The purpose of this research project was to identify differences in operation and safety of highways with service road access control as opposed to direct access; to determine which traffic and geometric factors affect the operation and safety of the service road and highway at their points of intersection; and to apply these factors as criteria for the layout of various service road configurations. Data necessary for the study was collected at eleven service road locations, as well as two direct access locations, found in seven Indiana cities. A total of 51 service road and highway intersections, classified into various types, were studied at these locations. The kinds of data required for the analysis were conflicts data, volume data, accident data, speed data, and questionnaire data. Analysis of a conflicts index developed from conflict and volume data was not reliable, but it is considered important that no conflicts were recorded on approaches of service roads terminating at a crossroad. Accident data was utilized to detect patterns of hazardous movements at both service road and highway intersections. A comparison of speed characteristics for five pairs of locations revealed little difference in speeds on service road highways versus direct access highways. The questionnaire responses indicated that establishments on service roads fare just as well, if not better, than those with direct access. For seven of the intersection types studied, regression analysis was used to determine those variables which best explained the conflicts index. For instance, the volume of traffic on the service road gave a reliable explanation of conflicts at three-legged service road connections to the highway. Evaluation of the range of operational experience at such intersections indicated a maximum service road volume of about 2000 vehicles per day for acceptable operation. In a similar manner the recommendation was made to prohibit four-legged service road intersections unless the expected service road volume is less than 400 vehicles per day. Finally, those three-legged service road intersections at a crossroad which require a left turn off the crossroad when entering from the highway should not be permitted where the expected crossroad volume exceeds 500 vehicles per day; where a right-turn entry is made this crossroad volume would be about 2000 vehicles per day.
Samenvatting