Objectives of this project were to: (1) determine the accident experience when STOP-controlled intersections are converted to YIELD control, and (2) develop guidelines for converting STOP control to YIELD control. A literature review and survey of highway agencies was conducted to determine current engineering practice and safety experience at STOP and YIELD controlled intersections. The survey showed that very few agencies had changed their control from STOP to YIELD and there appeared to be widespread reluctance to do so because of the perceived safety consequences and concern over liability. A total of 756 YIELD and STOP controlled intersections in six cities were analysed in terms of their accident experience. Three cities that recently converted some STOP control intersections to YIELD control were used in analysing accident experience before and after conversion. The other three cities were used in analysing established YIELD control intersections. Safety results were integrated with user costs and benefits to develop criteria for converting from STOP to YIELD. Key findings were: (1) Intersections converted from STOP to YIELD control are likely to experience an increase in accidents, especially at higher traffic volumes (about 1 accident every 2 years); (2) Accident severity and distribution did not significantly change after conversion from STOP to YIELD; (3) Converted YIELD control intersections have a higher accident rate than established YIELD control intersections; (4) Four-leg intersections with YIELD control have a higher accident rate than T-leg intersections with YIELD control; and (5) Because of reduced motorist delay, fuel cost and other vehicle operating costs, YIELD control is more cost effective than STOP control at all volume levels studied. The guidelines developed for conversion of STOP to YIELD are as follows: (1) Have adequate sight distance; (2) Intersection volume less than 1,800 ADT, major street volume less than 1,500 ADT, and minor street volume less than 600 ADT are potential conversion candidates; and (3) Intersections experiencing less than 3 accidents in 2 years are candidates for conversion. (A)
Abstract