In South Africa, Random Utility Maximisation (RUM) econometric theory based on the weighted linear addition of attribute vectors has been used in commuter trip choice models since the early 2000?s. The application of this theory assumes rational choice behaviour by the user, and requires the identification and quantification of the trip choice attributes and their relative weightings, as well as the specification of the random (error) component of the utility equation. One of the important derivatives of this approach is the value of travel time savings (VTTS), i.e. the monetary value a commuter is willing-to-pay to save a specific amount of travel time. However, VTTS and willingness-to-pay are still the subject of extensive research, controversy and debate internationally. In South Africa several attempts have been made to quantify the VTTS for various commuter modes since 2000, and this paper reviews the evidence of VTTS obtained from these studies. The large variation in values reflects a diversity of projects, modes, locations, and the use of a range of estimation methodologies. However not all the variation is explicable by these factors, raising questions about the appropriateness of using such values for demand forecasting and economic benefit estimation. This paper identifies promising advances in choice model specification, relating to the incorporation of commuter travel budgets (time and cost) and unobserved heterogeneity which might help to improve VTTS estimation and the performance of toll and public transport demand models in South Africa.
Paper presented at the 35th Annual Southern African Transport Conference 4-7 July 2016 "Transport ? a catalyst for socio-economic
growth and development opportunities to improve quality of life", CSIR International Convention Centre, Pretoria, South Africa.