Vehicle crashes on roads are caused by many factors. However, the influence of these factors is not necessarily homogenous across locations, which is a challenge for non-stationary modeling approaches. To address this problem, this paper adopts two types of methods allowing parameters to fluctuate among observations, that is, the random parameter approach and the geographically weighted regression (GWR) approach. With road curvature, curve length, pavement friction, and traffic volume as independent variables, vehicle crash frequencies are modeled by two non-spatial methods, including the negative binomial (NB) model and random parameter negative binomial (RPNB), as well as three spatial methods (GWR approach). These models are calibrated in microlevel using a dataset of 9415 horizontal curve segments with a total length of 1545 kilometers for a period of three years (2016–2018) over the State of Indiana. The results revealed that the GWR approach can capture spatial heterogeneity and therefore significantly outperforms the conventional non-spatial approach. Based on the Akaike Information Criterion (AICc), geographically weighted negative binomial regression (GWNBR) was proved to be a superior approach for statewide microlevel crash analysis.

Surface friction is currently the most common metric for evaluating the performance of high friction surface treatment (HFST). However, friction test methods such as the locked wheel skid tester (LWST) commonly provide a spot measurement. Large variations may arise in the LWST testing on curves. Based on 21 actual HFST projects, a study was performed to use a macrotexture metric, i.e., the mean profile depth (MPD) to evaluate HFST’s performance and improve its quality control (QC)/quality assurance (QA) procedures. The material properties were presented to understand the aspects of HFST. The method for calculating MPD was modified to account for the variations of macrotexture measurements. A vehicle-based test system was utilized to measure MPD periodically over an 18-month period since HFST installation. Statistical analysis was performed on the MPD measurements to identify the effects of influencing factors. Compared with the friction from LWST, MPD was equally effective in evaluating HFST performance. However, the use of MPD eliminated the errors as arisen in LWST testing and made it possible to detect surface distresses, including aggregate loss, delamination, and cracking. The expected overall MPD may be calculated by combining the MPD measurements made three months after installation at different HFST sites and used as a metric for evaluating HFST performance and QC/QA.