Smarter, Greener, Connected: The State of the Art in Multimodal Freight Innovation
The freight system in Europe is undergoing significant changes as a result of three major factors: the desire for decarbonisation, digitalisation and the need for seamless interoperability across transport modes. The political framework established by the European Union, which includes the European Green Deal, the Sustainable and Smart Mobility Strategy, the revision of the Trans-European Transport Network (TEN-T) and the ongoing mobile data space architectural development, is gradually but decisively altering what the logistics system is expected to deliver: fully zero-emission transport operations; real-time visibility; increased levels of automation; and sharing trustworthy data between the different stakeholders, across transport modes and borders. Technologies for enabling logistics operations are moving from concept to early implementation and are finding their way into the operations of ports, terminals, inland waterways and urban distribution networks.
HOLOGSTICS Deliverable 1.2 presents a two-part vision: The first part is a complete and thorough analysis of twelve emerging logistics technologies from across the entire multimodal spectrum; the second is a formalised and structured analysis of twenty-two European Union funded initiatives in order to provide lessons learned regarding the future implementation of these technologies.
Twelve innovations and one common thread
The technology scan analyses twelve innovations that collectively span electric fleet optimisation, autonomous inland waterway operations, AI-powered logistics scheduling, rail interoperability, digital data exchange, and IoT-enabled cargo tracking. Despite their diversity, a common technological substrate connects them all: the convergence of artificial intelligence, Internet of Things sensing, digital twins, and federated data-sharing architectures.
Take, for instance, the multi-agent EV shared resource optimisation platform, which uses multi-agent deep reinforcement learning and vehicle-to-grid (V2G) interactions to coordinate electric fleets, charging infrastructure, and distributed renewable energy in real time. Research cited in the analysis shows significant reductions in energy costs and in operational expenditure compared to conventional approaches, compelling outcomes that reflect the maturity of the underlying algorithms. Similar levels of empirical validation support innovations such as the Physical Internet (PI)-hub real-time AI logistics optimisation system and the PI-Logistics synchromodal AI freight bot, both of which harness AI-driven scheduling to reduce transit times and support modal shift decisions with measurable carbon co benefits.
The automated inland waterways zero emission service represents perhaps the most ambitious combination of these technologies: battery-electric or hydrogen-powered vessels, operating under remote supervision and progressively autonomous protocols, integrated into multimodal logistics planning as flexible, schedulable capacity nodes. The technology enablers, such as digital twins, ship shore data connectivity, automated routing, are demonstrably functional. The barriers, as the analysis is careful to note, lie elsewhere: in high capital costs, variable charging and refuelling infrastructure, and a regulatory framework for autonomous waterway operations that is still catching up with what is
technically possible.
Where technology meets its limits
A consistent finding across all twelve innovations is the gap between component-level maturity and systems-level integration. IoT sensors, edge computing, AI optimisation engines, and semantic data models are each well-established in isolation. Bringing them together, meaning converging in ecosystems across different operators, modes, legal jurisdictions, and legacy IT systems, is where complexity multiplies.
Interoperability is the central challenge. Whether the context is the smart box interoperable inlay container (which requires standardised physical and digital interfaces across logistics chains), the DTLF-compliant open standard data exchange platform (which must align with UN/CEFACT and GS1 data models while accommodating fragmented API ecosystems), or the vessel IoT digital twin integrator (which depends on harmonised ship-shore communication standards still under development), the pattern is the same: the technology exists, but the shared governance frameworks, semantic agreements, and regulatory anchors needed to make it work across actors are incomplete.
Cybersecurity also emerges as a cross-cutting concern. API-centric architectures, real-time data sharing, and remote asset management all expand the attack surface in ways that many logistics operators are not yet equipped to address. The analysis identifies this as one of the top deployment risks for the HOLOGISTICS system as a whole.
Learning from what came before
The second major contribution of D1.2 is its structured analysis of twenty-two previous and ongoing EU funded projects, from FENIX and AEOLIX in data-sharing governance, to TWIN-PORT and iTerminals 4.0 in port automation, to H2TowTractor and Couriers Go Green in zero-emission transport. The methodology combined desk research with a partner-driven questionnaire approach, ensuring that the
lessons captured reflect real operational experience.
The synthesis is direct and instructive, the technology readiness, while necessary, is not sufficient for adoption. Projects and initiatives that produced lasting impact shared four characteristics: robust governance models, harmonised KPIs and baseline metrics, regulatory alignment with frameworks such as eFTI and DTLF, and credible business models that survived the end of project funding. Conversely, isolated demonstrations, technically successful but disconnected from operational workflows, procurement cycles, or regulatory preparation, did not scale.
A particularly striking finding concerns the human-AI interface, this is that fully automated solutions have not yet achieved acceptance in the logistics environment, largely for liability and safety reasons. The dominant successful model, as evidenced by projects such as DIMI, SAFARI, and LUKA DT, is human-AI collaboration: artificial intelligence provides optimisation and decision support, while human operators retain oversight and intervention capability. This finding directly shapes the design philosophy of several HOLOGISTICS innovations.
A foundation for what comes next
In combination, the technology scan and lessons-learned analysis of D1.2 provide HOLOGISTICS with a solid foundation to begin work on developing the integration layer; the technologies required for implementation are ready, and the pathways for deploying them have been identified as well as the lessons learned from previous projects. What remains to be done is to construct an integration layer which incorporates shared standards, operational agreements, and regulatory connections necessary for the successful operation of the twelve innovative logistics systems proposed by HOLOGISTICS.
The size of this undertaking is large, but based upon all of the information provided by D1.2 it appears that it can be accomplished successfully.
About the author
Paulo Cantillano-Lizana is a former EU-funded Project Manager at ICOOR and a former research fellow at the University of Modena and Reggio Emilia. Specializing in mobility, logistics, and sustainable transport innovation, he leads and contributes to multiple Horizon Europe projects, bridging research and real-world impact across international consortia.
