Current methods of transport planning, in particular travel demand models, are inherently dependent on accurate representations of the land-use – transport system. Their established methods have been developing since the 1950s, exclusively in the global north, and require detailed descriptions of the transport system, including public transport infrastructure, routes and schedules. These methods have been implemented in most proprietary, as well as opensource, software and codes that transport professionals use globally as well as in Africa.
Two issues stand out with these established methods. First, these software tools are designed to analyse the functioning of transport systems, resulting in a focus on road congestion and solutions to ease congestion. This focus deepens transport-related social exclusion, especially in Africa where the majority population is dependent on walking, cycling and (informal) public transportation to access their activities, modes not (well) represented in these models. Second, most cities in Africa are home to informal public transport systems, which are typically not represented in these models at all. Because of their informal nature, they are difficult to model, as the models require fixed routes, stops and frequencies. Informal public transport systems are all but that. Some modellers have responded to this problem by using GPS surveys to capture a representative sample of paratransit routes and translate that into a ‘hard-coded’ public transport lines for further analysis. This research proposes to investigate alternative, low-cost, methods in appropriately modelling informal public transport systems and its associated data collection methods. We will develop and test a data collection protocol for paratransit and produce an open-source code to allow representation of flexible paratransit next to formal, scheduled, public transport services. We will implement these methods in the toolkit for transport planning based on principles of justice as developed in a previous VREF project by the same core team members.