Breaking data silos in multi-tier suppliers and designing intelligent collaborative trust

Data silos across multi-tier supply chains create significant barriers to operational efficiency and resilience, where information fragmentation undermines collaborative intelligence and increases disruption vulnerability. This research investigates data silo formation mechanisms and develops an intelligent collaborative trust framework leveraging artificial intelligence to address integration challenges. The study employs mixed-methods analysis across 47 manufacturing organizations selected through stratified purposive sampling from China's industrial regions. A hybrid architecture combining blockchain with federated learning enables secure cross-organizational information exchange while preserving competitive advantages through reputation-based smart contracts and algorithmic trust mechanisms. Network analysis identifies six primary data silo types, with technological barriers most prevalent at 31.4 percent and organizational barriers at 23.8 percent. Randomized controlled trials demonstrate significant performance improvements over conventional approaches. Supply chain visibility increases by 39%, while coordination costs decrease by 28%. The neural network ensemble achieves a 7.3-day average disruption prediction lead time improvement, with pharmaceutical manufacturers experiencing 9.8 days of early warning enhancement. Mean absolute prediction error reduces by 42 percent, and inventory optimization shows 156 percent cost efficiency improvement. This research contributes to supply chain digitalization theory by reconceptualizing trust as an algorithmically-mediated construct, establishing selective transparency frameworks that enable distributed intelligence architectures to achieve.