| dc.description.abstract | There is much discussion about the user-perceived value for various travel
features (Brownstone and Small, 2005; Fosgerau and Karlström, 2010), the public
bus system levels and its metrics (dell’Olio et al., 2010; Dell’Olio et al., 2011;
Hensher et al., 2003), and the factors that determine travel time and its variability
(Mazloumi et al., 2009; Moghaddam et al., 2011; Surprenant-Legault and El-
Geneidy, 2011). In this context, this thesis aims to quantify the effects of the main
design variables of a bus system over its service levels. It is considered as system
levels: the average bus speed, the coefficient of variation (C.V.) of the travel time,
and the C.V. of the times between buses of the same service. Additionally, it’s
intended to identify the savings and costs generated by the use of dedicated public
infrastructure and high-speed private highways, in the public bus system
Transantiago, of Santiago de Chile.
A method is proposed and applied, for the crossing of the georeferenced data of
dedicated infrastructure and highways, with the other services’ attributes. Then, a
weighted least squares with White’s standard errors model (Mukherjee et al., 1998;
White, 1980) is developed for each identified system level index.
It is found, that the use of dedicated infrastructure results in minor improvements
over each of the three system levels: for every 10 additional km, the average bus
speed increases in 0,1 km/h and the C.V. of the travel time and of the times
between buses are decreased in 9% and 2%, respectively. On another hand, for
every 10 additional km of highways, the average bus speed increases in 7 km/h;
but the C.V. increases for the travel time in 13%, and for the times between buses
in 4%. It’s also found, that most of the irregularity of the system is accountable to
its operators, through the coefficient of variation of the times between buses at the
start of the route. For each percentage point increase in the C.V. of the times
between buses at the start of the route, the C.V. at the end of it increases in 0,8%.
When looking at the Transantiago case, the savings generated by the use of
vii
highways stands out, while the savings generated by the use of dedicated
infrastructure are minor. | |
