Use of predictive analytics to enable modal shifts 3.3 Predictive demand analytics, powered by machine learning, forecast where and when demand will peak, helping companies pre-emptively allocate resources across more sustainable modes. This level of forecasting enables better bundling of shipments, making full use of available capacity in rail and sea freight. AI could also play a critical role in optimizing logistics hubs and intermodal connections. By streamlining how goods transfer between modes, AI may mitigate the inefficiencies that arise in first-mile and last-mile deliveries – where trucks are often necessary – ensuring that goods move as efficiently as possible through lower-carbon routes (see Box 4). Optimizing modes of transport across Europe to reduce costs and emissions BOX 4 DHL worked with one of its customers in Europe, a large automotive OEM company, to use AI to optimize modal shifts. The OEM, which transports car parts from the Czech Republic and Spain to Germany using large trucks, sought a more sustainable method that would maintain similar lead times while also offering transparency about bottlenecks in the supply chain. DHL devised a multi- modal solution, integrating optimized trucking operations with existing train routes between the Czech Republic and Germany, as well as Spain and Germany. This operational change was implemented within a few weeks. The project led to a 13% cost reduction on the Spain- Germany route and a 4% reduction on the Czech Republic-Germany route. Additionally, the overall carbon emissions per tonne-kilometre per trip decreased by approximately 58%, significantly contributing to the OEM’s sustainability goals. In another case, several European businesses have begun using AI to address the challenge of first-mile and last-mile inefficiencies in rail transport. By integrating AI-driven logistics platforms, they can synchronize transport schedules and reduce idle time, making rail a viable alternative, even for goods that require precise delivery timelines. The European Union (EU) Green Deal recognizes that while 75% of inland freight is carried by road, the region has advanced rail infrastructure, so a significant portion of road freight could shift to rail and inland waterways. Digitalization and AI-powered solutions could support this massive transition.27 As more companies adopt AI-powered solutions, the capacity to make transportation both greener and more efficient will likely continue to grow, driving progress towards global climate targets while supporting business growth. However, despite the high potential of modal shifts to increase decarbonization, this lever is more likely to be driven by increasing demand from customers to reduce scope 3 emissions across the value chain. As large retailers make a greater push towards net-zero emissions, modal shifts will gain momentum across the freight logistics industry. AI is already essential for the efficient routing of shipments across the oceans. We see further potential to minimize CO2 emissions and reduce costs through AI, especially if combined with an even closer integration across all transport parties. Bernhard Hersberger, Head of AI Hub Hamburg, Hapag-Lloyd Intelligent Transport, Greener Future: AI as a Catalyst to Decarbonize Global Logistics 19

Intelligent Transport Greener Future 2025