Drivers Lane-Changing Behavior at Bus Stops

Transcription

1 Drivers Lane-Changing Behavior at Bus Stops Jung-Taek Lee, PhD, PE Randy Dittberner, PE, PTOE Bobby Mangalath, PE Northern Region Operations Traffic Engineering Section VASITE 2012 Annual Meeting, June 29, 2012

2 Background Safety issue Undivided two-way streets with one lane in each direction and a two-way left-turn lane. Crash potential between cars turning left and cars in the opposite direction passing a bus stopped at a bus stop. Is there any difference between two-way left-turn lane and one-way left-turn lane for crash potential?

3 BUS

4 BUS

5 Legal violations? Is it legal to pass a stopped bus on this type of street? Does it matter whether it s atwo- way left-turn lane or a one-way left- turn lane?

6 Special regulations applicable on highways laned for traffic. 5. Wherever a highway is marked with double traffic lines consisting of a solid line immediately adjacent to a broken line, no vehicle shall be driven to the left of such line if the solid line is on the right of the broken line, but it shall be lawful to make a left turn for the purpose of entering or leaving a public, private, or commercial road or entrance. Where the middle lane of a highway is marked on both sides with a solid line immediately adjacent to a broken line, such middle lane shall be considered a left-turn or holding lane and it shall be lawful to drive to the left of such line if the solid line is on the right of the broken line for the purpose of turning left into any road or entrance, provided that the vehicle may not travel in such lane further than 150 feet; 6. Wherever a highway is marked with double traffic lines consisting of two immediately adjacent solid lines, no vehicle shall be driven to the left of such lines, except when turning left.

7 Study Approach Compare drivers behaviors when buses stop at bus stops in different locations. With one-way or two-way left-turn lanes. With or without bicycle lanes. Data collection for drivers behaviors. Observe bus arrivals. Count passing-over cars (violation). Count cars stopped behind stopped buses.

8 Data Collection Locations Locations Bicycle Lane Left-turn Center Lane Observation Time YES NO ONE-WAY TWO-WAY Soapstone Drive Durand Drive Purple Beach Drive Ridge Height Road 6 AM to 9 AM 6 AM to 9 AM 6 AM to 9 AM Lawyers Goldcup 4 PM to Road Lane 8PM Jermantown Road Fairhaven Ct Cavalies Ct 10 AM to 2 PM 10 AM to 2 PM

9

10

11

12

13

14

15

16 Street With Bicycle Lane

17 Street Without Bicycle Lane

18 Number of Bus Arrival Observations Locations Total Observations Valid Observations Valid / Total Soapstone Drive Lawyers Road Durand Drive Purple Beach Drive Ridge Height Road Goldcup Lane % % % % Jermantown Road Fairhaven Ct % Cavalies Ct % Total %

19 Violation Rate for Bicycle Lane Presence Bicycle Lane Left-Turn Lane Observations Violation* Violation Rate Two-Way % Yes One-Way % Total % Two-Way % No One-Way % Total % *Violation: One or more cars pass over a stopped bus illegally in a leftturn lane.

20 Violation Rate for Left-Turn Lane Types Left-Turn Lane Bicycle Lane Observations Violation Violation Rate Yes % Two-Way No % Total % Yes % One-Way No % Total %

21 Number of Violating Cars Bicycle Left-Turn Violating Total Cars Lane Lane Cars Violation Rate Two-Way % Yes One-Way % Total % Two-Way % No One-Way % Total %

22 Chi-square Tests Test a null-hypothesis (H 0 ) that a certain event is same between certain conditions. Use contingency table for events and conditions. Example: Bicycle lane vs. violation. Reject H 0 if X 2 value from observed data is greater than X 2 value from chi-square distribution at significance level, = X 2 = Chi-square value O i,j = an observed frequency n = the number of cells in the table E i,j = an expected frequency

23 Bicycle Lanes and Violation Violated? YES NO Total YES Observed NO Are Bicycle YES Lanes Expected Present? NO Total

24 Test Result H 0 : Lane-changing violations at a bus stop are the same for locations where a bicycle lane is present and where a bicycle lanes is absent. X 2 = > ( = 0.05). Reject H 0. Therefore, there is strong evidence that lane changing violations from drivers behind a bus stopped at a bus stop are more common where bicycle lanes are present.

25 Left-Turn Lane Type and Violation Violated? YES NO Total Left-Turn Type Observed Expected One-Way Two-Way One-Way Two-Way Total

26 Test Result H 0: Lane-changing g violations at a bus stop are the same for locations where a left-turn lane is two-way left-turn and where it is one-way left-turn. X 2 = < ( = 0.05). Do not reject H 0. Therefore, there is strong evidence that lane changing violations from drivers behind a bus stopped at a bus stop are not significantly different whether the left-turn is two-way or one-way.

27 Conclusions Lane-changing g violations are observed more at a bus stop where a bicycle lane is present. Locations at Soapstone Drive where bicycle lanes are present show almost 100% violations. Little difference in lane-changing violations between two-way and one-way left-turn lanes. Locations at Jermantown Road without bicycle lanes show low violation rates, 7% and 30% only.

28 Discussion Designers should consider left-turnturn conflicts at bus stops when installing new bicycle lanes. Near-side bus stops may help reduce the conflict. More data collection and investigating crash history at bus stops may produce more reliable results. Advanced data analysis using other factors (bus dwell time, traffic volume, speed, time of day, area type, driver type, etc.) may provide clues to improve safety at bus stops.

29 Q&A