Assessing regional accessibility within urban agglomerations: Divergence between opportunity-based and realized accessibility

Regional integration of transport and land-use systems has expanded opportunity access beyond individual cities, increasing the relevance of regional accessibility. Using the Beijing–Tianjin–Hebei urban agglomeration as a case study, we construct a multimodal transportation network covering roads and six types of rail services and assess regional accessibility using two measures. Opportunity-based accessibility represents the extent to which regional opportunities are theoretically reachable from a location, with destination attractiveness subject to impedance-based decay. Realized accessibility represents the integrated utility associated with reaching regional opportunities, inferred from realized destination choices under real-world constraints using a destination choice model calibrated with location-based service data covering more than five million daily intercity trips. Results show a distinct north–south pattern in modal advantage across the study region, with high-speed rail generally more advantageous in the northern part and road transport more dominant in the south. The empirical calibration identifies a peak in intercity trip frequency at a composite travel impedance of around 90 min before entering the decay range. The two measures produce broadly similar county rankings, with approximately three-fifths of the 161 counties showing rank differences within ± 20, although notable spatial divergence remains. Sensitivity analyses under alternative utility specifications further confirm the stability of the realized accessibility rankings and the observed spatial divergence. Urban centers in southern Hebei and counties along the Beijing–Shijiazhuang–Hengshui corridor tend to rank higher under realized accessibility, while outer Tangshan and the Chengde mountain belt tend to rank lower. These differences highlight the sensitivity of regional accessibility assessment to the choice of measurement approach and point to further questions concerning the valuation and substitution of regional opportunities, as well as potential biases arising from suppressed travel demand.