CHAPTER 2: COMPARISON OF THE RESULTS FROM THE ANALYSES CONDUCTED WITH THE 100-Car Data and the LOCAL/SHORT HAUL AND SLEEPER BERTH DATA

Recall in Chapter 1 that analyses were conducted with critical incident data that were collected during the 100-Car Study (Dingus et al., 2004). The method of analysis used in Chapter 1 is almost identical to the approach used in Hanowski, Keisler, and Wierwille (2004). The Hanowski, Keisler, and Wierwille (2004) study assessed LV-HV interactions from the perspective of local/short-haul (L/SH) and sleeper berth (SB) drivers. A total of 142 LV-HV interactions were identified in the L/SH study (see Hanowski, Wierwille, Garness, and Dingus, 2000, for a complete description of the L/SH study). Of these, 117 (82.4%) incidents were judged to have been the fault of the LV driver, while the remaining 25 (17.6%) incidents were the fault of the HV driver (remember that incidents in the L/SH and SB studies used "initiate" to indicate fault. From this point on, "at-fault" will be used rather than "initiate").

In the SB study, a total of 68 LV-HV interactions were identified (see Dingus et al., 2002, for a complete description of the SB study). Of these, 47 (69.1%) were assessed to have been the fault of the LV driver, while the remaining 21 (38.9%) were the fault of the HV driver. Taken together with the current research, these three studies consistently show that LV drivers appear to be responsible for the majority of LV-HV interactions. Of the 427 LV-HV incidents identified across the three studies (excluding the 29 Unknown at-fault incidents in the current study), 302 (70.7%) were the fault of the LV driver, while the remaining 125 (29.3%) were the fault of the HV driver (a 2.4:1 ratio).

However, one of the weaknesses in the current study was the lack of instrumentation in HVs. Conversely, one of the weaknesses in the Hanowski, Keisler, and Wierwille (2004) study was the lack of instrumentation in LVs. Thus, by integrating the results from the present study with the results from the Hanowski, Keisler, and Wierwille (2004) study, it is believed that a more complete understanding of LV-HV interaction problem can be gained. Chapter 2 has two primary aims: combine the data from the current study with the data from the L/SH and SB studies in Hanowski, Keisler, and Wierwille (2004) to illustrate the overall LV-HV interaction picture, and address the limitation of having only one vehicle instrumented by assessing the differences between the three studies (i.e., are the recorded LV-HV interactions fundamentally different as a function of which vehicle is instrumented?).

Incident Types

Table 34 illustrates the frequency, percentage, and rank ordering of Incident Types across all three studies. As can be seen in Table 34, there were a total of 456 LV-HV interactions across the three studies. The most frequent Incident Type was Lane Change Without Sufficient Gap (22.1%), followed by Late Braking for Stopped/Stopping Traffic (17.3%) and Roadway Entrance Without Clearance (7.9%). Figure 34 shows a bar graph of the 456 incidents, across the three studies, as a function of Incident Type.

Table 34. Frequency, Percentage, and Rank Ordering of the Incident Types

Across the Three Studies.

Figure 34. Percentage of Incident Types Across all Three Studies.

Tables 35-37 show the frequency and percentage of each Incident Type in the 100-Car, SB, and L/SH studies, respectively. In the 100-Car Study, the most frequent Incident Types were Late Braking for Stopped/Stopping Traffic (26.8%), Lane Change Without Sufficient Gap (22%), and Lateral Deviation of Through Vehicle (8.1%) . In the SB study, the most frequent Incident Types were Lane Change Without Sufficient Gap (25%), Turn Without Sufficient Warning (17.6%), Late Braking for Stopped/Stopping Traffic (14.7%). In the L/SH study, the most frequent Incident Types were Lane Change Without Sufficient Gap (21.1%), Roadway Entrance Without Clearance (14.3%), and Left Turn Without Clearance (14.8%). Figure 35 shows a bar graph that illustrates the percentage of incidents in the 100-Car, SB, and L/SH studies as a function of the Incident Types.

Table 35. Frequency and Percentage of Incident Types for the 100-Car Study.

Table 36. Frequency and Percentage of Incident Types for the Sleeper Berth Study.

Table 37. Frequency and Percentage of Incident Types for the Local/Short Haul Study.

Figure 35 . Percentage of Incident Types in the 100-Car, SB, L/SH Studies.

Figure 36 shows the percentage of HV driver at-fault incidents in the 100-Car, SB, and L/SH studies as a function of Incident Type. The black bars in Figure 36 represent HV driver at-fault incidents in the 100-Car data set, while the white and grey bars represent SB and L/SH driver at-fault incidents, respectively. The most frequent Incident Types for HV driver at-fault incidents in the 100-Car Study were Lane Change Without Sufficient Gap (26.6%), Lateral Deviation of Through Vehicle (21.5%), and Left Turn Without Clearance (13.9%). The most frequent Incident Types for HV driver at-fault incidents in the SB study were Late Braking for Stopped/Stopping Traffic (47.6%), Following Too Close (23.8%), and Turn Without Sufficient Warning (14.3%). The most frequent Incident Types for HV driver at-fault incidents in the L/SH study were Roadway Entrance Without Clearance (20%), Backing In Roadway (12%), Late Braking for Stopped/Stopping Traffic (12%), Wide Turn Into Adjacent Lane (12%), and Low Speed (8%).

).

Figure 36 . Percentage of HV Driver At-Fault Incident Types in the 100-Car, SB, and L/SH Studies.

Figure 37 shows the percentage of LV driver at-fault incidents in the 100-Car, SB, and L/SH studies as a function of Incident Type. The black bars in Figure 37 represent the 100-Car driver at-fault incidents in the 100-Car Study, while the white and grey bars represent LV driver at-fault incidents in the SB and L/SH data sets, respectively. The most frequent Incident Types for LV driver at-fault incidents in the 100-Car Study were Late Braking for Stopped/Stopping Traffic (41.3%) Lane Change Without Sufficient Gap (21.7%), and Aborted Lane Change (8%). The most frequent Incident Types for LV driver at-fault incidents in the SB study were Lane Change Without Sufficient Gap (32%), Turn Without Sufficient Warning (19.1%), and Low Speed (17%). The most frequent Incident Types for LV driver at-fault incidents in the L/SH study were Lane Change Without Sufficient Gap (24.8%), Roadway Entrance Without Clearance (17.9%), Turn Without Clearance (17.9%), and Improper Passing (5.1%).

Figure 37 . Percentage of LV Driver At-Fault Incident Types in the 100-Car, SB, and L/SH Studies.

As can be seen in Figure 37, the Incident Types differed markedly depending on the study (i.e., 100-Car, SB, and L/SH). One possible explanation for these discrepancies could be the frequency of travel on certain Road Types (i.e., it would be expected that a significant amount of the incidents would occur on the most traveled roadways, similar to exposure). Of course, this is an oversimplification. The geographical area, what the vehicle was used for, and driver preference all dictate the predominant Road Type used in driving. Figure 38 illustrates the percentage of incidents for each of the three studies as a function of Road Type. The black bars in Figure 38 represent 100-Car incidents, while the white and grey bars represent SB and L/SH incidents, respectively. In the 100-Car data set, the highest proportion of incidents occurred on the Urban Divided (60.2%), Urban Undivided (18.7%), and Rural Undivided (9.3%) roads. In the SB data set, the highest proportion of incidents occurred on the Rural Divided (55.9%), Urban Undivided (10.3%), and Rural Undivided (13.2%) roads. In the L/SH data set, the highest proportion of incidents occurred on the Rural Divided (38.7%), Urban Undivided (18.3%), and Rural Undivided (16.2%) roads.

Figure 38 . Percent of Incidents in the 100-Car, SB, and L/SH studies as a Function of Road Type.

Summary of Incident Types

Across the three studies, the most frequent Incident Types were Lane Change Without Sufficient Gap (22.1%), Late Braking for Stopped/Stopping Traffic, (17.3%), and Roadway Entrance Without Clearance (7.9%). These three Incident Types accounted for 47.3% of LV-HV interactions across the three studies.

As can be seen in Figure 36, the HV driver at-fault Incident Types differed with respect to the instrumented vehicle. The most frequent Incident Types for HV driver at-fault incidents in the 100-Car Study were Lane Change Without Sufficient Gap (26.6%), Lateral Deviation of Through Traffic (21.5%), and Left Turn Without Clearance (13.9%). These three Incident Types represented 62% of the HV driver at-fault incidents in the 100-Car Study. The most frequent Incident Types for HV driver at-fault incidents in the SB study were Late Braking for Stopped/Stopping Traffic (47.6%), Following Too Closely (23.8%), and Turn Without Sufficient Warning (14.3%). These three Incident Types represented 85.7% of the HV driver at-fault incidents in the SB study. The most frequent Incident Types for the HV driver at-fault incidents in the L/SH study were Roadway Entrance Without Clearance (20%), Wide Turn Into Adjacent Area (12%), and Backing in Roadway (12%). These three Incident Types represented 48% of HV driver at-fault incident in the L/SH study.

As can be seen in Figure 37, the LV driver at-fault Incident Types also differed with respect to the instrumented vehicle. The most frequent Incident Types for LV driver at-fault incidents in the 100-Car Study were Late Braking for Stopped/Stopping Traffic (41.3%) and Lane Change Without Sufficient Gap (21.7%). These two Incident Types represent 63% of the LV driver at-fault incidents in the 100-Car Study. The most frequent Incident Types for LV driver at-fault incidents in the SB study were Lane Change Without Sufficient Gap (31.9%), Turn Without Sufficient Warning (19.1%), and Low Speed (17%). These three Incident Types represent 68% of the LV driver at-fault incidents in the SB study. The most frequent Incident Types for the LV driver at-fault incidents in the L/SH study were Lane Change Without Sufficient Gap (24.8%), Roadway Entrance Without Clearance (17.9%), and Left Turn Without Clearance (17.9%). These three Incident Types represent 60.6% of LV driver at-fault incident in the L/SH study.

There were many differences across the three studies (100-Car, L/SH, and SB) as well between HVs and LVs within each study. This is possible as both the SB and L/SH trucks were instrumented in their respective studies, while the LV was instrumented in the 100-Car Study. Thus, those incidents recorded in the 100-Car Study reflect a diverse range of HVs. Further, the incidents in the SB and L/SH studies are likely to reflect difficulties specific to SB and L/SH operations. Support for this hypothesis was found when the authors assessed the location of each incident (see Figure 38).

Consider the Road Type comparison data shown in Figure 38. The bar graph illustrates the percentage of incidents as a function of different Road Types. Not surprisingly, the Road Types frequented by the 100-Car participants, such as major city roads and streets, are where the majority of 100-Car Study incidents occurred (60.2% of the incidents occurred on an Urban Divided road). The Road Types common to SB operations, such as interstates and highways, are where the majority of SB incidents occurred. That is, rural divided by median (i.e., interstate) and urban divided by median (i.e., highway) accounted for 74% of the SB incidents. Similarly, the Road Types common for L/SH trucks accounted for many of the L/SH incidents. On a percentage basis, there were more incidents for L/SH drivers in town settings (i.e., urban undivided, urban divided, one-way, and parking lot), which is where many L/SH delivery routes are located. By looking at the driving environments, it could be said that the majority of 100-Car and L/SH incidents occurred in and around town/urban areas (lower speeds and higher traffic areas). SB incidents, on the other hand, tended to occur on highways where speeds are relatively high and traffic density is relatively low.

For example, Roadway Entrance Without Clearance accounted for a substantial portion of L/SH incidents (18.3%), whereas this Incident Type only accounted for a small proportion of the incidents in the 100-Car (2.4%) and SB (5.9%) studies. This makes intuitive sense when one considers L/SH operations trucking operations. L/SH drivers have many deliveries during their workday and will, therefore, routinely exit parking lots onto roadways. This provides an opportunity for this particular Incident Type to occur, whereas this maneuver is not characteristic of 100-Car and SB drivers. Much of the time, SB drivers were on limited-access highways having no intersecting side roads, whereas 100-Car drivers were driving on major urban road going to and from their residence. The characteristics of the Road Types traveled by 100-Car, SB, and L/SH drivers appear to explain some of the discrepancies between the three studies.

Primary Maneuvers

Table 38 shows the frequency, percentage, and rank ordering of the Primary Maneuvers across the 100-Car, SB, and L/SH studies. The most frequent Primary Maneuver across all three studies was Changing Lanes (23.2%), followed by Braking (12.3%) and Left Turn (11.2%). Figure 39 shows a bar graph of the 456 incidents, across the three studies, as a function of Primary Maneuver.

Table 38. Frequency, percentage, and rank ordering of Primary Maneuvers across the three studies.

Figure 39 . Frequency of Primary Maneuvers Across the Three Studies.

Tables 39-41 show the frequency and percentage of each Primary Maneuver in the 100-Car, SB, and L/SH studies, respectively. In the 100-Car Study, the most frequent Primary Maneuvers were Braking (22.8%), Changing Lanes (21.1%), and Stopped (15%) . In the SB study, the most frequent Primary Maneuvers were Through Traffic (39.7%), Changing Lanes (32.4%), Roadway Exit (8.8%), and Left Turn (8.8%). In the L/SH study, the most frequent Primary Maneuvers were Changing Lanes (22.5%), Left Turn (20.4%), and Through Traffic (19%). Figure 40 shows a bar graph illustrating the incidents for the 100-Car, SB, and L/SH studies as a function of Primary Maneuver.

Table 39. Frequency and Percentage of Primary Maneuvers for the 100-Car Study.

Table 40. Frequency and Percentage of Primary Maneuvers for the Sleeper Berth Study.

Table 41. Frequency and Percentage of Primary Maneuvers for the Local/Short Haul Study.

Figure 40 . Percentage of Primary Maneuvers in the 100-Car, SB, and L/SH Studies.

Figure 41 shows the percentage of HV driver at-fault incidents in the 100-Car, SB, and L/SH studies as a function of Primary Maneuver. The black bars in Figure 41 represent the HV driver at-fault incidents in the 100-Car Study, while the white and grey bars represent HV driver at-fault incidents in the SB and L/SH studies, respectively. The most frequent Primary Maneuvers for HV driver at-fault incidents in the 100-Car Study were Changing Lanes (32.9%), Crossing Over Lane Line (26.3%), and Left Turn (15.2%). The most frequent Primary Maneuvers for HV driver at-fault incidents in the SB study were Through Traffic (71.4%), Changing Lanes (9.5%), and Left Turn (9.5%). The most frequent Primary Maneuvers for HV driver at-fault incidents in the L/SH study were Through Traffic (44%), Right Turn (24%), and Left Turn (12%).

Figure 41 . Percentage of HV Driver At-Fault Incidents in the 100-Car, SB, and L/SH Studies as a Function of Primary Maneuver.

Figure 42 shows the percentage of LV driver at-fault incidents in the 100-Car, SB, and L/SH studies as a function of Primary Maneuver. The black bars in Figure 42 represent the LV driver at-fault incidents in the 100-Car Study, while the white and grey bars represent LV driver at-fault incidents in the SB and L/SH studies, respectively. The most frequent Primary Maneuvers for LV driver at-fault incidents in the 100-Car Study were Braking (32.6%), Stopped (21.7%), and Changing Lanes (16.7%). The most frequent Primary Maneuvers for LV driver at-fault incidents in the SB study were Changing Lanes (42.6%), Through Traffic (25.5%), and Roadway Exit (10.6%). The most frequent Primary Maneuvers for LV driver at-fault incidents in the L/SH study were Changing Lanes (26.5%), Left Turn (22.2%), and Through Traffic (13.7%).

Figure 42 . Percentage of LV Driver At-Fault Incidents in the 100-Car, SB, and L/SH Studies as a Function of Primary Maneuver.

Summary of Primary Maneuvers Across Studies

Across the three studies, the most frequent Primary Maneuvers were Changing Lanes (23.2%), Braking (12.3%), and Left Turn (11.2%). These three Primary Maneuvers represented 46.5% of the Primary Maneuvers across the three studies. As can be seen in Figures 40-42, the Primary Maneuvers differed depending on whether the LV or HV was judged to have been at fault.

As can be seen in Figure 41, the most frequent Primary Maneuvers for HV driver at-fault incidents in the 100-Car Study were Changing Lanes (32.9%), Crossing Over the Lane Line (26.3%), and Left Turn (15.2%). These three Primary Maneuvers represented 74.4% of the HV driver at-fault incidents in the 100-Car Study. In the SB study, the most frequent Primary Maneuvers for HV driver at-fault incidents were Through Traffic (71.4%), Changing Lanes (9.5%), and Left Turn (9.5%). These three Primary Maneuvers accounted for 90.4% of the HV driver at-fault incident in the SB study. In the L/SH study, the most frequent Primary Maneuvers for HV driver at-fault incidents were Through Traffic (44%), Right Turn (24%), and Left Turn (12%). These three Primary Maneuvers accounted for 80% of the HV driver at-fault incidents in the L/SH study. From the data it appears that HV drivers had the most difficulties when traveling forward on a roadway or straight through an intersection and changing lanes.

As can be seen in Figure 42, the most frequent Primary Maneuvers for LV driver at-fault incidents in the 100-Car Study were Braking (32.6%), Stopped (21.7%), and Changing Lanes (16.7%). These three Primary Maneuvers accounted for 71% of the LV driver at-fault incidents in the 100-Car Study. In the SB study, the most frequent Primary Maneuvers for LV driver at-fault incidents were Changing Lanes (42.6%), Through Traffic (25.5%), and Roadway Exit (10.6%). These three Primary Maneuvers represented 78.7% of the LV driver at-fault incidents in the SB study. In the L/SH study, the most frequent Primary Maneuvers for LV driver at-fault incidents were Changing Lanes (26.5%), Left Turn (22.2%), and Through Traffic (13.7%). These three Primary Maneuvers accounted for 62.4% of the LV driver at-fault incidents in the L/SH study. From the data it appears that LV drivers had the most difficulties when braking or decelerating and changing lanes.

In the SB study, the Through Traffic (i.e., vehicle traveling forward on a roadway or straight through an intersection) and Changing Lanes Primary Maneuvers accounted for 72% of the total incidents recorded. The Primary Maneuvers associated with LV-HV interactions in the L/SH and 100-Car studies were more varied. Changing Lanes (22.5%), Left Turn (20.4%), Through Traffic (19%), Right Turn (12.7%), and Merge onto Roadway (10.6%) accounted for the majority of incidents recorded in the L/SH study, while Braking (22.8%), Changing Lanes (21.1%), Stopped (15%), and Crossing Over Lane Line (7.7%) represented the majority of Primary Maneuvers in the 100-Car Study. Further investigation of these maneuver types indicated that Through Primary Maneuver was the most frequent type for HVs in both the SB and L/SH studies: 71% and 44%, respectively. However, Changing Lanes (32.9%) was the most predominant Primary Maneuver for HVs in the 100-Car Study.

The Changing Lanes Primary Maneuver was the predominant type for LVs in both the SB and L/SH studies: 42.6% and 26.5%, respectively. However, in the 100-Car Study, the Braking (32.6%) Primary Maneuver was the most predominant type for LVs. Note that the predominant Primary Maneuvers for each group of drivers is consistent with the Incident Type classification presented in the previous section. For example, one would expect that the Primary Maneuver for Late Braking for Stopped/Stopping Traffic would be Through Traffic, which as indicated by the SB at-fault incidents, was the case.

Specific Contributing Factors

Table 42 shows the frequency, percentage, and rank ordering of the Contributing Factors across all three studies. The most frequent Contributing Factor across the three studies was Driving Techniques (41%), followed by Aggressive Driving (24.1%), Unknown (17.1%), and Distracted (10.5%). The reader should bear in mind that more than one Contributing Factor could be selected for a single incident in the 100-Car Study. Figure 43 shows a bar graph of the 456 incidents, across the three studies, as a function of the Contributing Factors.

Table 42. Frequency, Percentage, and Rank Ordering of the Contributing Factors across the 100-Car, SB, and L/SH Studies (n Total = 456).

Figure 43 . Frequency of Contributing Factors across the 100-Car, SB, and L/SH Studies.

Tables 43-45 show the frequency and percentage of each Contributing Factor in the 100-Car, SB, and L/SH studies, respectively. In the 100-Car Study, the most frequent Contributing Factors were Driving Techniques (49.5%), Unknown (24%), Distracted (18.7%), and Aggressive Driving (15%) . In the SB data set, the most frequent Contributing Factors were Driving Techniques (46%) and Aggressive Driving (35%). In the L/SH study, the most frequent Contributing Factors were Aggressive Driving (37%), Driving Techniques (24%), Roadway Alignment (11%), and Unknown (11%). Figure 44 shows a bar graph illustrating the 456 incidents in the 100-Car, SB, and L/SH studies as a function of the Contributing Factors.

Table 43. Frequency and Percentage of Contributing Factors for the 100-Car Study.

Table 44. Frequency and Percentage of Contributing Factors for the Sleeper Berth Study.

Table 45. Frequency and Percentage of Contributing Factors for the Local/Short Haul Study.

Figure 44 . Percentage of Incidents in the 100-Car, SB, and L/SH Studies as a Function of Contributing Factors.

Figure 45 shows the percentage of HV driver at-fault incidents for the 100-Car, SB, and L/SH studies as a function of Contributing Factor. The black bars in Figure 45 represent the HV driver at-fault incidents in the 100-Car Study, while the white and grey bars represent HV driver at-fault incidents in the SB and L/SH studies, respectively. In the 100-Car Study, the most frequent Contributing Factors were Unknown (68.4%), Driving Techniques (15.2%), and Distracted (11.4%). In the SB study, the most frequent Contributing Factors were Driving Techniques (52%), Aggressive Driving (24%), and Vehicle Kinematics, Physics (14%). In the L/SH study, the most frequent Contributing Factors for HV driver at-fault incidents were Driving Techniques (32%), Roadway Alignment (28%), and Aggressive Driving (12%).

Figure 45 . Percentage of HV At-Fault Incidents in the 100-Car, SB, and L/SH Studies as a Function of Contributing Factors for LV Drivers.

Figure 46 shows the percentage of LV driver at-fault incidents in the 100-Car, SB, and L/SH studies as a function of Contributing Factor. The black bars in Figure 46 represent the LV driver at-fault incidents in the 100-Car Study, while the white and grey bars represent LV driver at-fault incidents in the SB and L/SH studies, respectively. In the 100-Car Study, the most frequent Contributing Factors for LV driver at-fault incidents were Driving Techniques (70.3%), Distracted (22.5%), and Aggressive Driving (22.5%). In the SB study, the most frequent Contributing Factors for LV driver at-fault incidents were Driving Techniques (42.6%) and Aggressive Driving (40.4%). In the L/SH study, the most frequent Contributing Factors for LV driver at-fault incidents were Aggressive Driving (42.7%), Driving Techniques (22.2%), and Unknown (12.8%).

Figure 46 . Percentage of LV Driver At-Fault Incidents in the 100-Car, SB, and L/SH Studies as a Function of Contributing Factor.

Summary of Contributing Factors Across Studies

Across the three studies, the most frequent Contributing Factors were Driving Techniques (41%), Aggressive Driving (24.1%), and Unknown (17.1%). As can be seen in Figures 44-46, the Contributing Factors were relatively the same depending on whether the LV or HV was at-fault.

As can be seen in Figure 45, the most frequent Contributing Factors for HV driver at-fault incidents in the 100-Car Study were Unknown (68.4%), Driving Techniques (15.2%), and Distracted (11.4%). In the SB study, the most frequent Contributing Factors were Driving Techniques (52%), Aggressive Driving (24%), and Vehicle Kinematics, Physics (14%). In the L/SH study, the most frequent Contributing Factors were Driving Techniques (32%), Roadway Alignment (28%), and Aggressive Driving (12%). From the data it appears that the majority of at-fault HV drivers were classified with poor driving techniques.

It is important to note that the HV driver at-fault Contributing Factors in both the SB and L/SH studies are relatively the same, yet, different from the HV driver at-fault Contributing Factors in the 100-Car Study. The reader should bear in mind that the Contributing Factor was coded with respect to the instrumented vehicle, thus, the high frequency of Unknown Contributing Factors in the 100-Car Study. If a LV driver was not at-fault, it was unlikely they would be coded with a Contributing Factor in the LV-HV interaction (as supported by the data in the 100-Car Study).

As can be seen in Figure 46, the most frequent Contributing Factors for LV driver at-fault incidents in the 100-Car Study were Driving Techniques (70.3%), Distracted (22.5%), and Aggressive Driving (22.5%). In the SB study, the most frequent Contributing Factors for LV driver at-fault incidents were Aggressive Driving (42.6%) and Driving Techniques (40.4%). In the L/SH study, the most frequent Contributing Factors for LV driver at-fault incidents were Aggressive Driving (42.7%), Diving Techniques (22.2%), and Unknown (12.8%). From the data it appears that the majority of at-fault LV drivers were classified with poor driving techniques and aggressive driving.

Recall that Hankey et al. (1999) found that 77% of crashes in the Pennsylvania crash database from 1995-1996 were cited with "human error" as the primary factor in the crash. By adding the frequency of Contributing Factors associated with human error in the three studies, a total of 294 incidents out of 427, or 68.9%, of the LV-HV interactions had at least one human error as a Contributing Factor (excluding the Unknown at-fault incident in the 100-Car Study).

Yet, there were differences when only HV and LV driver at-fault incidents were considered. For example, 35.2% of the HV driver at-fault incidents across the 100-Car, SB, and L/SH studies had at least one human error Contributing Factor (for either driver), while 82.8% of the LV driver at-fault incidents across the 100-Car, SB, and L/SH studies had at least one human error Contributing Factor. Thus, LV driver at-fault incidents were found to be similar to what Hankey et al. (1999) found when they analyzed the Pennsylvania crash database. But, it also suggests that at-fault LV drivers were more likely than at-fault HV drivers to be coded with "human error" as a Contributing Factor in the LV-HV interactions. As the majority of crashes involve LVs rather than HVs (NHTSA, 2004), it is not all that surprising that LV drivers were more similar than HV drivers when comparing the results of the 100-Car, SB, and L/SH studies to the Hankey et al. (1999) study.

However, the Hankey et al. (199) study looked at all types of crashes. Stuster (1999) assessed driver-related factors in only LV-HV fatal crashes and found that 67.7% of LVs were cited with the driver-related factors similar to the Willful Behavior and Driver Proficiency Contributing Factor categories related to human error, while 23.8% of HVs were cited with the same driver-related factors. These results have a similar ratio to the LV (84.1%) and HV (33.6%) at-fault incidents attributed to human error across the 100-Car, SB, and L/SH studies.

Hankey et al. (1999) also conducted analysis with the 1996 FARS database. They found that 31.1% of the fatal crashes involved aggressive driving as a primary factor. This is relatively similar (within 10%) to what was found across the 100-Car, SB, and L/SH studies, where 24.1% of the LV-HV interactions were coded with the Aggressive Driving Contributing Factors.

Across the 100-Car, SB, and L/SH studies it was found that 33.1% of the LV at-fault incidents were coded with the Aggressive Driving Contributing Factor, while 7.2% of the HV driver at-fault incidents were coded with the same Contributing Factor (excluding the Unknown at-fault incidents). Stuster's (1999) analysis of fatal LV-HV interactions found that 2.1% of the HVs were cited with the driver-related factor "Erratic/Reckless Driving," while 4.3% of the LVs were cited with this same driver-related factor. Thus, when comparing the same type of event (i.e., LV-HV interaction), keeping in mind that the events were different in intensity, the current analysis resulted in different results from the Stuster (1999) findings.

Discussion of the Results Comparing the 100-car, SB, and L/SH Studies

The primary aims of Chapter 2 were to combine the data from the current study with the data from the L/SH and SB studies in Hanowski, Keisler, and Wierwille (2004) to illustrate the overall crash picture, and address the limitation of having only one vehicle instrumented by assessing the differences between the three studies (i.e., are LV-HV interactions fundamentally different as a function of which vehicle is instrumented?). The results from the 100-Car Study were presented in Chapter 1, while the results from both the SB and L/SH studies can be found in Hanowski, Keisler, and Wierwille (2004).

In the 100-Car Study, a total of 246 LV-HV interactions were identified. Of these, 138 (56.1%) and 79 (32.1%) incidents were found to be the fault of the LV and HV drivers, respectively. In the remaining 29 (11.8%) incidents it was unknown whether the LV or HV driver was at-fault. In the L/SH study, a total of 142 LV-HV interactions were identified. Of these, 117 (82.4%) incidents were the fault of the LV driver, while the remaining 25 (17.6%) incidents were the fault of the HV driver. In the SB study, a total of 68 LV-HV interactions were identified. Of these, 47 (69.1%) were the fault of the LV driver, while the remaining 21 (38.9%) were the fault of the HV driver. Taken together with the current research, these three studies consistently show that LV drivers appear to be responsible for the majority of LV-HV interactions. Of the 427 LV-HV incidents identified across the three studies (excluding the 29 Unknown at-fault incidents in the current study), 302 (70.7%) were judged to have been the fault of the LV driver, while the remaining 125 (29.3%) were the fault of the HV driver (a 2.4:1 ratio). The high ratio presented here emphasizes the role that LV drivers play in LV-HV interaction incidents. Given that LV drivers were more likely to have initiated an incident, it is believed that efforts at addressing the LV-HV interaction problem should include focusing on the LV driver.

There were a number of interesting findings from the comparisons between the 100-Car, SB, and L/SH studies. Comparisons were conducted with respect to the Incident Type, Primary Maneuver, and Contributing Factor. The Incident Type comparison indicated that Lane Change Without Sufficient Gap was the most frequent Incident Type across all three studies. A breakdown of incidents as a function of the at-fault driver showed that Lane Change Without Sufficient Gap incidents were primarily attributed to LV drivers. Critical incidents that involved a LV driver changing lanes in front of an HV, leaving the HV driver with very little headway between vehicles, were a common Incident Type that was captured in all three studies.

While the Incident Types for the LV driver at-fault incidents shared some similarities across the three studies, the Incident Types for the HV driver at-fault incident were more varied across the studies. In the 100-Car Study, 48.1% of the HV driver at-fault Incident Types included Lane Change Without Sufficient Gap and Lateral Deviation of Through Traffic. In the SB study, 71.4% of the HV driver at-fault incidents included Late Braking for Stopped/Stopping Traffic and Following Too Closely. In the L/SH study, 48% of the Incident Types included Roadway Entrance Without Clearance, Wide Turn Into Adjacent Lane, and Late Braking for Stopped/Stopping Traffic. One possible explanation for these differences was the predominant Road Type traveled (see Figure 38), as well as the type of trucking operations included in the SB and L/SH studies. It could be argued that the HVs in the 100-Car Study represent a more diverse population of HVs since they were not limited to L/SH and SB trucks. In fact, as shown in Table 3 (page 15), 25 different HVs were identified as being involved in LV-HV interactions in the 100-Car Study. Thus, it is likely the results for at-fault HV drivers in the 100-Car Study might be more representative of HV drivers in general, while the results for HV drivers in the SB and L/SH studies are more representative of drivers in those specific operations.

The Primary Maneuver comparison indicated that Changing Lanes was the most frequent Primary Maneuver across all three studies. A breakdown of incidents as a function of the at-fault driver showed that Through Traffic incidents were primarily attributed to HV drivers. Critical incidents that involved an HV driver traveling forward on the roadway or straight though an intersection were a common Primary Maneuver in the three studies. However, while Through Traffic was the most frequent Primary Maneuver for HV driver at-fault incidents in both the SB and L/SH studies, the most frequent Primary Maneuvers for HV driver at-fault incidents in the 100-Car Study were Changing Lanes and Crossing Over the Lane Line. This suggests that, for the general population of HV drivers, changing lanes and crossing the lane line are difficult maneuvers. This make intuitive sense, as HVs are likely to have blind spots that make it difficult to change lanes or attempt to change a lane.

The most predominant Primary Maneuver for LV driver at-fault incidents was Changing Lanes. While in all three studies the LV driver was likely to be coded with the Changing Lanes Primary Maneuver, there were also differences across the three studies. LV drivers in the 100-Car Study also had difficulties when they were braking or stopped. LV drivers in the SB study encountered difficulties in through traffic, while LV drivers in the L/SH study had difficulties when they were making left turns.

The Contributing Factors category allows researchers to describe why the incident occurred. The most frequent Contributing Factor across the three studies was Driving Techniques. A breakdown of incidents as a function of the at-fault driver showed that Driving Techniques were primarily attributed to HV drivers. Thus, when the Contributing Factor was known, this was the most frequent Contributing Factor for HV driver at-fault incidents in each of the studies.

The most frequent Contributing Factors for LV driver at-fault incidents across the three studies were Driving Techniques and Aggressive Driving. These two Contributing Factors accounted for a substantial number of the LV driver at-fault incidents across the three studies. However, a large proportion of the LV driver at-fault incidents in the 100-Car Study involved the Distracted Contributing Factor. In fact, the only time a LV driver at-fault incident was coded with the Distracted Contributing Factors was in the 100-Car Study. This is almost certainly due to the fact that the LVs in the 100-Car Study were instrumented (thereby allowing analysis of the LV drivers' behaviors while driving), while the LVs in both the SB and L/SH studies were not instrumented.

Summary

The results of the current study in conjunction with Hanowski, Keisler, and Wierwille (2004) indicated that LV-HV interactions represent a serious problem. While there were several differences across the three studies, the results consistently showed that LV drivers are more likely to be responsible for the LV-HV interaction than HV drivers. It is believed that the results from the 100-Car, SB, and L/SH studies provide a more complete description of the LV-HV interaction picture. Further, the comparisons among these three studies address the limitations of not having LVs and HVs instrumented. The detailed analyses that were conducted provide insight into how this problem might be addressed. Listed below are several suggestions that should be considered for reducing LV-HV interactions:

• Addressing the LV-HV interaction problem should focus on the driving behaviors of the LV driver. The LV driver was at-fault in 70.7% of the LV-HV interaction incidents recorded across the three studies.
• The primary area for LV that should be addressed involves their driving techniques and aggressive driving behaviors. Also, the instrumented LVs in the 100-Car Study showed that distraction was a significant problem in LV-HV interactions. Thus, the three studies identified three areas for LV drivers that should be targeted: distraction, aggressive driving, and driving techniques. Also, distraction, particularly from cell phones, appears to be a much bigger problem then has been reported in previous studies (e.g., Stutts et al., 2003).
• The primary area for HV drivers that should be addressed involves driving techniques. One possible method of addressing this is through improved truck driver training programs. For example, consideration should be given to ongoing (e.g., yearly) training courses. Given the high incidence of Aggressive Driving on the part of LV drivers, one of the primary areas of focus for a truck driving training program should be on defensive driving and hazard identification.
• Infrastructure was found to play a role in HV driver at-fault incidents in the L/SH study. Drivers and/or company dispatchers should be cognizant of problematic sections of routes, and avoid such locations to the greatest extent possible.
• Technology has progressed to the point where it is possible to collect data on almost any driving-related variable. I n situ data collection is one way to study a wide range of safety-related issues in a naturalistic environment. The video and performance/behavior data collected from the 100-Car, SB, and L/SH studies have been archived and provide a rich source of information that can be used for studying critical incidents, as was the case in the current effort, or other issues that might be identified at a later time.