A preliminary investigation of the relationships between historical crash and naturalistic driving

Abstract

This paper describes a project that was undertaken using naturalistic driving data collected via Global Positioning System (GPS) devices to demonstrate a proof-of-concept for proactive safety assessments of crash-prone locations. The main hypothesis for the study is that the segments where drivers have to apply hard braking (higher jerks) more frequently might be the "unsafe" segments with more crashes over a long-term. The linear referencing methodology in ArcMap was used to link the GPS data with roadway characteristic data of US Highway 101 northbound (NB) and southbound (SB) in San Luis Obispo, California. The process used to merge GPS data with quarter-mile freeway segments for traditional crash frequency analysis is also discussed in the paper. A negative binomial regression analyses showed that proportion of high magnitude jerks while decelerating on freeway segments (from the driving data) was significantly related with the long-term crash frequency of those segments. A random parameter negative binomial model with uniformly distributed parameter for ADT and a fixed parameter for jerk provided a statistically significant estimate for quarter-mile segments. The results also indicated that roadway curvature and the presence of auxiliary lane are not significantly related with crash frequency for the highway segments under consideration. The results from this exploration are promising since the data used to derive the explanatory variable(s) can be collected using most off-the-shelf GPS devices, including many smartphones.

Spatial prediction of traffic accidents with critical driving events – Insights from a nationwide field study

“…The thresholds suggested for identifying near-miss jerk events significantly vary between research projects (Aichinger et al, 2016). These thresholds range from very strong events between -9.9 m/s 3 and −12.6 m/s 3 (Bagdadi and Várhelyi, 2011), which occur very infrequently in our fleet of professional drivers, with less than 600 events in the 18-week period, to very low events between −0.15 m/s 3 and −0.61 m/s 3 (Pande et al 2017), which were triggered for almost every braking event that occurred in our study. As such, a threshold value of −2 m/s 3 was chosen to classify an event as high jerk.…”

“…horizontal and vertical alignments), weather, and seasonal effects on the basis of a multivariate analysis (Shankar et al, 1995). Additionally, the impact of horizontal curvature and an auxiliary lane has recently been investigated (Pande et al, 2017). The Crash Rate method is similar to the concept of Crash Frequency, but provides a measure of accident exposure of vehicles as it takes traffic volume into account (Hauer and Persaud, 1984).…”

“…Unfortunately, this option comes with difficulties, such as drivers interacting with the phone during the journey, triggering high acceleration values and thus leading to false positive events. In very recent work, a small naturalistic driving study of university staff members, using only data gathered from a GPS logger, went beyond smartphone data and modelled accident data on US Highway 101 (Pande et al, 2017). In the study, a total of 39 segments were analysed using percentage of high jerk (rate of change of acceleration) events, slope, average daily traffic, and other explanatory variables in a negative binomial model.…”

“…Following a recent study, high-jerk events were considered for classifying near-miss and CDEs (Pande et al, 2017). For the purpose of this analysis the jerk event behaviour was calculated from the vehicle's longitudinal acceleration sensor, which was sampled at 10 Hz.…”

“…Average Daily Traffic (ADT) (Hauer and Persaud, 1984). While there is research showing that the situational factors of crashes and near-misses are strongly related, there is limited empirical data on whether the Crash Frequency and Crash Rate of potentially dangerous locations can be reliably identified through analysis of CDEs (Pande et al, 2017).…”