Results from a Simulator-Based Analysis of Oregon Crash Data - - PDF document

Results from a Simulator-Based Analysis of Right-Hook Bicycle Crashes at Signalized Intersections

1

Chr hristophe her M. Mons nsere monsere@pdx.edu Portland State University monsere@pdx.edu Presented at ICTCT Workshop “Improving the safety of vulnerable road users” – Challenges and Opportunities Vancouver, Canada March 9, 2018

Oregon Crash Data (2007-2011)

2 3 4

Right-Hook Severity Summary Methodology

5

Simulator Experiment 2

Evaluate the effectiveness of four categories of treatments to mitigate right-hook crashes.

Simulator Experiment 1

Experimentally verify the influence

• f four factors that

potentially contribute to right- hook crashes.

Field Validation

Validate through field observations the motorist-bicyclist interaction exhibited in Simulator Experiment 1.

OSU Driving Simulator

6

View from outside the car View from inside car w/bicycle

Simulator Experiment 1 – Design

• 67 Participants
• 16 Simulator Sickness
• 51 Usable
• 1,071 total-right turn scenarios
• Visual attention
• Fixations (eye tracker)
• Situational awareness
• SAGAT responses
• Crash avoidance
• Position and speed of

vehicles, bicycles, and pedestrians

Simulator Experiment 1 – Variables

8

Name of the Variable Levels Relative position of bicyclist None One (1) bicyclist riding in front of the motorist in an adjacent bicycle lane to the right One (1) bicyclist coming from behind the motorist in an adjacent bicycle lane to the right Speed of bicyclist Lower (12 mph, 19 kph) Higher (16 mph, 26 kph) Presence of oncoming vehicular traffic None Three (3) vehicles Presence of conflicting pedestrian None One (1) pedestrian walking towards the motorist

Visual Attention Areas of Interest (AOIs)

9

Visual Attention – Avg. Total Fixation

10

Frequency

• f fixation

Bicyclist position Ahead Behind Total (n) 328 328 Fixated 284 145 % 87% 44%

11

Situational Awareness

Mean percentage of correct responses

Field Validation

• November 5,

2014 to February 12, 2015

• All days of week
• 144 hours
• Extraction of 43

events with measured PET < 5 seconds

Comparison of Field and Simulator Simulator Experiment 2 – Design

• 46 Participants
• 18 Simulator Sickness
• 28 Usable
• 616 total-right turn scenarios
• Visual attention
• Fixations (eye tracker)
• Crash avoidance
• Time-to-collision
• Crash severity
• Speed of turning vehicles

The most complex driver loading (bicycles approaching from behind at higher speeds)

15

Simulator Experiment 2 - Variables

Dashed white bike line with stencil, single line Dashed white bike line with stencil, double line Dashed green bike lanes with white

• utline

Full green bike lane with dashed white

• utline

ODOT OR10-15b “Turning Vehicles Yield to Bicycles” Larger curb radii, 30ft Smaller curb radii, 10ft With islands With islands and green pavement markings

PROTECTED INTERSECTIONS

Average Total Fixation Durations

16

Simulator Experiment 2- Crash Avoidance

17

Simulator Experiment 2- Crash Severity

18

Curb Radius Treatment 30 ft (9 m) 10 ft (3 m)

Final Comparison

• Each treatment was evaluated based on

the following:

– Visual attention

• Measurable change in longer AFTD towards

bicycle targets

– Crash avoidance

• Frequency of low and moderate TTC observations

– Crash severity

• Speed of turning vehicles and variance of speed

19 20

Final Comparison

21

Recommendations

Dashed white bike line with stencil, single line Dashed white bike line with stencil, double line Dashed green bike lanes with white

• utline

Full green bike lane with dashed white

• utline

ODOT OR10-15b “Turning Vehicles Yield to Bicycles” Larger curb radii, 30ft Smaller curb radii, 10ft With islands With islands and green pavement markings

PROTECTED INTERSECTIONS

Acknowledgements

• Co-PI: David Hurwitz,

Associate Professor, OSU

• PhD Student:

– Mafruhatul Jannat, PhD ’15, OSU

• MS Students:

– Ali Razmpa, MS ’15, PSU – Jennifer Warner, MS ’15, OSU

• Undergraduate Research Assistants:

– Amy Wyman, UHC OSU – Kayla Fleskes, BSCE OSU – Katie Mannion, BSCE OSU – Amber Meeks, BSCE OSU

Reports / Papers

• Warner, J, Hurwitz, D., Monsere, C., Fleskes, K. “A Simulator-

Based Analysis of Engineering Treatments for Right-Hook Bicycle Crashes at Signalized Intersections” Accident Analysis & Prevention, 2017 https://doi.org/10.1016/j.aap.2017.04.021

• Jannat, M., Hurwitz, D., Monsere, C. Funk, K. “The Role of Driver’s

Situational Awareness on Right Hook Bicycle-Motor Vehicle Crashes”. In revision, Safety Science, March 2018

• Jannat, M., Monsere, C., Hurwitz, D. “Measuring Driver’s Visual

Attention in Right-Hook Crashes Scenarios” In review, Transportation Part F, Submitted February 2018

• Hurwitz, D., Monsere, C., Jannat, M., Warner, J, Razmpa, A.

“Towards Effective Design Treatment For Right Turns At Intersections With Bicycle Traffic”. Oregon Department of Transportation SPR Project No. 767, June 2015 (link)

23

Questions?

24

David Hurwitz, PhD Associate Professor Oregon State University Email: david.hurwitz@oregonstate.edu and Chris Monsere, PhD, PE Department Chair and Associate Professor Portland State University Email: monsere@pdx.edu