Changes in road-user behaviours following % of the installation of - - PDF document

Changes in road-user behaviours following the installation of raised pedestrian crossings on urban arterials

Victoria Gitelman, Roby Carmel, Fany Pesahov, Sarit Chen

Transportation Research Institute Technion – Israel Institute of Technology

28th ICTCT Workshop in Ashdod, Israel

• n 29th – 30th October 2015

Pedestrian injury problem in Israel

% of total injuries % of total fatalities

Previous research:

• Over 80% of pedestrian accident sites: on arterial/collector multi-lane

streets, in city centers (Balasha et al. 2009)

• Infrastructure measures most effective for increasing pedestrian safety

(Zegeer and Bushell, 2012) 10 % 10 % 11 % 11 % 13 % 13 % 14 % 33 % 33 % 34 % 34 % 34 % 33 % 42 %

Raised pedestrian crossing solution

Guidelines for speed humps design and application (2002), for local streets:

Literature findings

Hawley et al (1992), Australia: installation of raised crossings → a reduction of 40% in the 85-percentile speed Harkey and Zegeer (2004), USA: raised pedestrian crossing → a reduction of the 85-percentile speed from 47 km/h to 42 km/h Hakkert et al. (2002), Israel: on collector residential streets, the mean travel speed on the streets with raised pedestrian crossings was lower than

• n the comparison streets - 42 km/h vs. 53 km/h

Ewing (2008), USA: raised pedestrian crossing → an increase in giving- way to pedestrians, from 20% to over 50% Elvik et al (2009): raised crossing is associated with 65% reduction in injury accidents [-83%;-27%], 42% reduction in pedestrian accidents [-70%;+11%]

Current study - Method

Topic: raised pedestrian crossings installed on multi-lane divided streets, presenting main traffic arterials, with high traffic volumes and high pedestrian activity, in the crossing areas Study’s purpose: to examine the impact of the measure on driver and pedestrian behaviors, and to deduce on a possible impact on pedestrian safety Method: a controlled field-study, where road user behaviors in the crosswalk areas are compared during "after" vs. "before“ periods No of sites: 8 sites in 4 cities Field observations: speed measurements + video-recording

Study’s sites

In 4 medium- and large-sized cities (population 45,000- 200,000), in the center and the north of the country

Site requirements:

• Dual-carriageway road segment with a built

median, 2 lanes per direction and a marked crosswalk

• Speed limit of 50 km/h
• 85-percentile speed above 50 km/h, at least in
• ne direction
• Medium to high volume of pedestrian activity

(over 30 crossings per hour)

• Straight and flat road segment
• Visibility distance of at least 50 meters ahead

8 sites X 2 parts (travel directions) = 16 crosswalks examined

Karmiel Netanya Herzlia Hod Hasharon

Building a raised crossing - components

Example: site 1 (Hod Hasharon, Jabotinsky street), direction from west

Installing a trapezoidal speed

hump on the crosswalk Building a circular speed hump 15-20 m before the crossing Adding traffic signs

Study sites’ examples: before and after

Before Site 2, Hod Hasharon After Before After Site 5, Netanya

Behaviours examined

Vehicle travel speeds (speed gun, 100 free-flowing vehicles) Vehicles’ yielding to pedestrians at the crosswalk area Pedestrians crossing in the designated area Conflicts between pedestrians and vehicles during pedestrian crossings Pedestrians’ following safe crossing rules

3 rounds of observations:

Before - prior to installation of the new arrangement After1 - 1-2 weeks after the installation After2 ~ two months after the installation, to examine long-term behaviour changes Each round’s duration (video- recording): 6 hours, incl. 3 day- and 3 evening hours

Example: Speed indicators - site 1 (Hod Hasharon)

Period - travel direction Mean speed* sd 85 Percentile speed % above the speed limit Before - to West 48.7 7.0 56.0 39% After1 - to West 26.1 4.3 30.9 0% After2 - to West 22.8 5.1 28.0 0% Before - to East 43.6 6.5 50.9 15% After1 - to East 25.3 6.0 31.0 0% After2 - to East 24.9 4.8 29.0 0%

*Sig. differences in before-after1, before-after2 comparisons, ANOVA test p<0.001, Tukey HSD p<0.001

Cumulative speed distributions, site 1 – to west:

Before After1 After2

Road user behaviours before

Site 3, Herzlia

Road user behaviours after

Site 3, Herzlia (After1)

Example: Behaviour indicators - site 1, to west

* Sig. differences at p<0.05, z-test Behavior Indicator Before After1 After2 After1 vs. Before After2 vs. Before After2 vs. After1 % of pedestrians stopped before the crossing 48% 32% 33%

↓* ↓ ns

% of pedestrians checked the traffic before the crossing 89% 96% 99%

↑ ↑ ↑

% of giving-way to pedestrians by first vehicle, on near lane 80% 95% 96%

↑ ↑ ns

% of giving-way to pedestrians by first vehicle, on far lane 63% 91% 98%

↑ ↑ ↑

% of conflicts between pedestrians and vehicles in the crossing area 2% 2% 0%

ns ↓ ↓

% of pedestrians performed a full crossing at the crosswalk area 79% 95% 98%

↑ ↑ ↑

% of pedestrians performed a full

• r partial crossing at the crosswalk

area 96% 98% 99%

ns ↑ ns

Summary of behaviour changes – speeds*

* Sig. differences at p<0.05 # in one direction only

Study's Site Mean speed, km/h 85-percentile, km/h Site 1 ↓ from 44-49 to 23-25 ↓ from 51-56 to 28-29 Site 2 ↓ from 46-47 to 26 ↓ from 52-53 to 32-33 Site 3 ↓ from 56-58 to 28-29 ↓ from 64-66 to 36-37 Site 4 ↓ from 42-50 to 22-25 ↓ from 50-60 to 29-31 Site 5 ↓# from 43 to 31 ↓ # from 47 to 39 Site 6 ↓ from 41-46 to 32-34 ↓ from 47-53 to 38-40 Site 7 ↓ from 51-52 to 36-37 ↓ from 60-61 to 44 Site 8 ↓ from 51-55 to 29-30 ↓ from 59-65 to 35-37

Summary of changes in other behaviours (values)*

*Sig. differences at p<0.05 are given in colors **Sig. difference at p<0.1 # not relevant Study's sites % of pedestrians stopped before the crossing % of pedestrians checked the traffic before the crossing % of giving- way to ped’s by first vehicle, on near lane % of giving- way to ped’s by first vehicle, on far lane % of conflicts % of ped’s performed a full crossing at the crosswalk area Site 1 to west ↓ from 48% to 33% ↑ from 89% to 99% ↑ from 80% to 96% ↑ from 63% to 98% ↓ from 2% to 0% ↑ from 79% to 98% Site 1 to east ns (1/3) ↑ from 86% to 99% ↑ from 80% to 98% ↑ from 62% to 100% ns (1%-0%) ↑ from 75% to 87% Site 2 to north ns (1/3) ↑ from 93% to 95% ns (93%-97%) ↑ from 51% to 96% ↓ from 4% to 0% ↑ from 46% to 85% Site 2 to south ↑ from 41% to 54% ↓from 92% to 86% ns (72%-78%) ↑ from 31% to 73% ns (4%-3%) ↑ from 60% to 90% Site 3 to north ↓ from 98% to 15% ns (100%) ↑ from 96% to 100% ns (100%) ns (0%) ns (98%-96%) Site 3 to south ↓ from 91% to 10% ns (100%-99%) ns (100%) ns (100%) ns (0%) ↑ from 92% to 98% Site 4 to north ↓ from 16% to 6% ↑ from 98% to 100%** ns (100%) ns (100%) ns (0%) ↑ from 76% to 85%** Site 4 to south ns (7%) ↑ from 97% to 100% ns (98%) # ns (0%) ↑ from 63% to 87%

Conclusions

Following the installation of the raised crossings:

• a significant decrease in vehicle speeds was observed which was

maintained over-time

• the rate of giving-way to pedestrians improved, the rate of

vehicle-pedestrian conflicts decreased (if observed before)

• the percentage of pedestrians conducting a full crossing in the

designated area increased (if was low before)

• no consistent changes in keeping safe crossing rules by

pedestrians

Conclusions

The changes in road user behaviors following the installation of the raised crossings were positive and associated with a safety improvement of pedestrian crossing conditions A positive impact was found for sites with various levels of traffic volumes and of crossing pedestrians The treatment is recommended for application on arterial/collector urban roads. Positive changes will be more substantial at sites with problematic road user behaviors in the pre- treatment conditions

Thank you!