AI in industry
AI might be transforming industries, but its ability to drive accurate workflows relies on a foundation of reliable data. For those working with supply chains, this data can generate assessments of global circumstances and highlight upcoming disruption to operations before it’s felt by the consumer.
In the past year, extreme weather, trade disputes, and geopolitics have tested the limits of business preparedness. For example, in October 2024, it was estimated that the storms that hit Valencia caused damage to its farming industry worth almost £1bn. That includes the produce lost and the rendering of underlying infrastructure as unusable. As the impact of the climate crisis drives an increase in natural disasters, supply chains must prepare for widespread disruption.
Looking to 2026 and beyond, this trend is unlikely to change for the better. To best future-proof business processes, AI will be fundamental. But where should organisations start?
Which data is good enough?
High-quality, accurate data is important for driving AI success in supply chains and providing users with accurate predictions. This enthusiasm is reflected in the expectation that the big data market will be worth over £300 billion by 2028. Despite this significant investment, most organisations, surveyed across industries, still face data-quality issues.
At present, only 12% of data and analytics professionals believe that their company’s data is ready for AI adoption despite 76% recognising data-driven decision-making as a priority. To drive success in supply chains, this lack of readiness needs to change.
Data preparation
Though action must be taken to remedy these concerns, companies shouldn’t view the quality of their own data as a blocker to innovation. Instead, they can ‘test’ the data before using it to drive insights.
As a first step, it’s essential to identify the format and quality of existing data assets. With complete knowledge of all the information available, corporations can integrate AI tools that work with their data, instead of trying to fit it into incompatible solutions.
Next, team leaders must be certain that their employees are trained on noticing hallucinations and changing processes to ensure accurate AI forecasting. Creation of the right procedures will feed into a successful long-term data governance strategy, ensuring full value is extracted by AI tools.
For ongoing insights, directly reflecting global circumstances, data must be continually fed into AI systems. By setting up the extraction of data from a reliable platform, companies can ensure that the insights they receive directly correspond with the most pressing logistical concerns.
Incompatible sources
Strategic partnerships can bring essential expertise for agile transformation, helping companies to scale at speed and improve their assessment of risks. For instance, by integrating data from a partner organisation, visibility across the global logistics landscape will be increased. Concerns arise, however, when data is formatted differently at each company. To mitigate the chance of hallucinations, data-trained workers should be proactively advised to scan insights for duplicates, misspellings, and inaccurate information.
Visibility
For operational success amid an ever-changing global landscape, the importance of preparing and ‘cleaning’, data should not be understated. To ensure accurate insights are produced by AI tools, integrated solutions should be compatible with current data-formatting, proactively mitigating the chance of hallucinations. To derive full value, the same ‘cleaning’ procedure should be used for partner data. By taking the right steps at the beginning of the adoption journey, business leaders can drive effective insights, consistently being updated, to support future growth.
- AI in Supply Chain