Conclusions of a Network-Wide Study in Bellevue, Washington

Author: Julie Levy, B.Eng. Project Delivery Manager

In response to rising road safety concerns, the City of Bellevue passed a Vision Zero resolution in 2015 to strive to eliminate traffic fatalities and serious injuries by 2030. The City of Bellevue’s downtown core is made up of a large proportion of pedestrians, and there is an increasing cyclist population throughout the city, which motivated the need to focus mainly on vulnerable road user safety and risk in their interactions with vehicles. In 2019, the City of Bellevue used Transoft Solutions’ TrafxSAFE Connect on 40 of the city’s 200 signalized intersections, detecting over 8 million road users in the span of one week.

Here are some of the main findings from the study:

Even though cyclist volumes were relatively low throughout the city, the severity level of incidents involving bicycles was significant compared to that of other road user types.

Cyclist volumes were relatively low throughout the city most likely because at the time of the study, the city didn’t have a complete cycling network in place. However, there are two major cycling corridor projects being planned which should help encourage cycling in the city. Furthermore, even though cyclist exposure was low, the conflict severity was significant.

Speed information at intersections was a new data source for the City of Bellevue.

Before the TrafxSAFE Connect study, speeds in Bellevue had been mostly analyzed using screen lines at midblock locations. Being able to obtain speed information at intersections instead of at midblock locations was a new and valuable data source for the city. It was also useful to be able to separate speed information based on turning movements. The speed study concluded that speeding incidence rates were more prevalent in residential areas compared to commercial areas, and some speeds observed at mainlines were higher than expected.

Being able to normalize conflict data using exposure helped provide a more complete picture and a different way to look at intersection hotspots.

Over 1.5 million interactions were observed throughout the week of analysis. While more than 80% of these were safe passage, around 20% of the interactions had potential to result in conflict. Looking at conflict information normalized by exposure instead of simply looking at the number of crashes helped pinpoint some hotspots that had previously not shown up as critical sites.