Left-Turn Traffic Organization at Intersections Based on Stochastic Processes

After the functional division of lanes at intersections, capacity constraints inevitably arise in certain directions as traffic volume increases, particularly for left-turn movements. When signal timing adjustments are no longer sufficient to address the issue, innovative intersection designs such as left-turn waiting areas and contraflow left-turn can be implemented to enhance the capacity of oversaturated movements without requiring major infrastructure modifications or significantly increasing delays for other directions. To overcome the limitations of relying solely on empirical methods in the implementation of these intersection designs, this study analyzes selected left-turn traffic organization strategies and signal timing characteristics. By applying stochastic process theories, the steady-state queue length of left-turn movements is estimated to guide the selection of appropriate traffic organization strategies.