Working Hours, Sleep, and Fatigue: Fatigue and Automation

Working Hours, Sleep, and Fatigue: Fatigue and Automation

Show Video

Ladies. And gentlemen thank. You for joining us for, today's installment, in the working hours sleep. And fatigue webinar. Series on. Fatigue. And automation, black. Swans and Lumberjacks. Redux i'd. Like, to turn the presentation over, to dr.. Naomi. Swanson, and dr., Ted Hitchcock, who will serve as moderators, for today's event, dr.. Hitchcock. Thank. You welcome. Everyone to the NIOSH director's webinar, series this. Series examines, various topics, related, to working hours sleep, and fatigue. Today, is our fourth webinar in this series and, it focuses, on fatigue and automation. I'm. Dr. Ted Hitchcock, and along with dr. Naomi Swanson. It is our pleasure to serve as moderators. For today's webinar, on behalf, of the NIOSH healthy, work desires program, and the working hours and fatigue, workgroup, we, are pleased to have you join us today with our featured, speaker, dr., Don Fischer, Don. Is the principal, technical, advisor in surface, transportation, human, factors at the US Department, of Transportation. Volpe Center his. Current research interests. Are in human factors and surface. Transportation, especially. Automated. Vehicles, distraction. And training. He. Is also editor, of the recently published handbook, of driving, simulation, for engineering, medicine. And psychology, and has, over 450. Scientific. Publications. And, presentations. Just. A few housekeeping items, before we start today's presentation. For. Those of you who are joining us through Adobe, Connect the audio, will come from your computer, speakers or your headphones, please. Ensure that the volume is turned up to a comfortable, level and if, you have any technical issues please contact, the, Adobe, Connect meeting support. Please. Note additional, information, is provided on the right side of your screen, for, example live caps, are also available at, the, bottom of your screen for anyone who would like to follow along, folks. Who are watching the recording, may email total, worker health to, request, the transcript, of today's presentation. To, supplement, their viewing experience. Much. Of what was covered in the previous slide I just presented is included, in the notes for attendees, box which, is located in the top right of your screen so. You can refer to that throughout, the webinar, also. During, today's webinar you can submit, questions comments. Or problems, to the Q&A box on the, bottom of your screen all, meeting, hosts, will be able to see what you have submitted and can respond, accordingly. You, may submit questions throughout. The presentation, however. Please, note that content. Related questions will only be addressed at the, end of our webinar webinar. During, the Q&A session. Additionally. We are also pleased to provide closed, captioning, that you can follow along with at the bottom of your screen if you would like to view the closed captioning, in a separate, larger, window you, can do so by clicking the, URL shown, at the bottom of this slide that can be found in the notes for attendees, box. For. Information, about the presentation. Or, to see previous. Presentations. In this series please, visit the NIOSH work schedules, webpage, again. Thank you for your participation, today and now I'll turn it over to our featured, presenter dr.. Don Fischer.

Thank. You Ted, as. Ted, said my name is Don Fisher and. To. Repeat I'm no principal technical, adviser the volte national, transportation Assistance. Center in Cambridge. Massachusetts which, is part of the department of transportation and, I'm. A former, department, head in mechanical and industrial engineering, at, the University, of Massachusetts Amherst, where. I spent the first 35, years of my professional, career, I want. You to thank you for inviting me to present at the NIOSH director's webinar. Working, hours sleep. And fatigue. I. Have. Roughly, 120. Slides I want to present I, spent. On average 30 seconds, on each slide I should go for roughly one hour at which point if I have not there's I promised, I will cut to the chase so you'll have time to ask questions, okay here, we go part one together. With Steve Popkin. The deputy, director for research and technology at the voltage Center and well known to many of you because of his research on shift work and fatigue. We've. Been, putting some, thought into how fatigue will interact, with automation, in the vehicle, cover. Before I delve more deeply into. That interaction, I first. Want to lay out the background for this as, you. All know fatigue. Is more than just sleepiness and its, effects, are more than just falling. Asleep. Using. The Department, of Transportation definition. Fatigue, is a complex, state, whose citizens are in part a lack. Of alertness, along, with the reduction, the levels of cognitive and physical performance, and the, causes of fatigue are many some of which are listed here, they, include lack of sleep and of. Course disruptive, word rest cycles, emotional, stress, and neurological. Conditions, like sleep apnea. Why. Worry about fatigue, well. In dollar terms alone, the, cost of Teague are staggering. Consider. Just the United States the cost was estimated at, 77. Billion in 1990, and had, it increased, to 200 billion in 2004. These, costs, can be attributed to decreases in productivity along, with increase in health care needs absenteeism. Turnover. And accidents. Of. Course. The costs can, be measured not only in dollars they, can also be measured in injury sustained, and lives, lost. Nitz. Estimates, that approximately 2.5. Percent, of fatal crashes involve driving during droughts involve. Drowsy driving however. These, numbers may be an underestimation. The problem by a considerable, margin, the. Triple-a foundation for, like safe be using, different, techniques. Estimates. That 21% of fatal crashes involved. The drowsy driver it's, surprising, that these figures are not even higher given that 60%, of Americans, have driven well faults while while, filming sleepily 37. Admits to actually having fallen asleep at the wheel in the past year and over, 78%. Of Americans do not work a regular Monday Friday shift. Note. That within transportation. Drowsiness. And more generally, fatigue is not, just a problem for passenger, car drivers. It, is also a problem for bus operators, taxi.

Drivers Truck, drivers, and on, and on. Okay. So. Why are we here today it, is not as if these problems are not well understood, indeed. They are well, understood, moreover. A great, deal is known about counter, measures that might be taken to reduce worker, fatigue in the transportation. Sector for, example the, US Department, of Transportation Safety. Council is sponsoring. Several, fatigue management products including, specifications. For the next generation, of fatigue and performance, models an accident. Investigation, methodology. And a, document, outlining, current, experience, with both hours, of service rule making and fatigue, risk management, systems, these, efforts are producing, approaches, that can be turned into tools for use by regulators, in industry. Transportation. Or Ganesh's. Organizations. Are themselves, employee. Fatigue. Risk management, systems and technologies. That include education. Health screening scheduling, and monitoring finally. Outreach to labor organizations. Vehicle operators, and their families and the, general, public about the dangers of fatigue, continues. On abated. Yet. Fatigue. Continues. To. Stalk us, why. Does, the elephant stay. In the room there are many reasons, here, are three first. It's economically. Convenient, for both the staff and the any second, there, have been long histories of collective, bargaining that, make it difficult to, get the elephant out of the road and three, despite, all that has been done there, is an appalling, lack. Of awareness and, appreciation for, the elephant, itself the film I showed at the start is just one example, by, and large drivers. Think that they can stay alert by opening the windows playing, loud music chewing, gum or any number of other strategies. For, which there exists no hard. Evidence. So. What is a, person, to do. The. Elephant's still, in the. Room. Well. Perhaps. The. Answer, to, the problem, of fatigue at least, in transportation. Is automation. At. First. Blush automation. Seems like a godsend. A vehicle. With advanced steering, braking. Accelerating. And decelerating. Capabilities. Could, take over if the driver fell asleep and guide. The vehicle to a safe landing, however. As we, will see the, relation between fatigue, and automation, really, needs to be much more, nuanced.

In. Part this is because we, are de Cades away, from the point in time when all vehicles, out in the road will be fully automated, automated note. That fully autonomous, vehicles what. Are known as l4, and l5 are. Not expected, to reach 50 percent of the vehicle fleet, penetration. Of the vehicle fleet until 2050, at the earliest the. Most optimistic, projection, being 60%, penetration. And the most pessimistic, being, 40%, that, still leaves millions, of existing, vehicles, on the road that are not fully. Autonomous. If. Full. Automation, is not, going to solve the fatigue problems, a day may. Be partial. Automation, can, do such so. Let's take, a step back. To. Me at least it. Appears that problems, and transportation. Are, cross. Modal. And, require. For their solution, not a only, a broad understanding of, different. Functional. Impairment. But. Also an understanding of the level of automation. At which, the problems are appearing, fatigue. Is one of these problems that I believe requires, such. A broad understanding. But. I don't think this broad understanding. Cannot can, be developed in a vacuum, rather. What, I believe, we need to solve complex, problems in transportation. Such, as fatigue is, an entity perhaps. An institute, that with its combined, expertise. Can, look across the, modes across. The functional, impairment and across, the level of automation. To create synergies. That do not exist today in pursuit. Of the solution, to these complex problems, the. Institute, for functional impairment. And transportation. Or I said, which, we are in the process of building here in the Volpe Center would be an attempt to do just this to, integrate what is known about the broad range, of impairments as they, arise in the different, modes and at the different levels of automation to, generate creative solutions, for a given impairment, that build on this integration, and then, to implement those solutions or help others to implement them I want, to take fatigue, as one, particular example. Of one, why why, we need to look across the mode across. The different functional, impairments, and across. The different levels of automation in, order, to find a solution. Okay. So. The next three parts are automation, functional. Impairments, and. And. Different. Different. Vehicle types and they. Correspond, with what I just, discussed. Let's. Go on to, part. One by, examining, fatigue in more detail across, the different levels, of automation. As. A. Very general state that statement, there are two different levels of automation at the first level our advanced driver assistance systems. Or a - a das. Which require the drivers full attention at all times to the driving environment though. Such systems, may intervene, for the driver like automatic, emergency braking or, more may warn the driver like a forward collision warning. At. The second, level our automated. Driving systems or, ABS. Here. The steering, braking and decelerating. Do not require the ongoing, attention of the driver though. The driver may, need to be brought back into the loop. You. May be, more familiar with, the fine-grain definition, of these two-level, systems, as they, are defined by the National Highway Traffic, Safety Administration, and, SAE. The. A dash systems, have three levels zero, one and two the. Levels of automation. Associated. With advanced driver systems, are in blue, the. ADF, systems, also have three levels three four and five those. Associated, with driving. System, with an automated driving systems are. In, green highlighted. In green the. Key factor, to remember with.

A Dash system, is the driver needs continuously. To monitor, the driving, environment whereas. With ADF, systems, the drivers can be out of the loop. The. Separate, levels of automation are defined. Defined. Primarily by four factors. One. Of which we have already discussed whether. The driver needs to continuously, monitor. The driving, environment. This. Is the column indicated, by the blue arrow to, repeat. No that levels, zero, one and two require. The driver to continuously, monitor the environments, level, three four, and five allow. The driver to exit, the loop the. Finer. Differentiation. Of levels within, the a dash and ABS, systems, is a turn determined, additionally, by. The. Control, the automated, driving suite. Has. Over, both steering, and acceleration and. Deceleration. It. Determines. Its determined, also. By. The entity, responsible, for. Performance of vehicle control if, the system, fails and. Finally. It's. Determined, by the environments. In which. The. Vehicle can operate, at its given, level that's, a lot to throw at you at once so. I'll make it simpler by. Saying, today. We're only going to focus on levels, zero one two and three since at least currently they, are the most likely to occur and I, will define each level as we make our way forward. Conveniently. We'll. Be a with level zero. Level. Zero as you remember requires the driver continuously, to, monitor the road ahead and the driver has full control of the steering the breaking the decelerating. And. Accelerating. An. Example. Of a level zero advanced, driver assistance system. Is a pedestrian collision. Warning system, pedestrian. Fatalities, on the ride they. Now represent about. One-quarter. Of all road. Fatalities. Again. Pedestrian. Fatalities. Represent, about one-quarter, of all road, fatalities. I'm. Going to focus here on bus pedestrian. Strikes as an example of where this technology, is useful this technology, beating the pedestrian. Collision warnings why, well, when calculated. By mileage, travel. Pedestrian. Fatality rates, per. Vehicle mile traveled, are much, greater for buses compared. To passenger, vehicles, and large trucks in, 2002. The, relative, risk of attempted, estrin being killed by a bus was about thirteen, point seven five times higher than, that of a car when, the mileage was considered, is considered, only in urban areas. Intersections. Are particularly, problematic because, of all that needs to be scanned by the bus driver, this, is a top-down. View of a. Crash which occurred and an, intersection, with. A bus and other vehicles a. Simple. Example can, make clear why bus pedestrian style strikes. Are, sometimes difficult to avoid. As. An example I will step you through a top-down, cartoon. Drawing of a natural, bus pedestrian. Strike at a why intersection. So. Imagine, you are. That. Vehicle right there and you're, going to take a left, turn at the Y intersection, okay. That. Is the why intersection. You're going up here you're going to take a left turn. Now. I'm going to step you through the, process. The. First thing you need to look at is where, is opposing, traffic in the, lane, adjacent, to you. The. Second, thing you need to look at is opposing. Traffic in the. Lane two lanes over. The. Third things you need to look at is possible. Pedestrians. Which, might be crossing from. The right. In, summary, the driver needs to keep track of vehicles. In both opposing lanes and of, pedestrians, on the crosswalk on the right. But. That, is not all as some, of you I'm sure have guessed from the diagram. Not. Only do, drivers, need to pay attention to the, areas, I just defined, but. Perhaps. Hardest. Of all the. Driver needs, to keep track of any, pedestrians. That might, be emerging, from the left as the driver turns into the intersection. These. Pedestrians. Could be all but invisible to, the driver because, the vehicles in the opposing lane as the, driver was approaching, the intersection and, because.

Vehicles, Stopped at the crosswalk on, the left are, indeed. Obscuring. The, drivers view of the pedestrian, who might be behind them, this, is an almost impossible task. So. Predicting. Where hazards. Might materialize can be difficult for transit bus drivers, and for all of us given, all that they have to do in addition to what we as passenger car drivers have to do pedestrian. Collision warning systems, especially, in transit buses where, blind spots are a real problem our proposed is one potential countermeasure, again. Pedestrian, collision warning systems are considered an example of level, zero, automation. This. Actual. Bus. Has three sensors. One there one, there one, right and one on the one forward and one on the far right just. In case what you wondered what those four different, shaded. Areas were not. Only, is predicting, where latent hazards. Might, materialize a problem. For drivers who are fully. Alert. Research. Indicates, that fatigue, impairs. Hazard. Anticipation. This. Object. Right here is not a flying saucer okay, it's, actually, a, image. Of an, eye that's closed, but I've been told it looks like a flying saucer. Please remember. That it is not such instead it's a tie that's closed in other words keeping, track of this is going to be much even harder much, harder even, for. Fatigued. Drivers so could a pedestrian, collision warning system increase. The likelihood that a fatigued, driver detected. The pedestrian. Absolutely. The. Example, I just gave is, one. Of many such examples of, latent, or hidden hazards. Where a pedestrian collision, warning system could, prove useful here. Is an example of a hazard which is clearly visible. It. Would seem that a fatigue driver may if alerted have avoided the horrible. Crash with a family of three pedestrians, just. Seconds, before it. Occurs. So. We're the first of many diagrams. You're going to see like this we've. Got benefits, on the Left. Diss. Benefits, on the right the. Benefits, are outweighing. The disc benefits, let, me explain if we, went to consider on to consider the benefits of religion pedestrian. Collision warnings and other safety warnings I believe, that we would include the fatigue, driver the benefits easily outweigh the disc benefits for, the driver who is fatigued because he or she is drowsy, overloaded, or simply, mind wandering, the warnings could potentially, serve as an alerting function, for, the driver was actually in a microsleep the alert, could actually awake, the driver the, one diss benefit, might be that, it could startle a fatigued driver thereby. Delaying a quick appropriate. Response, this, needs more research. But, in short the net benefits, would appear to outweigh the knit net diss benefits. But. What does, research actually. Say. Above. I was just speculating. On the benefits of alerting drowsy, drivers it seemed reasonable, to do such since a number of simulator studies have shown that forward. Collision warning systems, can we do this both collisions, and response times for distracted. Drivers, presumably. The same would be true for drowsy drivers. So. It was an experiment, was run at the, University, of Iowa on the national, advanced driving simulator, to, test just this, pollicis. Participants. Were asked to remain awake from 7:00, 7:00. In the morning on the day of the visit until, the arrival for, the study sometime between 5:00 and 7:00 in the evening and asked, to refrain from caffeine. Beginning, at 12:15. The day of the visit to. Manipulate, drowsiness, level, each participant, completed, one drive between 10:00 at night and 2:00 in the morning and the second drive between 2:00 in the morning at 6:00 in the morning the. Study was a between subjects design with, three warning conditions no forward collision warning an audio. Alert which was a series of several loud beeps and a, half book warning consisting, of series of rapid pulses in the front portion of the seat in. The simulator, the driver was following. A lead vehicle. Which, suddenly swerved, into, you jacent, lane to. Avoid a stopped, vehicle so this lead. Vehicle suddenly, swerved into the adjacent lane to avoid a stopped vehicle. Response. Times to the alert were measured both by accelerator, release the. Response initiation. And brake, pest press, the, response execution.

Surprisingly. Given. Previous, effects, a forward. Collision warning on. The response, times and crashes, of distracted. Drivers, there. Was no effect, of forward. Collision warnings, on the response times of fatigued drivers, what, is going on the. Author's speculated. There, could be a number of potential, explanations. I'll discuss. Just one, the, day gave, unlike. Distracted. Drivers, drives, the drivers they already, have been, looking forward toward, the location, of the lead vehicle reveal event this. May have mitigated, the potential, orienting. Benefits, of forward collision warnings the. Vision, for action processing, steam stream. Insensitive, to basic visual stimuli, such, as loving, that operates, largely outside, of conscious processing. Such. A visual, processing, stream may have been at play with drowsy drivers, allowing. Drivers to respond, appropriately to. The lead vehicles threat in spite. Of the fact that they were, impaired. They, were fatigued. So. Maybe it is not the case that warnings. Will have net benefits. We. Don't know exactly what will be the case we, need more, research. At level, zero, of automation. Let's. Next consider a level one. Automation. Problem, with. A level zero car the, driver manure the vehicle entirely himself, or herself, with. A level one system, if the steering is controlled by the vehicle, or the braking, accelerating, and decelerating are. Controlled by the vehicle but not both so the driver either controls the steering or, control, controls, the longitudinal, position or, the lateral position but, not both the driver still needs to be in the loop. With. This in mind let's. Move from level one to a particular level one system, and a particular adaptive. Cruise control. Adaptive. Cruise control which many of you may have in your vehicle senses, where the vehicle in front of you is relative, to your own vehicle and slows down and speeds up your vehicle to maintain consistent, spacing, unlike. Traditional cruise. Control, which only can be set to a single speed, adaptive. Cruise control or, ACC. Can, adapt when other vehicles change their speed a fatigued. Driver who, engages adaptive, cruise control would, potentially, benefit in two scenarios in highway driving when, vehicles are, traveling. Over 25, miles, per hour, first. When. A. ACC. Equipped vehicle, approached, a slower vehicle, traveling.

At A constant velocity, or. When. A lead vehicle slowed in front of an ACC, equipped, vehicle. Why. Would a fatigued driver benefit, in these two scenarios, there, are many reasons consider. Just one the, top row represents the duration of the perception, comprehension, and response time in an, alert driver the. Bottom row represents the duration of the, perception, comprehension, and response times in a fatigued driver the. Perception, comprehension. And decision time are slowed in the fatigued driver each of these stages is required in order to make a correct, response without. An ACC, quit vehicle, thus, the, fatty driver could easily benefit, from the automation. So. It looks like indeed. The. Benefits. Outweigh. The disk. Benefit, but. This ignores two. Essential. Factors. At. Least, in. My way of third thinking, first. It ignores the fact that automated, vehicles, with level 1 automation. May lead to mind wandering in general, not. Just mine wandering, with respect, to adaptive. Cruise control. Internal. Distraction, or mine watering, is now being recognized, as a significant. Source of driver. Distraction that. Requires rigorous, study and this, sort of internal distraction. Is especially. Likely in driving, with a low workload and, delayed, feedback, exactly. Those characteristics. Which I argued define, level, 1 automation. So. Where do we go with this let's. Consider two studies. First. Consider, the following simulator. Study on the effects of mind wandering on drivers, scanning, performance, in low, and high, workload scenarios. Low workload being like automation, high workload being, like manual the, question, asked was the following when compared, with alert drivers, would, be glance, behavior, of drivers, who were mind wandering. Be, more effective, in the low workload scenarios. And in the high workload, scenario. The. Actual tasks of the following in, the driving simulator participants. Were asked to maintain lateral, control, and a safe headway distance, while, following a lead. Vehicle. That's, indicated, by your blue, arrow and keeping. Ahead of a trailing car that's, indicated. By the. Red arrow at the bottom. The. Trailer vehicle was used to motivate participants. To check their mirror the lead car drove at an average speed of 45 miles per hour accelerating. Or decelerating when. The range of 40 to 50 at random intervals the. Participants. Car and in case the. Colors. Aren't clear, the orange arrow. Began. Each stride position mid made between the lead and the following distance, a vehicle. And maintain a constant distance of 200, meters from each other, the. Measurement, excuse, me the. Measurement of mind-wandering was subjective. Participants. Were told to keep their attention on the driving task is much impossible as, much as possible and we were instructed to press, a button on the steering wheel to report, any time that they found themselves mind-wandering. Participants. Eyes were tracked throughout the experiment the. Standard, deviation, of the horizontal. Gaze position. Standard. Deviation how far they looked left and right, was. Used to the dependent, variable, the, less broad was the scanning the, less safe it was assumed was the driver. These. Right here for those of you who don't recognize it are actually. The, lightweight eye cracking, goggles one can use now in studies. Okay. These. Are the two conditions in some. Segments. Of the drive there, were heavy winds. That. Corresponded to a. High workload in other. Segments, of the drives there, were no work no, wind that correspondent. To, low workload. The. Findings, of the behavior, the. Findings on glance behavior in the low and high workload, conditions. Are displayed. Here what. Do we see the. Standard deviation. Of, the. Horizontal, gaze position. Is on the y axis the. Dark grey bars represent. The condition which the participant, reports, himself, or herself as mind-wandering the. Light grey bars represent, the situation, in which the participant is attentive note. That in both the high workload condition. On the right and the, lower workload, condition, on the left the drivers are scanning less broadly, when their mind, wandering, but. That wasn't my. Point. Exclusively. For bringing in this information. Importantly. For our purposes. The difference, in. The. Standard deviation. Of the horizontal, gaze position. Between the alert and mind-wandering, drivers. Is greater. When. The workload. Is. Low. The. Green box then. The workload, is high, the. Red box in, short. The, relative, effect of scanning on mind-wandering. Is greater, in the autonomous driving, light condition, low, workload than, it is in the manual driving. Light condition, high, workload. So. I talked about the effect of, mind.

Wandering, On, scanning. Read horizontal, a standard, deviation of horizontal gaze condition, let's, talk about response, time now. Next. Consider a simulator, study of the effects of mind wandering on the drivers response time in delayed and immediate, feedback scenarios. The. Question asked was the following when, compared with alert, drivers, with the response types of drivers who were mind wandering, be, affected more in the delayed feedback scenarios. Than, the immediate feedback scenarios. The, task was a relatively, simple one, but. In a simulator, study, the drivers car would suddenly change into, the next Lane, and. Orient. Yourselves here you, see the. Drivers car right here suddenly, it starts changing into the next Lane the. Driver was supposed, to return his or her vehicle. To. His or her Lane. One. Group of drivers was, given the equivalent, of. Delayed. Feedback on, the change in the position, of their vehicle that if they were given only visual cues the, other group of drivers was given the equivalent, of immediate, feedback, motion. And visual. Cues. Moreover. The, measurement, of mind wandering in this case was an objective, one. EEG. Recordings, were made to determine in both the delayed and immediate feedback conditions, whether the drivers mind-wandering, whether, the drivers mind was wandering or attented, at, the. Time the. Response was made. The. Dependent, variable, was said to the time between the car first wandered out of its way and when the driver returned, the car to its lane so you see that right. Here the. Car first wandered out of its lane and now the drivers returning, it that was the dependent, variable or the response time. So. What did the results show the, response, time is on the y-axis I'll, cut, to the chase here, the. Average, response time of the drivers, who are mind wandering are indicated, by the plus signs in the blue boxes. The, average, response time of the drivers who are alert are indicated, by the plus signs in the orange, boxes, importantly. For our purposes, the, differences, in the response times between, the alert and mind-wandering. Drivers. Is greater when the feedback is delayed. Green. Box then, when the feedback is immediate, in. Short. And analogously. The relative. Effect of response times of mind 1 is greater in the autonomous, driving light condition, delayed feedback than, it is in the manual driving, light condition, immediate. Feedback. In. Summary. When. Looking at the disc benefits, of, adaptive. Cruise control we cannot ignore the fact that automation.

Decreases. Situation, awareness in general not, just with respect to scenarios, in which, adaptive. Cruise control would, help to. Repeat first. A loss of situation, awareness leads, to increases, in response time, the. Left hand side of the figure the. Orange arrow. Second. A. Loss. Of situational. Awareness also. Leads to decreases, in the breath of scanning the. Orange arrow or the right hand side of the figure, both. Changes, increase, the likelihood that a driver will. Crash. But. What does this have to do with fatigue. Well. We know that automation. Decreases. Situation. Awareness. We. Know that the t decreases. Situation. Awareness. So. When fatigue, is present, both, are operating. To decrease situation. Awareness. So, when the fatigue isn't. Present. When. Fatigued when when there's no automation, we no fatigue drivers are going to do worse than alert. Drivers. However. When, automation, is. Present. Fatigue, may. Disproportionately. Increase. The number of drivers who, lose situation awareness, in the presence of automation the. Figures are meant to be illustrative only. So, we have the worst of the worse Automation. Increases. The loss of situation, awareness which, leads to increases, in response time and decreases, in scanning and fatigue. May disproportionately. Increase the number of such drivers who lose situation awareness, and, therefore the number of drivers with increases, in response time and decreases. In scanning we need research to confirm this hypothesis. There. Is a second, caveat to the statement that level automation, level, 1 automation, such as adaptive cruise control has, no dis, benefit, in particular it ignores the fact that ACC, may not work effectively in certain types, of weather. Conditions, additionally. These symptoms may not work in some environments, such as tunnels. What. Is the impact on fatigued drivers, well, we know that memory is impaired when drivers are fatigued so fatigue drivers are less likely to remember the details of the operational, design domain, than alert drivers, automation.

Over Time will. Decrease drivers likelihood, that they will remember the details, of the operational, design demand, and automation. Is likely to disproportionately. Increase the number of fatigued drivers who failed to remember the operational, design domain research. Is needed to confirm this hypothesis. As well. So. Now we've got a complex situation. With. Situation. Awareness. Throw. It into the mix and the operational, design domain thrown into the mix we need a slightly more sophisticated model. Of the benefits and disc benefits of level 1 automation, 40 50 drivers we. Need to start by differentiating, between those, cases where the ACC, is engaged and no CT cases, where the ACC, has disengaged. If. We. Look at just the cases where, the ACC, is engaged the, ACC, will keep the distance between the drivers vehicle, the lead vehicles Floyd if necessary, something, which could would not happen without a cc. If the, driver lost situation. Awareness these. Are the targeted crashes, however. Because the, use of a cc does not increase, but. Does increase the loss of situation, awareness ittan could increase the likelihood of other crashes, even when it were engaged what we are calling not, targeted, crashes here so. We've got targeted. Crashes, it decreases. Their likelihood, but, because you lose situation awareness when. You are when. ACC is operating, or let's the hypothesis. The the crashes that aren't targeted, by ACC, may in fact increase. Similarly. Suppose that, the ACC. Is now disengaged, either. Because it is operating, outside its design domain or because the sensors, become. Jam then, the fatigue drivers less likely, to notice this and a. Driver has lost situation, awareness either because of mind watering or micro speed microsleep. Is more likely, to be in a rear-end crash for, all or any type of crash. We. Are entering a no-man's land and we need answers. Okay. Let's go to level 2. You. May remember. For. Those of you have a level 2 car, that, would level 2 automation, the. Longitudinal. Control. Are no longer being performed by the drivers they under the control of the vehicle this. Can offer additional safety benefit for fatigued individuals, actually staring braking and accelerating, decelerating. If, the driver, starts mind wandering or actually. Falls asleep however. The same two caveats apply to level 2 automation is applied to level 1 automation, since. The driver must monitor the vehicle, at all times. First. Mind-wandering. Is even, more likely with level 2 automation, than it is with level 1 automation, because now not only the longitudinal control.

But Also the lateral control is determined, by the automated driving suite when, increases, in mind with increases, in mind-wandering come increases in the response time and the decreases, in scanning activity and these changes, are apt to be amplified, for fatigue drivers. Recognizing. The importance, of drivers, being in the and remaining, situation. Where the, drivers alertness is now monitored, if only indirectly, with level 2 automation, for, an example an alarm may sound if a driver's hands are off the wheel for too long the, end result sometimes being as long as 5 to 7 minutes however. This is way too long to capture a microsleep. Additionally. A second. Caveat the, driver now needs to keep track of the operational. Design domain, of 2 general, vehicle functions steering. On the one hand and braking decelerating, accelerating, on the other hand here, is one of 30 exceptions. Where automated, steering does not work for one manufacturer. I want to read it, to. Prevent serious injury, or death be alert and pay special attention when, passing freeway exits, and entrances with. Super crews and be ready to take control of the vehicle unnecessary, changes. In lane markings, around freeway, exits, and entrances can, monitor, momentarily. Cause super crews to. Not detect, the correct lane if this occurs super, crews may attempt steering inputs to bring the vehicle back into the collect crackling, and in rare circumstances, could. Overcorrect, and cause the vehicle to momentarily. Cross into a lane next to your vehicle unless you manually steer, to maintain your lane position mind, you the drivers told throughout, the manual that they must pay attention at all time. But. Can you imagine a fatigue driver trying to deal with a lane change forgetting. The operational. Design domain, and then having this car, steer into the other lane again. Increase, in automation, are likely to make it more difficult for alert drivers to remember the details of the OD D and even, more difficult for fatigued drivers to do such. So. What about the benefits and dismisses, of automation, again, we need to consider the, case where l2 is engaged level 2 Automation engaged and where it's not engaged. First. Let's assume, that. Level 2 automation is engaged and the driver is fatigued, losing. Situation awareness when, the vehicles stay in the lane and slow down and speed up appropriately. This should decrease the targeted, crashes, moreover. The driver is issued alerts that he or she does not even, stay. Engaged with the driving task this may decrease crashes, as well however, there are hazards that - Automation does not mitigate, even when it's engaged such.

As A pedestrian, walking out into a crosswalk, partially, obscured, by a large vehicle in a parking lane stop right in front of a crosswalk. What. A camera can't see it can't predict, that's, a fatigue driver is much more likely to lose situation awareness without, to automation and therefore much more likely with l2 automation, to strike a pedestrian, or other latent hazard than without such automation, that's. Their disincentive. It's to automation, l2 automation, even, when it's engaged those. Are the not targeted, crashes. Now. Let's imagine that the l2 has disengaged for, whatever reason, if, it's disengaged. The fatigued driver is less likely to notice this I have homicide, there, are no benefits and because the logit situation, can be almost complete, for, the fatigued driver there are dips benefits, that lead to increases, in all types of crashes. Finally. Let's consider level 3 automation. With. Level 3 automation, the driver no longer needs to be continuously, moderating the driving environment however. The driver does need to serve as a fob fallback, if automation sail a. Number. Of studies have been run to determine how long it takes a driver of a level 3 car who is out of the loop to recover enough situation, awareness in. Order to be able to drive safely estimates. Run from 10 to 20 seconds and I'll talk about this shortly but. These are relative or a relatively alert driver. But. What is to prevent a fatigued driver from falling asleep in a level 3 vehicle, as of yet nothing, and how long will it take to wake the driver and for him or her to become situational. Situation. Aware we, simply have no idea. So. We've seen the disorder to compute the benefits of automation we need to consider whether the automation is engaged, or disengaged to, begin consider the case where, it is engaged. So right here consider this case where it is engaged and. Assume that the crashes are targeted, ones so we differentiate it when it's a cage and it's targeted. With. The increase, in the level of automation the fatigue driver who loses, situation, awareness for whatever reason, is less likely to be in a crash which, the automation is dying to decrease and the risk increases of the level of automation increases. Next. Consider the situation where. The. Crashes. Are not targeted, by, the automation. With. Increase, in the level of automation, situation. Awareness decreases. And the likelihood of being in the crash not targeted, by the automation, increases. And. These. Were situations, where the automation was engaged. What. Happens when the automation is disengaged. Well. Then both. For targeted. And not. Targeted, crashes, we're, going to see a, potential. Increase in crashes, because, the driver losses increasing, self-awareness, as the, automation, increases. We. May believe in our heart of hearts that the likelihood of a driver crashing, with no automation, is much higher than a driver crashing, with automation installed, in the vehicle but the bottom line not only for fatigue and automation but for many functional impairments, and automation is we simply do not know what the net benefits, will be when the automation is engaged or disengaged. When. The automation is engaged there's good reason to believe that the benefits will substantially, outweigh, the dissonance when. The automation is disengaged, and the disengagement, is not detected, there's good reason to believe that the disc benefits, will outweigh the benefits, but, we cannot conclusively, determine. The net benefits. Of automation for a given impairment, unless, you know the probability, of the automation is gauged and disengaged, these, are the probabilities highlighted. In red, there. And there. And. The. Conditional, likelihood that a crash, the. Conditional, likelihood of a crash given that the automation is engaged and disengaged, those. Are those probabilities, those conditional, probabilities. But. As today we do not know these probabilities. In, summary, hopefully. Steve and I have provided you with a framework for thinking about how to maximize the benefits and minimize the dissonance for fatigue, drivers at each, level of automation, keeping, in mind whether the automation is engage to disengage, and keep it in mind whether the targets are the, automation.

Is For target target crashes, or not, targeted crashes I. Said. At the very beginning that, indeed. In. Order to think creatively about solutions, to fatigue and automation, we need to consider the different levels of automation the, different functional impairments, and the different modes I just discussed why it's important to understand the different levels of automation if, one is going to compute the true benefits of automation to the fatigue driver I now. Want to argue how knowledge about other functional, impairments, can inform countermeasures. We might implement for, fatigue. So. I looked at very problems. In transportation. That is addressed across the lifespan and. I, focused on novice drivers in, particular I want you to think about, what are the particular types of driving skills which, are impaired when the driver is fatigued and whether, the impaired skills remember, the set of skills that are impaired in novice drivers and if so whether, training can be used as a countermeasure to impairment, with. This broad perspective, in mind our research has shown that petite drivers hazard anticipation hazard. Mitigation and, attention. Maintenance skills are impaired these. Skills have established, links, to crashes. Well. It turns out that these skills are also the very ones that are impaired in novice drivers when compared with experienced, drivers. So. We asked whether training, could improve these, three sets of skills among fatigued drivers we know it did sit so among novice. Drivers, we wanted to know whether it did, so among. Petite, drivers we, ran an experiment to test this upon hypothesis. There were 36, participants all licensed, nurses, nurses, their. Hazard anticipation Hazard. Mitigation attention. Maintenance skills were, evaluated, in the morning on a driving simulator using, an eye tracker half. Were then trained on a PC on these skills half were given to osebo training, 12, hours later they were brought back for a second simulator, evaluation. Before. Giving you the results, I need to define what I mean by each of these three skills let's start with latent hazard anticipation, consider. First the definition of hazard anticipation in, this, scenario there, is a mark mid-block crosswalk. The orange arrow. You. Can just see it there a. Pedestrian. Could well be in the crosswalk but, obscured, from the driver by, the then stopped, in the parking lane. Immediately. Upstream of, the. Crosswalk, the yellow arrow this, is a particularly dangerous crosswalk. Because, it's located right near a library, which has a large, children's. Section a slow, driver a safe driver would slow move slightly to the left and glanced to the right as he or she passed the crosswalk, of. Course, we don't evaluate latent, hazard anticipation on, the road it's simply too dangerous, could, you bring up. The. Next video read, when you get a chance instead. The evaluation, occurs on a driving simulator. We. Used an eye tracker to monitor the drivers fixation, to each point in time you can play it anytime. Here. You're going to see the. Crosshairs. Indicating. The. Driver. I didn't see it on mine read if you see it on yours. Are. You seeing it anyway, I'm. Working, on it, okay. Crosswalk. Unsafe, either. Ones fine at. A ditch truck crosswalk unsafe right, try sake. We'll, spend only a second we work these yesterday people. Are sitting out there I apologize that. They're just not working now try, once more read then we'll move on. Okay. Let's let's move on this is oh there you go okay so that's what we see and you, saw the cross banks read you can take them off no those, crosshairs were, a luster. To that we're in a luster those crosshairs indicated, where the driver the, drivers six-nation point and that was a safe driver the driver looked to the right in the, video on the left the truck crosswalk. Unsafe, the driver just stared straight ahead and you never see. It. So. That's what we mean that's how we evaluate latent, hazard anticipation how do we train it. Well, PowerPoint was used as a training software, a desktop, PC, would use of the training platform, and what we ask the participant, to do was to drag an arrow to the. Area. Of the scenario, where a latent, hazard might here and of course that's right their. Driver. Then had to click on this arrow in order to move forward in the training program if they missed it they got another chance they got three chances then they were told where. It actually was. Here. Are the hazard anticipation. Results. You. Will see the, probability of glancing, at the latent hazard is on the y-axis that is the probability of glancing, towards, the truck, and of towards, the pedestrian, in front of the band. In. One of the scenarios of, the greater the probability the safer is the driver looking at the dark bars, the untrained, group we see as hypothesize. That fatigue, depresses, hazard anticipation, so.

Fatigue On the, pre there's, the morning evaluation, there's the evening the value fatigue. Depresses, hazard anticipation, looking. At the light gray part, bars, morning. Versus. Evening. We. See that hazard anticipation training. Almost double the number of hazards anticipated. From 35%, in the morning to most 60%, in the evening, or roughly. Four times what, the untrained, groups were in the evening. Next. Let's go to. Hazard. Mitigation, same. Scene as before you could have a latent hazard. Here. The. Participate, notification. Slow as he or she approaches, the crosswalk, and steer slightly to the left and we can gather this information on, the driving simulator but we don't train. On the, driving. Simulator. We. Do the evaluation, on the simulator, and we actually look, at the vehicle position when they're in the simulator. Training. Occurs, in, something, like this it begins with a slide showing, scenario. With a latent threat and they're asked to point. To where the latent threat would be and we help them if they don't get it this time and then. Up. Come the hazard. Mitigation training. Slides and here, you'll see, that. They could steer. The. Same course they could steer slightly to the left slightly to the right of far to the right and farther left. And they have to click on one of those, Aeros ah and. They, also have to do something about speed, they could stay the same because it's Ella rate slightly, or quite a bit they could decelerate, slightly or quite a bit and they had to choose which of those was the appropriate, one for that scenario here. Are the hazard mitigation results. The, results of straining training, were also striking when looking at hazard mitigation on, the, y-axis you. See. How. Many feet they move to, the left of their starting, position, 10 seconds, upstream of the latent hazards so we start measuring, that we look at their initial position then we measure how far they look. To the left of that as they, approach the hazard so, on average as the. Untrained. Drivers, approach the hazard which was on the right they move closer and closer. Whereas. The trained drivers, as they approach the concert steered, slightly to the left, Curley hazard mitigation training. What's working for the fatigue drivers, finally. We have attention maintenance, training could, you play this video. If you've got it if, you don't that's fine too. If. You can play. That's. An example of what we call attention maintenance, or. The lack thereof, the lack of attention management. For our purpose. That driver Willie assumed me and obtained attention, to the forward roadway as long as no glances, are longer, than two seconds down inside the cabin of the automobile, glances, longer. Than two seconds away from the forward roadway are, associated, with an increase, in crash risk by a factor, of three again. Glances, longer. Than two seconds down inside of the cabin of the automobile, are associated. With an increasing crash risk by a factor, of three that's the same increase in crash risk for a BAC of 0.08%. Later. With an eye tracker I won't. Ask Reed to play. The video you, see oh go ahead and play it Reed if it goes it goes that's great you'll, see the, driver tried to perform a secondary task no what. You see wrong here if you see anything that's where the driver, is glancing. And. I. Wish I could raise hands now or whether have you raised hands at the end of this I delivered this lecture at West Point the other day and asked whether there was anything, wrong. And. The. Students said well maybe, the. Driver ran. A red light I don't think, that was the case I asked, you whether you see anything wrong and if you count the number of seconds, the crosshairs, are down right there it's up to five seconds, so while performing the secondary, task the. Glance is clearly drivers, clearly doing something unsafe, okay Reed you can advance, thank you. So. How. Did we train attention, maintenance we had a video you're, not going to show the video I didn't it, knew would have been too large to play here but then the, driver's seat clicks on this and seized forward.

Roadway Unfold, in real-time, we. Can't train with an eye tractor, inspect too widely disseminate, the training but how are we going to capture the duration, of glances, away from the forward roadway, well. We did so as follows the participants, watched the video of a simulated Drive which should pull unfolded, before them in real-time they, were asked to look at the map and find the. Name of a street that intersected, the street on which they were currently. Driving, whenever, they felt safe so when they felt safe they. Click. On that button and what, would happen is the view of the forward roadway would disappear, and they'd. See a map and they, had to indeed. Find, the. Name of street that intersected, the street they were driving, them when. They were. Decided. They glanced long enough down at the map they pressed this box which, was Drive you can't see it because I'm covering up which was Drive and they went back to the previous view. So. We could measure without measuring eye movements how long they looked away from the forward roadway because that was the duration of time the map was up and they could take multiple glances, at the map here. Are. The results. I'll. Move this up here the cursor up here because I think I'll need it let's. First. Consider the untrained drivers in the morning and the evening the to reddish bars on the left the average number of glance is greater than two seconds, in each map drive is on the y-axis for. The untrained, drivers. The. Average number of glances, inside, the vehicle greater than two seconds in the evening, was, statistically. And practically, significantly, greater than it was in the morning okay. That's the undine remans. Consider. Next the train drivers in the morning in the evening the two greenish bars on the right these two greenish, bars on the right. For. The train drivers. The. In the excuse, me for the train drivers the number of classes over two seconds actually decreased, in the evening. So. We see in the evening it decreased, over, what it was in the morning. Again. I hope this example illustrates, the benefits, of an integrative approach to the understanding, of fatigue by. Thinking about countermeasures, that worked with some, functional, impairment, other than fatigue in this, case the functional impairment, being, an experienced, or novice drivers we potentially, identified, some new countermeasure, for fatigue I. Said. At the very beginning that in order to think creatively about solutions, to fatigue and transportation. We, needed to consider the different levels of automation the, different functional impairments, and the different modes.

Finally. I want to explain why it is I believe that fatigue, requires a multimodal, solution. The. Particular problem, that requires a multimodal, solution, among several is microsleeps, when it comes to fatigue a fleeting. Uncontrollable. Brief episode of sleep which can last anywhere from a single fraction one second up, to a ten full seconds. I'm. Going to talk about a three step approach to the problem of micro-sleeps but, only after I talk about what. The three-step process might look like for, distraction, the, first step is to measure the, duration of glances inside the vehicle the, second, step is to measure the glasses. Measure. The duration of glances outside the vehicle and the, third step is to keep running track of the duration of glasses inside, and outside the vehicle over, the last 25, seconds, or so of driving, as. A driver glances, back and forth between the inside and the outside of, the vehicle tasks. Where a driver is switching, glances back, and forth are. Common enough and. Perhaps much more frequent, today than 20 years ago now we all have GPS units, you see one here mounted on the. Dashboard. Now. We all have GPS units, and many of us choose to look at the visual display for directions, in addition, to or instead of listening. To the audio display for, direction. In. Vehicle, distractions, have, long been associated with, crashes. Up, until, recently. Researchers. Assumed, was the mean duration that, was a problem. But. It turns out that, it is the especially. Long, glances. The, tail of the distribution that. Is creates the problem. How. Do we know this. Two. Studies are worth noting I've mentioned, one I didn't, describe it and I'll just briefly stated in a naturalistic, study of 100 cars reported, by Clower at all in 2006. The, authors attempted, to determine the relationship, between driver, inattention and crash, near, crash read they, concluded, that drivers engaged in secondary, visually, or manually, complex, tasks had, as I said before a three times higher near crash crash, risk than drivers who were attentive. In. A simulator, study by hurry and wicken in 2007. It was found that, 80%. Of, the crashes, on the simulator, were due to the 20%, of, glances inside, the vehicle longer than two seconds, but. Long glances inside the vehicle are only half the problem. Dr.. Sippy samuel ran an experiment to determine just what was the duration of the minimum glance on the, forward roadway. Not. Only do, we need to worry about how long the glances, inside, the vehicle we need to worry about the minimum duration of those glances, on the forward roadway and that's, the experiment that dr. sippy samuel ran. It. Was run on the driving simulator, with three forward screens there were three conditions, each participant, was assigned to only one condition in the, contingent, continuous, Edition which you see here the, simulator. World. Was displayed on all three screens in front of the driver at all points in time in, the. Alternating. Condition, which. You see here or unloaded. Condition the drivers view, of the, forward, roadway on the center screen is interrupted. Multiple, times, by. A blank, black. Screen. The. Driver did not need to do anything when the blank black screen appeared which is why it's referred to as the unloaded condition note, that the durations, of the alternating, glances on the forward roadway.

And At the blank black screen were entirely controlled by the experimenter, in this case. In. The final condition, the alternated, loaded condition the, drivers view of the forward roadway in the center screen is interrupted, multiple, times during, the performance, of a secondary. Task by a target display. It. Blinks, on and off on and, off the driver had to count the number of T's when the target screen was present, in the loaded condition. There. Were. Nine. Groups the continuous condition, for, alternating, unloaded. That's where the blank black screen for alternating, loaded, b1. Was. 2, seconds down you. Had the blank black screen for 2 seconds in one second up b2. Was 2 seconds down 2 seconds up alternating, with, alternating low second low II one with 2 seconds down 1 seconds, up the. 2 seconds down was a targeted, screen so. On and so forth so they had their continuously, viewed it, alternately. The front screen was replaced by a blanks black, screen or alternately. It was replaced by a screen with T's. You. Can probably already guess what we use to, determine, whether. The. Drivers had the time they, needed in order to regain situation. Awareness when they looked up on the road in fact we measured latent, hazard, anticipation by. Looking at their eye movements. Here. You see the latent hazard again is the pedestrian that might emerge from behind the truck and, here. We see, the. Results, on. The. Y-axis is the proportion, of latent hazards that were detected, the blue bar represents, the group that was able continuously, to, see the forward roadway it, is clear they are more than like more likely than any of the other groups to detect the latent hazard so. This, group right, here detects. 80 percent of the hazards and, in. No condition at no. Time is. Any. Other group at 80 percent. These. Three. Four groups in green. Are. The group, that had the alternating blank black, screen. It is, clear, looking. At the progressive increase, in the poor has is anticipated. That. This. Increases. As the duration of the forward, glance increases. So here the duration of the forward glance was only one second, here it was four seconds, has, an anticipation, increases.

But. The performance, in the alternating group never approaches, the continuous, conditioned. Sitting, model to the data we find that it was taken a full seven. Seconds. For, performance, in the unloaded, condition to reach the level of performance in the continuous, condition, that means that, you would have looked down for two seconds up for seconds down for two up for seven if indeed, you're going, to detect, at the same level, the, latent hazard in. The alternating, condition, as you did in the contingency condition. So. What does all this have to do with microsleeps. Well. Bobby sepals, and others have shown that if one keeps running track of the 25 seconds, prior. To and. Up. To the current moment one can reliably differential. Nationalistic. Data between crashes and air crashes again. One, can reliably differentiate. Between crashes and near crashes, and naturalistic data if one keeps running track of, the 20 seconds five seconds prior to and up to the current moment this. Is extraordinary no, one has been, able to come close to doing this the key is to note the duration, of glances down and glances up the, glances, down right. Here glances inside, reducing. The level of attention the level, of attention starts at one goes down and then. The glances, up increasing. The. Level of attentions, up but the glances up increases, attention less quickly than the glances down decreased, attention at some, point attention gets so low when glances were relatively, long periods of time inside the vehicle and relatively, short periods of time outside, the vehicle that a crash is likely as an occasion occurs which require, the drivers full attention, I think, the same critique procedure. Could potentially, use to predict microsleep, but, as with all things it, needs more research. Ok. I've run five minutes over I'm in, my conclusion, and that will take just a few seconds ah. First. Consider. The importance of the level of automation to the understanding, of the effects of fatigue Crick, glistens, in a now famous talk, title, black swans and Lumberjacks noted that a little bit of automation helped the lumberjack, and if the automation failed the lumberjack, could recover however. Well a great deal of automation, could help the lumberjack, even more if it sailed a disaster. Was awaiting the. Same is true of automation, the Black Swan here is the rare event with, a little bit of automation if the driver failed to see the rare, event he, or she is normally partly out of the loop but with a great deal of automation, say level 3 the driver is almost completely. Out of the loop and disaster, potentially, aways this, is doubly true for, the fatigue driver or, so high of Rd. Second. Consider the importance of thinking broadly about the types of functional impairments, we know that there are situations where automation, is likely to prove troublesome for, individuals. I noted. That by considering training, programs for an entirely different functional.

Impairments, In that case lack of experience, one could potentially train drivers so that when fatigued they were much likely to drive safely. Anticipate. Hazards pay attention and mitigate, hazards correctly when they did occur here, we are providing, help before the driver. Then, how you could take the elephant out of the room have proved useful thank. You for your attention. Reid. You run or Ted. And Naomi's No, thank you dr. Fisher that was a very interesting presentation you. You, gave us a lot to unpack there. Before. We jump to the questions, just a quick reminder to, the folks who are joining us online to, please type in your questions, in the Q&A box on the lower right hand corner of your screen and, as, we start to compile the questions I'd like to remind you that the webinar and, presentation. Slides will be available on, our NIOSH. Work schedules, website, along. With the archived presentations. In this, working hours sleep and fatigue, webinar. Series, again. If you would like a copy, of the unedited, transcript, please, email total. Worker health at the, email link below. While. We wait for questions to to load I have a quick one for you dr., Fischer you mentioned, early on in your presentation that. A, lack of awareness and appreciation for. The fatigue elephant, in the room NIOSH. Has made, some, some inroads into some translational. Products, in long-haul. Truck drivers, and in other. Industries. Such as nursing, healthcare etc, but. I was wondering if you could take. A minute and and give us your idea, as to what types of products, or training, do you think would be best to address this lack of awareness in, a general driving population. There. Were there's, a recent. Review. Of the literature that's. Being done for the national highway safety. National. Highway Traffic Safety Administration and. I. Can't. Imagine the. Problem. Being solved, for the general public without something. Like the publicity campaigns. That. Have gone on for designated. Drivers, and. Other related. Problems. But. I haven't given this anywhere near the far, have a knock.

So, That. Would be my first stab. And, it's. Maybe. You've done something like that Ted, you're. Better, better. Able. To. Define that well. I know I know my colleagues, in our division of safety research for their center of motor vehicle safety and health they've done a lot in terms of, translation. Safety, behind the wheel is one of their newsletters. That goes out it has a wide circulation and, we've looked at th

2019-02-17 18:15

Show Video

Other news