Linking Poverty-Based Inequalities with Transportation and Accessibility Using Mobility Data: A Case Study of Greater Maputo

Abstract

Accessibility is widely considered as the most crucial benefit of any transportation system. Low accessibility may cause compromise on living conditions, low economic growth, high unemployment, social seclusion, and long-term social inequalities. In developing countries of Sub-Saharan Africa, surveys fail to keep up with the pace of rapid urbanization. Additionally, numerous location-based data sets, including mobile phone location data and Google Maps travel time, enable near-real-time observations of actively changing mobility dynamics. In this study, we use these novel data sets to assess various facets of accessibility and corresponding poverty-based inequalities within the Greater Maputo region in Mozambique. A data-based approach is seldom used in the Sub-Saharan African context. Consistent poverty-based inequalities in access to opportunities by driving, transit, and walking are observed. The transit system was ascertained to be more inefficient in the poorer regions. The richer regions witnessed most decreased access to opportunities because of high congestion. People living in the poorest regions were observed to travel much longer distances to access facilities than did those living in the richest regions. These differences were particularly significant for essential services of education, healthcare, and employment. These results can help urban planners and policymakers identify disadvantaged communities and develop policies to ameliorate the conditions. This assists in measuring the progress of Sustainable Development Goals (SDGs) and investigating areas of concern. Moreover, this study shows the applicability of various data sets and methods in a data-deprived scenario in Africa, and motivates more scholars and practitioners to partake in this technology leapfrogging.

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.