Analysis of Naturalistic Driving Study Data: Offset Left-Turn Lanes (2014) / Chapter Skim
Currently Skimming:
Pages 27-55
The Chapter Skim interface presents what we've algorithmically identified as the most significant single chunk of text within every page in the chapter.
Select key terms on the right to highlight them within pages of the chapter.
From page 27... ...
27 C h a p t e r 4 This chapter provides a description of the data used in the analysis, the methodology used to address the primary research hypotheses, and the statistical analysis results. Results from secondary analyses are also discussed.
|
From page 28... ...
28 intersections were found for either of the positive-offset categories or for the offset category of -1 ft to -5 ft; a substantial majority of trips were observed at intersections with left-turn lane offset between -11 ft and -15 ft. This limited the potential for identifying differences between offset categories in the analysis.
|
From page 29... ...
29 be long. An accepted gap was considered "long" when no oncoming opposing vehicles were present in the forwardfacing camera at the time the left-turning vehicle began the turning maneuver, or when the rear camera showed that no opposing through vehicle passed through the intersection for at least 12 s after the turn was made.
|
From page 30... ...
30 Repeated Measures The NDS data provide an opportunity to know how many of the total turning maneuvers at any given intersection, or within any given offset category, were made by the same driver during the study period, as well as how many study intersections a given NDS driver was observed passing through. All statistical analyses documented in this report assumed statistical independence of the observations.
|
From page 31... ...
31 • Available gap length by offset category (4) for two-way stopcontrolled intersections (Figure 4.2)
|
From page 32... ...
32 Two-Way Stop-Controlled Intersections Red dot = mean; Horizontal line = median; Colored box = mid 50% of data; Blue circle = extreme value (a) -16 ft or less (b)
|
From page 33... ...
33 the view of oncoming vehicles for the left-turning NDS driver is quite high. For all available gaps, the percentage of opposing left-turn drivers that create a sight obstruction for NDS drivers is 86% for all negative-offset categories combined; for accepted gaps, it is 71%.
|
From page 34... ...
34 durations of rejected gaps represent crash risks judged unacceptable by the left-turning driver. The analysis of choice is therefore regression analysis in which the distributions of the accepted and rejected gap durations are analyzed to establish the critical gap duration (tc)
|
From page 35... ...
35 All analyses were performed separately for signalized and two-way stop-controlled intersections, using PROC LOGISTIC and PROC PROBIT, two statistical procedures of SAS Version 9.3 (SAS Institute 2013)
|
From page 36... ...
36 the turning vehicle closer to the opposing through lanes of traffic and, therefore, shorten the travel distance (and time) required to clear the intersection.
|
From page 37... ...
37 Figure 4.6. Critical gaps and 95% confidence intervals by offset category for signalized intersections.
|
From page 38... ...
38 From the model, the critical gap length, t50, and its 95% confidence interval were estimated, separately for each offset category. The results, along with the number of accepted and rejected gaps in each offset category, are shown in Table 4.7 and plotted in Figure 4.8.
|
From page 39... ...
39 intersections than for positive- and zero-offset intersections. In fact, there were few observations at positive-offset left-turn lanes in which an opposing vehicle restricted sight distance.
|
From page 40... ...
40 Figure 4.9. Predicted probability of accepting gap as function of gap length and presence of sight obstruction for signalized intersections with offsets of 16 ft or less.
|
From page 41... ...
41 Figure 4.11. Predicted probability of accepting gap as function of gap length and presence of sight obstruction for signalized intersections with offsets between 10 ft and 6 ft.
|
From page 42... ...
42 stop-controlled intersections are presented here. The following results were obtained: • The interaction term between gap length and sight obstruction was not significant at the 10% level in either of the two analyses; thus, parallel lines logit models were assumed.
|
From page 43... ...
43 Table 4.8. Critical Gap Estimates by Offset Category and Presence of Sight Obstruction for Signalized Intersections Offset Category Is Sight Distance Obstructed?
|
From page 44... ...
44 Figure 4.16. Predicted probability of accepting gap as function of gap length and presence of sight obstruction for two-way stop-controlled intersections -- all offsets combined.
|
From page 45... ...
45 unobstructed view versus 7.7 s for drivers with an obstructed view resulting from the presence of a left-turning driver (difference of 2.1 s)
|
From page 46... ...
46 traffic was obstructed by an opposing left-turning vehicle was low; this does not warrant a comparison of the differences in postencroachment times on the short end of the distribution between accepted gaps with and without sight obstructions. When considering all accepted gaps, Table 4.10 shows that the 1st-, 10th-, and 15th-percentile accepted gaps were shorter at negative-offset left-turn lanes than at zero- or positive-offset left-turn lanes at both signalized and stop-controlled intersections.
|
From page 47... ...
47 Driver age categories were designed to match SHRP 2's age categories shown on the NDS data website. However, some age groups were combined to provide a large enough sample size for comparison.
|
From page 48... ...
48 Figure 4.18. Distribution of postencroachment time for accepted gaps by NDS driver age group and gender at signalized intersections.
|
From page 49... ...
49 Figure 4.19. Distribution of postencroachment time for accepted gaps by NDS driver age group and gender at two-way stop-controlled intersections.
|
From page 50... ...
50 was red at the time of arrival in the queue, this time spent at the red light was included in the time spent waiting for a gap. The literature suggests that drivers who wait longer than 30 s for a gap tend to become impatient and select a shorter, riskier gap.
|
From page 51... ...
51 Presence of a Following Vehicle The presence of a following vehicle has the potential to make left-turn drivers accept a gap as soon as possible so as not to prolong the wait time of the driver behind them. Table 4.14 shows the percentage of all events (rejected and accepted gaps)
|
From page 52... ...
52 the distribution of postencroachment time for accepted gaps by the posted speed limit of the opposing approach. The plots do not indicate a decreasing trend in the times drivers leave between the execution of their turning maneuver and the arrival of the next opposing vehicle and decreasing posted speed limit on the opposing approach increases.
|
From page 53... ...
53 Two-Way Stop-Controlled Intersections Red dot = mean; Horizontal line = median; Colored box = mid 50% of data; Blue circle = extreme value 30 35 40 45 50 55 0 5 10 15 20 25 Po st -e nc ro ac hm en t t im e (se c) fo r a cc ep ted ga ps Opposing Approach Posted Speed Limit (mph)
|
From page 54... ...
54 Summary of results The analysis found that drivers accept longer gaps at intersections with negative offset than with positive offset. This makes intuitive sense, given that the distance drivers must travel to complete the left-turn maneuver is longer at intersections with wider medians and negative-offset left-turn lanes.
|
From page 55... ...
55 was found. For most of the offset categories, the presence of a following driver resulted in a lower average accepted gap length and a lower median accepted gap length; however, sample sizes were small.
|
Key Terms
This material may be derived from roughly machine-read images, and so is provided only to facilitate research.
More
information on Chapter Skim is available.