Dubai Airports implements AI-powered inventory forecasting

Dubai Airports (DA) has implemented a demand planning and forecasting initiative that is transforming behind-the-scenes supply management operations.

Image: Dubai Airports Company

The approach has enhanced service levels, streamlined inventory management, and boosted operational efficiency across the organisation, contributing to a superior experience for airport guests.

Over the past year, the introduction of a sophisticated Material Requirement Planning (MRP) tool, powered by artificial intelligence and machine learning, helps the inventory team predict future spare parts consumption. With parts on hand, the team can promptly execute engineering work orders, resulting in quicker maintenance responses and fewer disruptions for airport guests.

The MRP tool has improved forecast accuracy by 30%, allowing for precise inventory planning and reducing excess stock by 12%. This optimisation enhances capital utilisation and ensures that resources are available when needed.

Emmanuel Augustin, VP of Supply Management at Dubai Airports said: "We leveraged AI and real-time data to streamline our inventory management, improve efficiency and enhance the quality of service we provide to our key customers, both our internal business functions and the millions of guests who travel through the airport. As the first airport in the region and the first government entity in Dubai to implement this system, it underscores our commitment to innovation and operational excellence.”

Improved forecasting has led to a 24% increase in service levels, enabling faster response times for facility maintenance and enhancing the overall guest experience. The reduction of ageing work orders by 82% ensures that maintenance tasks are addressed promptly, minimising downtime and maintaining high standards of airport operations.

The automation of the inventory ordering process has improved efficiency by 400%, integrating the creation of purchase requisitions and significantly reducing manual workloads.