Open Collections

Probabilistic calibration of highway geometric design : theoretical issues and applications

University of British Columbia

Road safety is an evolving area of research that underpins the continuous attempt to build safe and cost-efficient roads. Despite the considerable growth of the road safety literature, some concern remains regarding the safety level associated with standard geometric design models. The central concern is that the level of safety delivered by design standards is implicit and largely unknown. Accordingly, it remains difficult to form a judgment regarding the acceptability and the consistency of the safety level built in design standards. In order to account for the previous concern, road safety is quantified in the present practice by explicitly developed analytical tools or safety evaluation models. Those analytical tools, to some extent, guide the designer in investigating the safety consequences of different dimensioning scenarios for a highway design. However, in order to elicit and examine the implicit safety level in design standards, a quantifiable measure must be used to assess road safety. A parallel system is formulated for the assessment of safety levels in standard design outputs by tracing the propagation of uncertainty. Uncertainty can be incorporated into the design process through a probabilistic framework. Reliability theory, a subset of probability theory, offers a rational foundation for calculating the propagation of uncertainty throughout the design process. The main proposition that underlies the analysis presented in this thesis is that design safety level associated with standard design outputs should be consistent and close to a premeditated level. Several discussions are presented that suggest different methods of selecting a target design safety level. A general framework for calibrating standard design models is presented in accordance with the previous preposition. The thesis contains an application of the calibration framework to the standard design model of crest vertical curves. In order to study the effect of model uncertainty, represented by the combination of horizontal and vertical curves, a new sight distance model is formulated that enables the calculation of available sight distance in three-dimensional environment. The sight distance model is further augmented to the process of reliability analysis. The calibrated design charts are constructed in order to yield consistent design safety level.

Text

2010-01-20

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/18733

Civil Engineering

University of British Columbia

Graduate