Rouholamin Mahdi Pour - PhD, Department of Civil Engineering, Auburn University, Auburn, AL, USA
Huaguo Zhou - PhD, P.E, Department of Civil Engineering, Auburn University, Auburn, AL, USA

Single-vehicle crashes on rural two-lane highways impose a considerable risk to road users due to their higher severity outcome compared to other crashes on these facilities. Furthermore, considerable variation in the severity among various age groups (young, middle-aged, and older drivers) has been noticed, corroborating the need for analyzing age-classified data. Crash data from Alabama was compiled and classified based on the age group. For each age class, a generalized ordered logit model was developed to identify the effect of various variables on injury severity. This model can consider ordered nature of severity as well as provide flexibility in calculating the parameter estimates. Driver gender, seatbelt use, damage to the vehicle, driving on county roads, hitting a fixed object and animal, and speeding were found to be significant in all developed models. Intoxication is a significant factor that affects injury severity for young drivers. Time of day also significantly affects the injury severity for older drivers. Vehicle age and driving with invalid license were not found to affect injury severity for older drivers, while they affected the other age groups. It was shown that some factors have significant effect on the injury severity for all age groups while others have varying effect across different age groups. The results of this study highlight the importance of considering separate injury severity models for different age groups, specifically separating older drivers from others, as the difference among older drivers and others are substantial.

Age Difference; Single-Vehicle Crash; Rural Two-Lane Highways; Generalized Ordered Logit Model