Estimating accessibility using a Markov chain model accounting for traffic on the South African road network

Abstract

Accessibility modelling has been investigated since the early 1960s. Accessibility is defined as the ease with which people can access destinations when using a particular mode of transport. It is largely important in urban planning, freight business, policy-making and sustainable development as it forms part of the SDG goals of rural accessibility. This research models accessibility in and between administrative units using the South African road network accounting for real life traffic data. Hence, this research investigates whether traffic volume has a direct impact on accessibility. This will be done through extending an existing Markov chain accessibility model by incorporating traffic data. The addition of traffic data can assist us in addressing several issues related to the South African road network, such as routes ambulances could take to avoid treacherous traffic conditions. The different available South African traffic data sources have been extensively investigated. This study indicates that traffic may have an impact on accessibility. This is displayed by the difference in transition probabilities between the current model and the total travel time model. Accessibility studies can truly add value in different aspects as it can be used as a factor in decision making that affect road networks in South Africa. We have provided a novel extension of an accessibility model.