Abstract
Most serious crashes involving bicyclists occur at non-intersecting road locations. Over the past decade, there has been a steady increase in the number of bicyclist fatalities. According to the U.S. Department of Transportation National Highway Traffic Safety Administration (NHTSA) crash report [1] and Fatality Analysis Reporting System (FARS) [2], there was a record low 623 bicyclist fatalities in 2010 and it climbed to 966 fatalities in 2021, which is the highest it has been since 1975. This is a surprising trend given that many states, cities, and municipalities have been installing bicycle lanes to accommodate the increasing number of bicyclists. Overall, bicycle lanes have reduced crashes up to 49% on urban 4-lane roads and 30% on 2-lane urban roads [3] and reduced fatalities for all road users [4]. However, these studies aggregated data from 12 different cities and therefore, due to a variety of factors, some individual cities did not see crash reductions of this magnitude. For example, when looking at shared bicycle lanes, there was an 18% risk reduction in New York City [5] but no benefit Chicago [6].
We propose a study to identify factors for improving bicycle lane safety in Pittsburgh. Over the past 20 years, bicycle lanes in Pittsburgh have increased from 10 miles to over 100 miles. We will assess bicycles lanes in Pittsburgh and provide valuable information to the city to aid in their plans to expand bicycle lanes by 150 miles over the next decade. We will take a three-pronged approach to assessing bicycle lane safety (Figure 1). First, crash data will be analyzed and compared to data from similar cities. Second, surveys will gather information (e.g., attitude, incidents, etc.) based on personal experiences using bike lanes. Finally, we will develop and deploy a camera-based platform that automatically computes bicycle lane analytics, which will include information on when, where, and how often the bicycle lanes are most often used. We will also automatically identify hazards in or near bike lanes (Figure 2). Hazards may include stopped or parked vehicles, potholes, crashes, near-misses, fallen debris, snow, etc. These computed analytics may be shared with bicycle riders via a smartphone app to aid in planning and with Pittsburgh’s Department of Mobility and Infrastructure to address any issues and aid in planning the expansion of Pittsburgh’s bicycle network. For example, vehicles frequently stopped in an unprotected bike lane in a commercial district may indicate that delivery drivers are using the bike lane as temporary parking, which can be resolved by making the bike lane protected.
Description
Timeline
Strategic Description / RD&T
This project directly addresses the Data-Driven System Safety research priority of the RD&T Plan as specified below.
Safety Grand Challenge, Safe Design (page 19):
- Evaluate the safety performance of infrastructure design and develop and promote
the use of effective safety countermeasures
- Identify and support strategies to increase vulnerable road user safety (e.g., pedestrians,
bicyclists, motorcyclists, and people with disabilities).
Safety Grand Challenge, Safe Data (page 19): Research and develop new methodologies and tools for safety data collection, management, analysis, and evaluation:
- Develop safety data collection methods and advanced safety data and risk analysis techniques to identify and analyze emerging safety issues.
- Provide the scientific and engineering basis for policy decisions, improved industry standards, and enforcement and compliance matters.
- Assess safety incident trends and causes to enhance safety requirements and best practices.
Deployment Plan
Survey Deployment
Q1: Develop survey questions and identify locations for participant recruitment
Q2-Q4: Deploy survey and analyze responses
Camera Deployment
Q1: Identify camera location(s) for observing bicycle lane activity near CMU's campus
Q2: Install cameras, if necessary, at identified location(s) and capture images
Q3/Q4: Develop algorithms for identifying bicyclists and detecting hazardous conditions and apply to images
Expected Outcomes/Impacts
It is anticipated that using a combination of crash data and survey data will provide a robust understanding of bicycle lane safety in Pittsburgh. The results will be shared with DOMI, which may prove useful for their bicycle network design plans. Hazard info will be computed for the bike lanes covered by the deployed camera system(s). While we cannot install enough cameras to capture the 100+ miles of bicycle lanes in Pittsburgh, it is expected that this proof of concept platform will provide real-time information about the state of the bike lane. Expansion of the platform could provide broad monitoring of bike lanes for damage (potholes, infrastructure), obstructions (stopped vehicle), and other hazards. Damage can be quickly repaired and obstructions can be quickly cleared to keep the bike lane clear for riders.
Expected Outputs
We anticipate the following outputs:
- Comprehensive analysis of crash data in bike lanes in Pittsburgh. Analysis, will break down crashes based on variables such as proximity to intersection, type of bike lane, etc.
- Survey data from the 4 types of bicycle riders
- Algorithm for detecting bicycles and hazardous activity and conditions in bike lane(s) where camera(s) are present.
- Statistics of hazards in bike lane(s) where camera(s) are present
- Bike lane dataset of images where camera(s) are present
TRID
A search for "Bicycle Lane Safety in Pittsburgh" yielded 5 results. Three of the results are about bicyclists and their perceptions of safety when sharing the road with autonomous vehicles and are irrelevant to this project. The other two results are related to a project about complete streets in Pittsburgh; one being the project itself and the other about a supporting dataset. The project focuses on evaluating travel and air quality impacts resulting from a complete street redesign on a section of Forbes Avenue. The redesign centered on reducing traffic lanes and adding bicycle lanes. The major findings were that vehicle volume and vehicle speeds reduced, and bicycle counts increased. Crash counts are reported from the Pennsylvania Department of Transportation up to 2019-- prior to completion of the retrofit project.
The proposed project focuses specifically on bicycle lane safety in Pittsburgh. Forbes Avenue will be included in the analysis, but so will many other areas in the Pittsburgh area that have undergone complete streets retrofitting. Our analysis will also include post-COVID pandemic data which will reflect changes in personal driving and bicycling behaviors. Our work would provide an update to some of the results reported in the prior work while uniquely characterizing bicycle lane safety in Pittsburgh.
Individuals Involved
| Email |
Name |
Affiliation |
Role |
Position |
| srinivas@andrew.cmu.edu |
Narasimhan, Srinivasa |
Carnegie Mellon University |
Co-PI |
Other |
| rtamburo@cmu.edu |
Tamburo, Robert |
Carnegie Mellon University |
PI |
Other |
| kvuong@andrew.cmu.edu |
Vuong, Khiem |
Carnegie Mellon University |
Other |
Student - PhD |
Budget
Amount of UTC Funds Awarded
$95000.00
Total Project Budget (from all funding sources)
$99201.00
Documents
Match Sources
No match sources!
Partners
| Name |
Type |
| Pittsburgh Department of Mobility and Infrastructure |
Deployment Partner Deployment Partner |
| BikePGH |
Equity Partner Equity Partner |
| dashcam.bike |
Deployment Partner Deployment Partner |
