Artificial Intelligence in Collaborative Data Sharing for Intra-Company Logistics
embedded in broader multi-partner supply-chain networks. Decisions slow down, small delays compound into lost shifts, and opportunities to prevent breakdowns or cut idle time slip away. We distinguish raw multi-source data streams— discrete, unprocessed observations that lack inherent meaning or context— from the information that AI-supported analytics derive for decision support( data that has been organized, contextualized and endowed with meaning, relevance and purpose) [ 1 ].
The ICL Testbed grew directly out of this practical challenge. Over 41 months an interdisciplinary team from three industrial companies ran 55 workshops to design, build, and operate a cooperative Track and Trace system for intralogistics. An industrial service provider served as the main integrator and hosted the central platform. A sensor technology manufacturer delivered the IoT hardware, the connectivity layer, and the digital twin functions. A heavy equipment manufacturer contributed operational expertise on how wheel excavators actually move through yards and provided the live test sites. Instead of replacing existing tools, the partners linked their current products and data sources so that information could flow across company boundaries without forcing anyone to abandon what already worked [ 2, 3 ]. This multi-stakeholder orchestration illustrates a broader design principle: ecosystem value emerges from connecting existing assets under shared governance rather than from consolidating them onto a single platform. Today the testbed runs live with 15 vehicles and assets exchanging data across the three organizations in real time. Every partner sees the same up-to-date picture of asset location, status, and movement history. The project also clarified the real-world requirements for data quality, governance rules that all sides accept, and a structure that lets new asset types or new partners join later without major rework.
The ICL Testbed thus furnishes a uniquely authentic intra-company, multi-partner data-sharing architecture featuring comprehensive Track and Trace functionalities, high-granularity sensor data streams, and a robust substrate for AI-enhanced analytical applications. At its core lies a Track and Trace system that connects 15 vehicles and assets across organizational boundaries through real-time IoT-enabled data exchange [ 2, 3 ]. This architecture crystallizes tangible challenges and empirically validated solutions concerning( inter-) organizational integration, data provenance and quality management, governance design, and the cultivation of scalable ecosystem value; precisely the issues that animate the three AI value-creation dimensions central to this study. These dimensions encompass intelligent data provisioning and sensor data preparation to achieve profound operational integration across partner systems; AI-supported analytics that enables decision support and propel continuous process optimization; and the horizontal expansion of AI affordances to encompass adjacent assets, complementary data sources, and innovative business scenarios – a dynamic herein termed the horizontal expansion logic(“ One Step Further”) and examined in depth in subsequent sections.
By interrogating these dimensions through the lens of the ICL Testbed, the research addresses the guiding question of how AI can effectively scale and amplify the value generated by collaborative data-sharing architectures within logistics contexts. The overarching objective is to distill practice-oriented, actionable insights and recommendations tailored to industry practitioners. In advancing this agenda, the contribution explicitly engages the reimagined supply chain paradigm by demonstrating pathways toward smart operations predicated on AI-driven
128