Planning for equal transit-based accessibility of healthcare facilities: A case study of Shenzhen, China

Abstract

Recent location-allocation studies have made considerable progress in optimizing the equality of facility accessibility but are focused on automobile transport to facilities. In cities, however, the transit-based accessibility of essential services is crucial for social equality and sustainable development. In this study, we develop a modified transit-based maximal accessibility equality (MAE) model for optimizing the equality of the transit-based accessibility of healthcare facilities. In this model, equality is quantified as the weighted mean absolute deviation (WMAD) of accessibility across locations. Two scenarios are set up to reallocate resources or allocate newly added resources. The results reveal that the equality of transit-based healthcare accessibility can be significantly improved in both scenarios. A dispersed planning strategy for facilities is suggested to achieve equal accessibility. However, the transit-based optimization results significantly differ from the car-based optimization results, with more supply allocated to facilities close to transit corridors. This finding implies that the traditional car-based MAE model might generate unequal healthcare accessibility for transit-dependent populations and thus lead to biased recommendations for healthcare planning. Furthermore, it shows that traditional car-based optimization may engender a misallocation of healthcare supply, exacerbating the inequality in healthcare accessibility. The necessity of incorporating public transit into public facility planning is highlighted. The improved MAE model can be applied in cities where the supply of public services is relatively adequate and public transit plays an important role in daily mobility.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 20 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.