J T. Siques

Document Type

Journal Article

Publication Date


Subject Area

operations - traffic, planning - methods, planning - safety/accidents, mode - rail, mode - tram/light rail, mode - pedestrian


Warning devices, Videotapes, Traffic signs, Traffic markings, Statistical methods, Statistical analysis, Road markings, Risk taking, Railroad grade crossings, Portland (Oregon), Pedestrian safety, Pavement markings, Mathematical statistics, Light rail transit, Level crossings, Highway signs, Highway railroad grade crossings, Highway rail intersections, Grade crossings, Countermeasures, Channelization, Carriageway markings, Before and after studies, Automatic gates


The effects of pedestrian treatments on risky pedestrian behavior at light rail transit grade crossings were examined. Five pedestrian treatments were evaluated--(a) pedestrian automatic gates, (b) a prototype active pedestrian warning device, (c) a prototype active "Look Both Ways" sign, (d) barrier channelization at a skewed crossing, and (e) a "Stop Here" pavement marking. Pedestrian grade-crossing treatments were installed at three grade crossings along the Tri-County Metropolitan Transportation District of Oregon MAX light rail system in Portland, Oregon. The pedestrian treatments and the crossing geometry varied at the three locations, providing for three unique evaluations on the effectiveness of different pedestrian treatments at grade crossings. The grade crossings were videotaped for at least 1 week both before and after the installation of the pedestrian treatments. The data were evaluated using a before-and-after statistical approach to determine the effects of the treatments on risky pedestrian behavior. The statistical evaluation of the data shows that pedestrian treatments result in a statistically significant reduction in risky pedestrian behavior. The greatest reductions were found with pedestrian automatic gates. The results, however, also demonstrate that various pedestrian treatments can sometimes increase risky pedestrian behavior.