Statistical Modeling of User Perceptions of Infrastructure Condition: Application to the Case of Highway Roughness

Abstract

In determining certain infrastructure rehabilitation needs, it is sometimes important to consider user perceptions along with physical measures of infrastructure condition. Pavement roughness is one such case. A critical determinant of public satisfaction, user perception of pavement roughness can potentially play a critical role in the allocation of resources to competing highway resurfacing projects. In this paper, to gain a better understanding of user perceptions of pavement roughness, users were placed in real-world driving conditions and asked to rank the roughness of specific roadway segments. Coupled with individual-specific, pavement-specific, and vehicle-specific data, users’ roughness rankings were modeled using a random effects ordered probit specification. The model identified a number of key factors influencing user roughness rankings. The results indicate that, while physical roadway-roughness measurements, such as the measured International Roughness Index, provided a strong indication of user roughness rankings (as one might expect), other factors such as the type of vehicle used, vehicle speed, individual’s age, individual’s gender, and interior vehicle noise levels were also significant. This study fills an important gap in the literature by linking physical infrastructure measurements with individual perceptions of infrastructure condition.