appropriate access rules, is therefore a precondition for analytics success rather than an optional add-on.

The third element is managerial: the ability to convert information into decisions. The review highlights that analytics creates value through a chain that runs from data to insight to decision to action, and the chain breaks wherever responsibility for decisions is unclear or where organisational incentives discourage acting on the evidence. Companies that treat an AI investment as an IT procurement exercise, delegating it to a technology function without changing decision rights, processes and performance metrics, tend to see sophisticated outputs ignored on the ground. By contrast, organisations that embed analytical findings into routine planning cycles, and whose managers understand and trust the tools, capture measurable improvements in forecasting accuracy, responsiveness and resilience.

On the basis of the evidence, the authors recommend that businesses strengthen their data governance, analytical skills and decision-making processes before expanding their use of advanced analytics. In practical terms, that means auditing data quality across the supply network, investing in training that bridges the technical and operational divide, clarifying who owns which decisions, and building the trust that allows front-line managers to act on what the systems tell them. Only then, the research suggests, can investments in AI, big data, cloud computing and the Internet of Things translate into faster, more resilient and more sustainable supply chains rather than shelf-ware.

The study, titled ‘Big data analytics value creation in operations and supply chain management: a review of capabilities, conditions and constraints’, is published in Benchmarking: An International Journal by Emerald Publishing. For British firms weighing further technology spending against continuing supply chain turbulence, the paper’s core lesson is deliberately unglamorous: artificial intelligence amplifies the quality of an organisation’s data, people and management rather than substituting for them, and firms that skip the foundational work should expect their AI ambitions to underperform the hype.

Subject of Research: The role of big data analytics capabilities in supply chain management and operations

Article Title: Why AI won’t fix Britain’s supply chains

Article References: Why AI won’t fix Britain’s supply chains. (n.d.). Original publication

Image Credits: AI Generated

DOI: Not provided

Keywords: artificial intelligence, big data analytics, supply chains, supply chain resilience, data governance, Industry 4.0, Internet of Things, cloud computing, decision-making, operations management, University of East London, literature review