Key points are not available for this paper at this time.
A persistent challenge in road network optimisation is prioritising new routes under budget constraints. Existing connectivity bottleneck identification methods lack integrated macro–micro analysis. This study proposes a multi-scale connectivity bottleneck identification framework using complex network community detection. It classifies network connectivity bottlenecks as macro-scale Community Connectivity Bottlenecks (CCBs, weakest inter-community links detected via iterative spectral bisection) and micro-scale Traffic Convenience Bottlenecks (TCBs, localised through Origin-Destination (OD) detour degree (ODDD) heatmaps). Applied to Fujian Province's expressway network, the method achieves modularity optimisation in two iterations, reducing the average shortest path length by 15.1%, the network diameter by 13.6%, and the clustering coefficient by 1.1%, while improving traffic convenience in high-ODDD segmented spaces by 5.2% to 9.8%. By integrating multi-scale analysis, this data-driven approach enhances traditional connectivity bottleneck identification frameworks, enabling cost-effective infrastructure planning that maximises connectivity gains.
Xu et al. (Sat,) studied this question.