Predicting supply chain bottlenecks with real-time data: a bibliometric review of transport and logistics research from 2016 to 2025

Abstract Supply chain bottlenecks are critical constraints that disrupt the smooth flow of goods, often triggering delays that cascade through interconnected processes and weaken overall resilience. In an era of digital transformation, the ability to anticipate such constraints through real-time data and predictive analytics has become increasingly important, yet the existing literature remains fragmented across transportation, logistics, and information systems, leaving conceptual boundaries blurred and practical insights underdeveloped. Addressing this gap, we conducted a bibliometric review of 3638 Scopus-indexed publications from 2016 to 2025, using Bibliometrix and VOSviewer to analyse publication patterns, keyword clusters, and thematic evolution. The findings reveal a strong concentration in intelligent transportation systems, with traffic congestion emerging as the most frequent theme, and four clusters identified around traffic management and detection, congestion and travel-time modelling, data infrastructure and analytics, and short-horizon flow control. A clear shift is observed between 2016–2020 and 2021–2025, marked by the growing prominence of machine learning and deep learning approaches. The study contributes by distinguishing transport-level prediction from supply chain-level bottleneck management. It further proposes a research agenda that emphasises cross-context validation, prediction-to-action integration, and explicit tests of forecast usefulness, thereby offering both theoretical clarity and practical guidance for strengthening supply chain performance.