Incorporating Speed in a Traffic Conflict Severity Index to Estimate Left Turn Opposed Crashes at Signalized Intersections

Abstract

Rigorous evaluation of implemented safety treatments, especially for innovative treatments and those targeted at rare crash types, is challenging to accomplish with conventional crash-based analyses. This paper aims to address this challenge for treatments at urban signalized intersections by providing a methodology that uses surrogate measures of safety obtained from video analytics to predict changes in crashes. To develop this approach, left turn opposed traffic conflicts based on post-encroachment times, along with corresponding conflicting vehicle speeds, are first measured from video observations at signalized intersections. The conflicts are then classified into three severity levels using a risk score function defined by these measures. Multiple linear regression models are developed to relate left turn opposed crashes at the same intersections in the period 2009–2014 to the correspondingly classified conflicts. The results show strong relationships between the classified conflicts and crashes (adjusted [Formula: see text] of 85% and 94% for total and fatal/injury crashes, respectively). The results also reveal that the contribution of conflicts to the risk of crashes varies based on speed dimension of their severity, suggesting that neglecting speed as a factor in conflict severity levels may be at the expense of losing meaningful information. The models can be applied to estimate the change in crashes following a safety treatment by observing, through video analytics, the change in conflicts and speeds and using the crash-conflict-speed model. The methodological approach is viable for quickly evaluating all treatments and, in particular, innovative ones for which knowledge on safety effects is sparse or non-existent.

Advancing Crash Prediction Models Based on Simulated Conflicts and Exploring Their Predictive Capabilities and Transferability

“…Chapter 2 presents a literature review conducted to provide the foundation in the development of this thesis. The major areas discussed in this section cover the statistical modelling of crashes, definitions related to surrogate safety measures, the traffic conflict technique, transferability of crash prediction models, different levels associated with autonomous vehicles, a brief discussion of the software packages used in this study and a discussion of a similar study performed by Anarkooli et al (2020).…”

“…Table 2.1: Parameters for Crash Models using Video Analytics (Anarkooli et al, 2020) (Gettman et al, 2008) 7.16: Coefficient Estimates for Crash -Conflict and Average Speed (TTC ≤ 1 sec) ........ Table 7.17: Coefficient Estimates for Crash -Conflict and Average Speed (TTC ≤ 0.5 sec) ..... Table 7.18: Coefficient Estimates for Crash -Conflict and Maximum Speed (TTC ≤ 1 sec) ..... Table 7.19: Coefficient Estimates for Crash-Conflict and Maximum Speed (TTC ≤ 0.5 sec) .... 8.1: Calibration Goodness-of-Fit Ranking (Hayward, 1972) and PET (right) (Lord and Washington, 2018) .............. Figure 2.2: Hydén's Safety Pyramid (Laureshyn & Varhelyi, 2018) (Gettman et al, 2008) .......…”

Proceedings of the 2021 Australasian Road Safety Conference - Melbourne

“…While some conflict detection technology primarily considers temporal separation of road users, the approach introduced herein considers kinetic energy principles to identify events that are most likely to cause a severe injury or fatality. This approach has resulted in strong relationships between conflicts and fsi crashes (adjusted 𝑅𝑅 2 of 94%) (Anarkooli et al, 2021), and has been deployed in 65 communities for diagnostics, safety improvement planning and evaluations.…”

Exploration of Some Key Issues in Developing and Applying Crash–Conflict Models for Signalized Intersections

“…The main points of departure of the current study are the investigation of speed as a predictor in conflict-based crash prediction models, and the use of microsimulation to derive those speeds. Speed measures have been proposed for defining the severity of conflicts and developing relationships with crashes, but the speeds and conflicts were derived from observations ( 4 , 16 , 17 ). For example, Anarkooli et al ( 17 ) estimated models to relate left-turn opposed crashes to conflicts based on PETs and conflicting vehicle speeds all determined by video analytics.…”

“…Speed measures have been proposed for defining the severity of conflicts and developing relationships with crashes, but the speeds and conflicts were derived from observations ( 4 , 16 , 17 ). For example, Anarkooli et al ( 17 ) estimated models to relate left-turn opposed crashes to conflicts based on PETs and conflicting vehicle speeds all determined by video analytics.…”