Our objective is to determine how various types of truck accidents are related to traffic flow conditions and roadway characteristics on urban freeways. The case study involves data on 19,000 accidents that occurred over two years on six freeways in Orange County, an urbanized area of Southern California; over 10% of these accidents involved trucks with at least six wheels on the road. The propensity of truck-involved accidents is found to be a decreasing function of the number of lanes and the average annual daily traffic (AADT) per lane, and an increasing function of truck percentages of AADT, all factored by time of day and day of week effects and weather conditions. The likelihood of a truck being involved in an accident is particularly sensitive to proportion of large (five axle or more) trucks. Accidents involving lane changes are the prevailing type of truck accident, and about 42% of truck-involved accidents involve one of the first two vehicles changing lanes or merging, compared to only about 17% of accidents not involving trucks. In contrast, 45% of accidents not involving trucks are rear end collisions in which the involved vehicle are proceeding straight ahead, compared to only about 18% of truck-involved accidents. A multinomial logit model is used to determine differences in the traffic and roadway conditions conducive to three types of truck accidents: weaving, runoff, and rear-end accidents. Results show that patterns of truck and non-truck involved accidents vary substantially, and we point out some implications regarding mitigation strategies. (Author/publisher)
Samenvatting