Supply chain leaders prioritise capital but underuse AI for risk

A new survey of supply chain professionals has revealed a gap between supply chain investment priorities and the adoption of artificial intelligence for risk management.

The research, conducted by FourKites in collaboration with ABI Research, encompassed 490 supply chain professionals across manufacturing, retail, and logistics sectors in Germany, the United States, Mexico, and Malaysia. Findings indicate that although nearly a third of supply chain executives consider working capital optimisation as their primary driver for investment, only 37% employ AI specifically for risk management, which is vital for reducing operational disruptions and costly delays.

AI focus lacking in risk prevention

The study found that 28% of supply chain leaders cite working capital optimisation as their top investment goal, significantly surpassing strategic aims like gaining competitive advantage (14.9%) or advancing sustainability (8.4%). Despite this, the majority of companies are focusing their AI applications on demand forecasting or inventory management rather than on preventing disruptions that directly affect working capital.

Mathew Elenjickal, FourKites Founder and Chief Executive Officer, commented on the mismatch between goals and technology application:

"Executives want working capital improvements, yet they deploy AI for demand forecasting instead of disruption prevention. They're analyzing problems instead of preventing them. In contrast, the 27% of organizations willing to use AI for autonomous execution can prevent detention fees before they occur, eliminate expedited freight by managing exceptions proactively, and reduce safety stock by guaranteeing reliable operations. These are direct hits to the balance sheet, delivered through AI that acts, not just analyzes."

The report highlighted regional differences, particularly between Germany and the United States. Only 33% of German respondents reported using AI for risk management, compared to 48% of their American peers. In terms of inventory management, 31% of German firms utilise AI, while 55% of American companies deploy AI in that area. There is also a discrepancy in plans for future AI adoption for autonomous execution, with just 20% of German firms intending to do so versus 31% of U.S. firms.

Key drivers and organisational readiness

Ryan Wiggin, Senior Analyst at ABI Research, discussed the importance of organisational factors and data infrastructure in determining whether AI investments yield the desired outcomes. He said:

"The survey identified key factors that determine whether AI investments achieve strategic goals like working capital optimization. Success requires data interoperability across systems, defined processes for action, and organizational readiness - elements that many companies currently lack."

The findings suggest that, while the need for improved working capital is well recognised, companies often misdirect their AI efforts, focusing on analysis rather than real-time operational action that could prevent cash flow issues associated with disruptions and unplanned logistics costs.

Integration barriers noted

Barriers to AI deployment were also explored. The survey indicates that integration challenges-rather than data quality concerns-are the main obstacles. A total of 46% cited legacy system integration and the problem of fitting new tools into established workflows as the primary hurdles, with data quality more frequently appearing as a secondary concern.

The report identifies a subset of respondents (156 participants) who strongly support autonomous decision-making. These organisations have moved beyond analysis, connecting AI systems to existing infrastructure to prevent costly disruptions, such as detention fees and emergency freight requirements. In contrast, other organisations continue to deliberate about the reliability of AI for roles beyond demand forecasting.

The research concludes that working capital optimisation depends on proactive disruption prevention, and that achieving this requires technological integration and organisational preparedness beyond the simple use of AI for data analytics.