Editorial: Digital transformation in transport and logistics

The digital transformation of transport and logistics has moved from a strategic aspiration to an operational necessity. The intellectual origins of this shift trace back to the Industry 4.0 initiative of the early 2010s, when the integration of cyber-physical systems placed digital-physical convergence at the center of industrial strategy (Kagermann, 2013; Sullivan and Kern, 2021). For much of that decade, however, the practical distance between vision and reality remained wide.What narrowed this gap was the rapid development of enabling technologies. Internet of Things (IoT) devices, cloud computing, machine learning and autonomous systems such as Automated Guided Vehicles (AGVs) and warehouse robots moved from experimental settings into live logistics operations (Fottner et al., 2021; Fragapane et al., 2021; Escherle et al., 2023; Lam et al., 2024). A growing body of research has documented the resulting improvements in visibility, asset utilization and decision-making capability across the value chain. Industry 4.0 has positioned digital technologies as central enablers of this transformation (Barreto et al., 2017). IoT devices now provide real-time visibility into the movement and condition of goods across extended transport chains (Barreto et al., 2017; Feng and Ye, 2021). Artificial intelligence (AI) and machine learning algorithms support demand forecasting, route optimization and predictive maintenance (Hoffmann et al., 2022; Nitsche et al., 2024). Cloud computing and digital platforms facilitate the integration of fragmented logistics ecosystems by connecting shippers, carriers, terminal operators and end customers through shared data environments (Winkelhaus and Grosse, 2020).The 2020s then introduced a different kind of pressure. The COVID-19 pandemic and subsequent geopolitical disruptions revealed the fragility of efficiency-optimized but resilience-deficient logistics networks. Contact restrictions, acute labor shortages and simultaneous demand shocks forced many organizations to compress years of planned digitalization into months (Ivanov, 2021; Hellweg et al., 2023). The resulting recalibration was structural. End-to-end supply chain visibility and data-driven decision-making became operational necessities rather than competitive advantages, and the need to collect, integrate and act on operational data intensified across the industry.The consequence is a rapidly growing number of interfaces among physical assets, digital tools and humans throughout the transport and logistics value chain. Figure 1 illustrates them as a three-level architecture: at the physical level, physical assets (e.g. vehicles and facilities), field and IoT devices generate ongoing data about what is happening in the logistics system; at the digital level, these data are ingested, stored, integrated and exchanged across systems through application programming interfaces (APIs) and related system connections, where they support event management, interoperability and consistent digital representations of operations; these representations provide the basis for data-driven analytics and decision-making at the analytical level, such as diagnostic assessments of current system states, predictive analytics for future developments, reality-based scenario analysis and higher-level decision support. The analytical outputs, in turn, move downward to shape physical behavior through accurate and reliable control signals (Albrecht et al., 2024). This bidirectional architecture has considerable productive potential. However, realizing it requires ensuring that digital systems can accurately capture the complexity of the physical world and that analytical outputs can be effectively translated into operational decisions.This special issue of Logistics Research was designed to address these dual challenges. It invited analytical, conceptual, empirical, experimental and optimization contributions that advance the understanding of digital and technological interfaces in transport and logistics at both the technological and the organizational-planning levels. In particular, the integration of digital technologies into physical assets and processes, as well as the application of data-driven analytical methods for supply chain decision-making, were identified as areas of primary interest. The two contributions of this special issue, Wehner (2026) and Doege (2026), address two essential and complementary dimensions of this transformation though different in scope and methodology: the physical implementation of autonomous technologies and the analytical use of network data for resilience.Digital interfaces in transport and logistics differ fundamentally in the nature of the problems they address and the level at which those problems arise. To position this heterogeneity, we propose a two-dimensional conceptual framework that distinguishes between the nature of the digital interface, depending on whether digital technologies primarily sense, translate and control physical assets (physical-digital interface) or act on digital representations of them (analytical-digital interface). This framework also distinguishes between the levels of analysis at which challenges and decisions arise for specific operations and technologies or across network-level coordination. These two dimensions are analytically necessary because operational and network-level challenges differ in scope, time horizon and coordination requirements, while the two interface types raise distinct technological and organizational issues. Table 1 maps this space through selected examples. The two contributions in this special issue occupy distinct positions within this space, as discussed in Section 2.3.This framework reflects observations drawn from a broader community, including discussions within the production, logistics and transportation research community, including exchanges at the 7th Interdisciplinary Conference on Production, Logistics and Traffic (ICPLT), held in Darmstadt in March 2025. These discussions covered the full spectrum from physical automation to network-level analytics across rail freight, road transport, maritime logistics, urban delivery and terminal operations.Physical-digital interfaces refer to the points at which physical assets (e.g. robots, vehicles, containers and terminal equipment) become readable by digital systems and at which digital instructions are translated into physical behavior. Lee et al. (2015) proposed a five-layer architecture for cyber-physical systems that precisely captures this bidirectional flow from physical sensing to digital action, spanning connection, conversion, cyber, cognition and configuration. At both the operational and the network levels, the challenges are primarily technological. For example, how can an autonomous system reliably perceive and act in a dynamic, unstructured environment, and how can the resulting information be safely shared and made interoperable across various systems (Kembro et al., 2017; Fottner et al., 2021)? Beyond the technical dimension, physical-digital interfaces raise equally significant organizational questions. The introduction of autonomous systems into existing workflows requires process re-engineering, personnel retraining and often fundamental rethinking of task allocation between humans and machines (Cichosz et al., 2020; Hellweg et al., 2023).Analytical-digital interfaces are the boundaries between digital representations of the real world and the computational methods used to extract useful information from them. The central challenge here is epistemological: What data and metrics are sufficient for effective decision-making, and how much network visibility is actually needed? The growing availability of various data sources has considerably expanded the scope of these interfaces. Yet data availability alone does not guarantee usefulness. Supply chain networks are characterized by information asymmetries, data quality issues and coordination failures, and autonomous entities typically make decisions based on local and incomplete information, so that what any single actor can reliably infer and influence remains limited (Delfmann et al., 2018; Fottner et al., 2021). Therefore, the practical value of digital analysis at this interface depends on whether data-driven decision rules and performance indicators can be translated into operationally feasible actions in specific organizational and network contexts.The two contributions in this special issue occupy complementary positions within the framework proposed above. Together, they illustrate how digital transformation in transport and logistics unfolds through two distinct but interconnected modes: the physical implementation of autonomous technologies and the analytical use of network-level data for resilience-oriented decision-making.Wehner (2026) are positioned at the physical-digital interface on the operational level. Their empirical study investigates the integration of automation and digital technologies in air cargo logistics through the empirical testing of the O3dyn pallet transport robot at Munich Airport. This study documents a ten-day real-world trial in which the robot's performance was assessed across multiple transport scenarios, with data collected on travel times, load weights, manual interventions and operational challenges.Doege (2026) are positioned at the analytical-digital interface on the network level, addressing one concrete analytical question: How can supply chain networks be made more resilient against disruption propagation, and how much network visibility is actually required for this purpose? Using agent-based modeling combined with a susceptible-infected-recovered epidemiological framework, the authors simulate disruption cascades across ten real-world supply chain networks and evaluate five node-selection strategies derived from social network analysis metrics.Digital transformation in transport and logistics has moved past the stage of technology demonstration into a period that requires empirically grounded and organizationally sensitive research. The two-dimensional framework proposed here offers a means of positioning and connecting contributions across this field. The papers in this special issue represent two reference points in that space. Viewed through this framework, their respective scopes point directly toward the most consequential open research directions. Yet the scope of the present issue necessarily remains selective. Some important dimensions are not addressed directly.The interfaces among the physical, digital and analytical levels are not independent. Digital twins represent the most developed form of this interdependence, functioning simultaneously as physical-digital synchronization mechanisms and as analytical simulation environments (Tao et al., 2018; Fottner et al., 2021). While a growing body of research has begun to explore their potential in logistics and supply chain management, this dimension falls outside the scope of the present issue and warrants continued attention, particularly in multi-actor and cross-modal settings where the integration of operational data streams with network-level decision support remains technically and organizationally demanding.Furthermore, the rapid emergence of generative AI and large language models represents the most visible frontier. Applications such as natural language interfaces for freight procurement, generative scenario planning for supply chain risk and AI-assisted coordination across carrier networks demonstrate the remarkable potential of these disruptive technologies in logistics decision-making. While these applications are situated at the analytical-digital interface, they also introduce new challenges regarding model interpretability and human oversight. As the submission timelines preceded this period, these developments are not addressed in this special issue. However, they do deserve more focused attention in future research.The contributions in this special issue encourage further inquiry at the intersection of physical automation and network intelligence, advancing both the technological and the organizational foundations of digital transformation in transport and logistics.During the preparation of this work, the authors used Grammarly and Claude in the writing process to improve the readability and language of the manuscript. After using this tool/service, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.The Guest Editors wish to thank all authors who responded to the call for papers, the anonymous reviewers whose expertise ensured the quality of the review process and the Editor-in-Chief of Logistics Research for the opportunity to organize this special issue. They also wish to acknowledge the organizers and participants of the 7th Interdisciplinary Conference on Production, Logistics and Traffic, whose presentations and discussions on digital transformation in transport and logistics provided a broader intellectual context for this special issue.