Citation

Lu G, Noyce DA. Transp. Res. Rec. 2009; 2140: 63-78.

Copyright

(Copyright © 2009, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

10.3141/2140-07

PMID

unavailable

Abstract

The increasing number of midblock vehicle-pedestrian crashes has led traffic engineers to consider safer treatments for pedestrian crossings while preserving flow efficiency. One of the existing solutions is to install signalized crosswalks. Using a microsimulation approach, this study first assesses three signal systems for a typical midblock crosswalk (MBC) with varied geometries, with the aim to explore how different signalization schemes and crosswalk geometries affect measures of effectiveness from all user perspectives. The results indicate that two-phase timing out-performs one-phase timing and the innovative high-intensity activated crosswalk significantly improves vehicle operations over actuation by pedestrians and pedestrian light control. Of existing signals, the pedestrian user-friendly interface (PUFFIN) is more functional because of its dynamic pedestrian clearance interval, but it still does not account for enough safety and human factors in its control logic and thus lacks an adaptive ability in fulfilling competing objectives. Fuzzy logic control (FLC) has proved effective for a complex optimization problem with multiple goals, uncertain information, and vague decision criteria. Traffic signal timing lies in this realm. To model the range of variables affecting MBCs, a user-friendly FLC counterpart is developed and then evaluated against PUFFIN to quantify potential safety and efficiency benefits. The results show that with straightforward logic and tractable parameters, FLC manages the MBC signal timing effectively and outperforms PUFFIN in terms of a compromise among enhanced safety, ameliorated operations, and lessened social cost from crashes and delays.