| 1. |
- Bálint, András, 1982, et al.
(författare)
-
Multitasking additional-to-driving: Prevalence, structure, and associated risk in SHRP2 naturalistic driving data
- 2020
-
Ingår i: Accident Analysis and Prevention. - : Elsevier BV. - 0001-4575. ; 137
-
Tidskriftsartikel (refereegranskat)abstract
- Objective This paper 1) analyzes the extent to which drivers engage in multitasking additional-to-driving (MAD) under various conditions, 2) specifies odds ratios (ORs) of crashing associated with MAD, and 3) explores the structure of MAD. Methods Data from the Second Strategic Highway Research Program Naturalistic Driving Study (SHRP2 NDS) was analyzed to quantify the prevalence of MAD in normal driving as well as in safety-critical events of various severity level and compute point estimates and confidence intervals for the corresponding odds ratios estimating the risk associated with MAD compared to no task engagement. Sensitivity analysis in which secondary tasks were re-defined by grouping similar tasks was performed to investigate the extent to which ORs are affected by the specific task definitions in SHRP2. A novel visual representation of multitasking was developed to show which secondary tasks co-occur frequently and which ones do not. Results MAD occurs in 11 % of control driving segments, 22 % of crashes and near-crashes (CNC), 26 % of Level 1–3 crashes and 39 % of rear-end striking crashes, and 9 %, 16 %, 17 % and 28 % respectively for the same event types if MAD is defined in terms of general task groups. The most common co-occurrences of secondary tasks vary substantially among event types; for example, “Passenger in adjacent seat – interaction” and “Other non-specific internal eye glance” tend to co-occur in CNC but tend not to co-occur in control driving segments. The odds ratios of MAD using SHRP2 task definitions compared to driving without any secondary task and the corresponding 95 % confidence intervals are 2.38 (2.17–2.61) for CNC, 3.72 (3.11–4.45) for Level 1–3 crashes and 8.48 (5.11–14.07) for rear-end striking crashes. The corresponding ORs using general task groups to define MAD are slightly lower at 2.00 (1.80–2.21) for CNC, 3.03 (2.48–3.69) for Level 1–3 crashes and 6.94 (4.04–11.94) for rear-end striking crashes. Conclusions The number of secondary tasks that the drivers were engaged in differs substantially for different event types. A graphical representation was presented that allows mapping task prevalence and co-occurrence within an event type as well as a comparison between different event types. The ORs of MAD indicate an elevated risk for all safety-critical events, with the greatest increase in the risk of rear-end striking crashes. The results are similar independently of whether secondary tasks are defined according to SHRP2 or general task groups. The results confirm that the reduction of driving performance from MAD observed in simulator studies is manifested in real-world crashes as well.
|
|
| 2. |
- Bärgman, Jonas, 1972, et al.
(författare)
-
How does glance behavior influence crash and injury risk? A ‘what-if’ counterfactual simulation using crashes and near-crashes from SHRP2
- 2015
-
Ingår i: Transportation Research Part F: Traffic Psychology and Behaviour. - : Elsevier BV. - 1369-8478. ; 35, s. 152-169
-
Tidskriftsartikel (refereegranskat)abstract
- As naturalistic driving data become increasingly available, new analyses are revealing the significance of drivers’ glance behavior in traffic crashes. Due to the rarity of crashes, even in the largest naturalistic datasets, near-crashes are often included in the analyses and used as surrogates for crashes. However, to date we lack a method to assess the extent to which driver glance behavior influences crash and injury risk across both crashes and near-crashes. This paper presents a novel method for estimating crash and injury risk from off-road glance behavior for crashes and near-crashes alike; this method can also be used to evaluate the safety impact of secondary tasks (such as tuning the radio). We apply a ‘what-if’ (counterfactual) simulation to 37 lead-vehicle crashes and 186 lead-vehicle near-crashes from lead-vehicle scenarios identified in the SHRP2 naturalistic driving data. The simulation combines the kinematics of the two conflicting vehicles with a model of driver glance behavior to estimate two probabilities: (1) that each event becomes a crash, and (2) that each event causes a specific level of injury. The usefulness of the method is demonstrated by comparing the crash and injury risk of normal driving with the risks of driving while performing one of three secondary tasks: the Rockwell radio-tuning task and two hypothetical tasks. Alternative applications of the method and its metrics are also discussed. The method presented in this paper can guide the design of safer driver–vehicle interfaces by showing the best tradeoff between the percent of glances that are on-road, the distribution of off-road glances, and the total task time for different tasks.
|
|
| 3. |
- Dozza, Marco, 1978, et al.
(författare)
-
Real-world effects of using a phone while driving on lateral and longitudinal control of vehicles
- 2015
-
Ingår i: Journal of Safety Research. - : Elsevier BV. - 0022-4375. ; 55,, s. 81-87
-
Tidskriftsartikel (refereegranskat)abstract
- Technologies able to augment human communication, such as smartphones, are increasingly present during all daily activities. Their use while driving, in particular, is of great potential concern, because of the high risk that distraction poses during this activity. Current countermeasures to distraction from phone use are considerably different across countries and not always widely accepted/adopted by the drivers.This study utilized naturalistic driving data collected from 108 drivers in the Integrated Vehicle-Based Safety Systems (IVBSS) program in 2009 and 2010 to assess the extent to which using a phone changes lateral or longitudinal control of a vehicle. The IVBSS study included drivers from three age groups: 20-30 (younger), 40-50 (middle-aged), and 60-70 (older). Results from this study show that younger drivers are more likely to use a phone while driving than older and middle-aged drivers. Furthermore, younger drivers exhibited smaller safety margins while using a phone. Nevertheless, younger drivers did not experience more severe lateral/longitudinal threats than older and middle-aged drivers, probably because of faster reaction times. While manipulating the phone (i.e., dialing, texting) drivers exhibited larger lateral safety margins and experienced less severe lateral threats than while conversing on the phone. Finally, longitudinal threats were more critical soon after phone interaction, suggesting that drivers terminate phone interactions when driving becomes more demanding.These findings suggest that drivers are aware of the potential negative effect of phone use on their safety. This awareness guides their decision to engage/disengage in phone use and to increase safety margins (self-regulation). This compensatory behavior may be a natural countermeasure to distraction that is hard to measures in controlled studies. Intelligent systems able to amplify this natural compensatory behavior may become a widely accepted/adopted countermeasure to the potential distraction from phone operation while driving.
|
|
| 4. |
- Dozza, Marco, 1978, et al.
(författare)
-
Understanding driver self-regulating behavior: how does phone use influence vehicle control in real world?
- 2013
-
Ingår i: Proceedings of the 3rd Conference of Driver Distraction and Inattention, Gothenbrug, 4-6 September, 2013.
-
Konferensbidrag (övrigt vetenskapligt/konstnärligt)abstract
- In the last few years, research on driver distraction focused on assessing the extent to which usingmobile phones is compatible with safe driving. Many studies employing driving simulators suggestedthat using a phone while driving is definitely unsafe. Nevertheless, so far, naturalistic studies, as wellas aggregate crash statistics, did not match these results, keeping open the debate, and promisingguidelines for new countermeasures to distraction alternative to bans and based on understandingdriver behavior in the real world.Naturalistic data collected from 108 drivers in the Integrated Vehicle-Based Safety Systems (IVBSS)program in 2009 and 2010 was employed to determine how drivers change their vehicle controlwhen engaging in a conversation on, or manipulation of, a phone. Drivers were also divided intothree age groups 20-30 (young), 40-50 (middle-age), and 60-70 (older) to determine the possibleinteraction between age and phone use while driving on vehicle control.Using a phone for calling affected lateral control differently than manipulating a phone (as whiledialing or texting). However, no difference was found for longitudinal control. Young drivers used aphone while driving more often than older and middle-age drivers. In addition, young driversexhibited smaller safety margins while using a phone as well as faster reactions. Finally, the resultssuggest that drivers tend to interrupt phone interactions when the driving context becomes morecomplex.In conclusion, this study suggests that driver self-regulating behavior is the key to assess the netsafety effect of using a phone while driving. Consequently, countermeasures able to support thedrivers’ inherent self-regulating behavior may be a more successful, and more widely adoptedsolutions, than phone bans toward addressing the potential for distraction posed by phones whiledriving.
|
|
| 5. |
- Flannagan, Carol A., et al.
(författare)
-
Analysis of SHRP2 Data to Understand Normal and Abnormal Driving Behavior in Work Zones
- 2019
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- This research project used the Second Strategic Highway Research Program (SHRP2) Naturalistic Driving Study(NDS) to improve highway safety by using statistical descriptions of normal driving behavior to identify abnormal driving behaviors in work zones. SHRP2 data used in these analyses included 50 safety-critical events (SCEs) from work zones and 444 baseline events selected on a matched case-control design. Principal components analysis (PCA) was used to summarize kinematic data into “normal” and “abnormal”driving. Each second of driving is described by one point in three-dimensional principal component (PC) space;an ellipse containing the bulk of baseline points is considered “normal” driving. Driving segments without-of-ellipse points have a higher probability of being an SCE. Matched case-control analysis indicates that thespecific individual and traffic flow made approximately equal contributions to predicting out-of-ellipse driving. Structural Topics Modeling (STM) was used to analyze complex categorical data obtained from annotated videos.The STM method finds “words” representing categorical data variables that occur together in many events and describes these associations as “topics.” STM then associates topics with either baselines or SCEs. The STM produced 10 topics: 3 associated with SCEs, 5 associated with baselines, and 2 that were neutral. Distractionoccurs in both baselines and SCEs. Both approaches identify the role of individual drivers in producing situations where SCEs might arise. A countermeasure could use the PC calculation to indicate impending issues or specific drivers who may havehigher crash risk, but not to employ significant interventions such as automatically braking a vehicle without-of-ellipse driving patterns. STM results suggest communication to drivers or placing compliant vehicles in thetraffic stream would be effective. Finally, driver distraction in work zones should be discouraged.
|
|
| 6. |
- Flannagan, Carol. A. C., et al.
(författare)
-
Comparing motor-vehicle crash risk of EU and US vehicles
- 2018
-
Ingår i: Accident Analysis and Prevention. - : Elsevier Ltd. - 0001-4575 .- 1879-2057.
-
Tidskriftsartikel (refereegranskat)abstract
- This study examined the hypotheses that passenger vehicles meeting European Union (EU) safety standards have similar crashworthiness to United States (US) -regulated vehicles in the US driving environment, and vice versa. The first step involved identifying appropriate databases of US and EU crashes that include in-depth crash information, such as estimation of crash severity using Delta-V and injury outcome based on medical records. The next step was to harmonize variable definitions and sampling criteria so that the EU data could be combined and compared to the US data using the same or equivalent parameters. Logistic regression models of the risk of a Maximum injury according to the Abbreviated Injury Scale of 3 or greater, or fatality (MAIS3+F) in EU-regulated and US-regulated vehicles were constructed. The injury risk predictions of the EU model and the US model were each applied to both the US and EU standard crash populations. Frontal, near-side, and far-side crashes were analyzed together (termed "front/side crashes") and a separate model was developed for rollover crashes.For the front/side model applied to the US standard population, the mean estimated risk for the US-vehicle model is 0.035 (sd = 0.012), and the mean estimated risk for the EU-vehicle model is 0.023 (sd = 0.016). When applied to the EU front/side population, the US model predicted a 0.065 risk (sd = 0.027), and the EU model predicted a 0.052 risk (sd = 0.025). For the rollover model applied to the US standard population, the US model predicted a risk of 0.071 (sd = 0.024), and the EU model predicted 0.128 risk (sd = 0.057). When applied to the EU rollover standard population, the US model predicted a 0.067 risk (sd = 0.024), and the EU model predicted 0.103 risk (sd = 0.040).The results based on these methods indicate that EU vehicles most likely have a lower risk of MAIS3+F injury in front/side impacts, while US vehicles most likely have a lower risk of MAIS3+F injury in llroovers. These results should be interpreted with an understanding of the uncertainty of the estimates, the study limitations, and our recommendations for further study detailed in the report.
|
|
| 7. |
- Flannagan, Carol A.C., et al.
(författare)
-
Mutual Recognition Methodology Development
- 2015
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- Phase 1 of the Mutual Recognition Methodology Development (MRMD) project developed an approach to statistical modeling and analysis of field data to address the state of evidence relevant to mutual recognition of automotive safety regulations. Specifically, the report describes a methodology that can be used to measure evidence for the hypothesis that vehicles meeting EU safety standards would perform similarly to US-regulated vehicles in the US driving environment, and that vehicles meeting US safety standards would perform similarly to EU-regulated vehicles in the EU driving environment. As part of the project, we assessed the availability and contents of crash datasets from the US and the EU, as well as their collective ability to support the proposed statistical methodology.The report describes a set of three statistical approaches to “triangulate” evidence regarding similarity or differences in crash and injury risk associated with EU- and US-regulated vehicles. Approach 1, Seemingly Unrelated Regression, tests whether the models are identical and will also assess the capability of the data analysis to detect differences in the models, if differences exist.Approach 2, Consequences of Best Models, uses logistic regression to develop two separate models, one for EU risk and one for US risk, as a function of a set of predictors (i.e., crash, vehicle, and occupant conditions). The two models will then be exercised on a standard population for the EU and a standard population for the US. Approach 3, Evidence for Consequences, turns the question aroundto measures the overall evidence for each of a set of possible conclusions. Each conclusion is characterized by a range of relative risk on a single population. Evidence is measured using a weighted average of likelihoods for a large group of models that produce the same outcome. That evidence is then compared using Bayes Factors.
|
|
| 8. |
- Flannagan, Carol A., et al.
(författare)
-
Comparing motor-vehicle crash risk of EU and US vehicles
- 2015
-
Rapport (övrigt vetenskapligt/konstnärligt)abstract
- This study examined the hypotheses that vehicles meeting EU safety standards perform similarly to US-regulated vehicles in the US driving environment, and vice versa. The analyses used three statistical approaches to “triangulate” evidence regarding differences in crash and injury risk. Separate analyses assessed crash avoidance technologies, including headlamps and mirrors. The results suggest that when controlling for differences in environment and exposure, vehicles meeting EU standards offer reduced risk of serious injury in frontal/side crashes and have driver‐side mirrors that reduce risk in lane-change crashes better, while vehicles meeting US standards provide alower risk of injury in rollovers and have headlamps that make pedestrians more conspicuous.
|
|
| 9. |
- Flannagan, Carol, 1962, et al.
(författare)
-
Replacement of distractions with other distractions: A propensity-based approach to estimating realistic crash odds ratios for driver engagement in secondary tasks
- 2019
-
Ingår i: Transportation Research Part F: Traffic Psychology and Behaviour. - : Elsevier BV. - 1369-8478. ; 63, s. 186-192
-
Tidskriftsartikel (refereegranskat)abstract
- As Automated Vehicles (AVs) enter the fleet at lower levels of automated (SAE, 2018), the need for human drivers to remain engaged in the driving task will continue. Thus, understanding driver distraction and estimating the reduction in risk associated with removing distractions is important as AV technology develops. While previous research (e.g., Dingus et al., 2016) has estimated large odds ratios (i.e., 3–4) for using cell-phones while driving, countermeasures directed at reducing cell-phone use have not realized large crash reductions. One reason may be that drivers may replace cell-phone use with other risky activities and that odds ratios (ORs) have often compared cell-phone use to ideal driving rather than a realistic reference. Using data from the second Strategic Highway Research Program (SHRP2), we developed two cell-phone propensity models, one with age and one without, to develop weights for events without cell phone use. Using these weights, we estimated the probability of engagement in a variety of tasks in place of cell-phone use. We also estimated weighted odds ratios for cell-phone use (all uses) and cell-phone talking only. Weighted ORs are lower than unweighted ORs and much lower than ORs compared to ideal driving. This is consistent with the idea that in practice, even if cell-phone bans are effective at reducing cell-phone use, they may not greatly reduce risk because drivers may replace cell-phone use with other distracting activities in the same situations in which they normally use cell phones while driving. We also discuss the influence of young drivers on our results. Younger drivers in the dataset are more likely to use cell phones and thus are influential in the propensity model results.
|
|
| 10. |
- Flannagan, Carol, 1962, et al.
(författare)
-
What Are Drivers Doing When They Aren't on the Cell Phone?
- 2018
-
Konferensbidrag (refereegranskat)abstract
- Cell-phone bans have been motivated by previous research estimating large odds ratios (i.e., 3-4) for using cell-phones while driving. However, large crash reductions have not been realized. One reason may be that drivers may replace cell-phone use with other risky activities and that ORs have often compared cell-phone use to ideal driving rather than a realistic reference. Using SHRP2 data, we developed two cell-phone propensity models, one with age and one without, to develop weights for events without cell phone use. Using these weights, we estimated the probability of engagement in a variety of tasks in place of cell-phone use. We also estimated weighted ORs for cell-phone use (all uses) and cell-phone talking only. Weighted ORs are lower than unweighted ORs and much lower than ORs compared to ideal driving. This is consistent with the idea that in practice, even if cell-phone bans are effective at reducing cell-phone use, they may not greatly reduce risk because drivers may replace cell-phone use with other distracting activities in the same situations in which they normally use cell phones while driving. We also discuss the influence of young drivers on our results. Younger drivers in the dataset are more likely to use cell phones and thus are influential in the propensity model results.
|
|
