Driver Workload Research Articles

Mental workload accumulation effect of mobile phone distraction in L2 autopilot mode

As automated vehicles become more common, there is a need for precise measurement and definition of when and in what ways a driver can use a mobile phone in L2 autonomous driving mode, for how long it can be used, the complexity of the call content, and the accumulated mental workload. This study uses a 2 (driving mode) × 2 (call content complexity) × 6 (driving stage) three-factor mixed experimental design to investigate the effect of these factors on the driver's mental workload by measuring the driver's performance on Detection response tasks, pupil diameter, and EEG components in various brain regions in the alpha band. The results showed that drivers' mental workload levels converge between manual and automatic driving modes as the duration of driving increases, regardless of the level of complexity of the mobile phone conversation. This suggests that mobile phone conversations can also disrupt the driver's cognitive resource balance in L2 automatic driving mode, as it increases mental workload while also impairing the normal functioning of brain functions such as cognitive control, problem solving, and judgment, thereby compromising driving safety.

Modeling Driver Responses to Automation Failures With Active Inference

Automated vehicle (AV) technologies promise to improve traffic safety and reduce driver workload in the near future. However, most current implementations have limited capabilities and require transition of control between the vehicle and the human driver during automation failures. For this reason, models of driver behavior have been widely studied to assist the design and development of AV technologies. Recent works have shown that driver behavior models grounded in human cognitive information processing achieve better generalization than prior methods. In this work, we applied active inference, a framework of human perception, cognition, and behavior based on the predictive processing theory, to model driver emergency braking responses to automation failures. We estimated the model parameters from experimental data and examined the model parameters using a factor analysis. We verified the model’s braking response prediction capability in counterfactual scenarios. Our results show that the model effectively captured braking reaction times and provided insight on the correspondence between the variations in driver parameters and behavior.

A Study of Driver Eye-Glancing Behavior at Two Different Types of Unsignalized Intersections Based on SHRP2 NDS Data

The study aimed to analyze driver visual workload at unsignalized intersections with wide medians on high-speed divided highways. The study focused on two types of traffic movements: direct left-turns from minor roads at conventional intersections and right-turns followed by U-turn movements at restricted crossing U-turn (RCUT) intersections. A total of 430 left-turn trips and 40 right-turn followed by U-turn trips were collected from the Second Strategic Highway Research Program Naturalistic Driving Study database. Researchers analyzed the data from driver eye-glancing videos, demographic information, and driving history. The entropy rate of each trip was calculated as an indicator of the driver's visual workload and was treated as the dependent variable for the statistical analysis. Entropy rate is the metric of the randomness associated with drivers’ scanning patterns. The higher the entropy rate, the higher the workload. A comparative study of these two movements at these two types of intersections was conducted. Statistical analyses indicated that drivers at RCUT intersections engaged in less random scanning and longer average fixation and spent more than 70% of the time looking forward during the whole movement. Younger drivers at both types of intersections had higher entropy rates. Additionally, drivers at conventional intersections with higher annual average daily traffic (≥30,000) had higher entropy rates. These results could provide a better understanding of driver workload and its safety implications at different types of unsignalized intersections.

Eye-Gaze Metrics for Cognitive Load Detection on a Driving Simulator

Automated driving systems (ADSs) are becoming ubiquitous to reduce the workload of drivers and improve road safety. However, present-day ADS lacks accurate and effective driver monitoring systems. Driver monitoring systems use physiological measurements, such as pupil dilation, eye-gaze, and eye-blinks, in order to monitor the cognitive load experienced by the drivers. With advances in eye-tracking technology, pupil dilation is emerging as a reliable measure of cognitive load in ADS. However, pupil dilation as a measure of cognitive load suffers from many factors, such as confounding effects, noise, and personal attributes, to name a few. Hence, in order to improve cognitive load estimation in ADS, other noninvasive measures must be studied and incorporated. In this article, various eye-gaze metrics are studied and evaluated as a measure of cognitive load based on data collected from 16 drivers in a simulated driving scenario using a driving simulator.

Large-scale dockless bike sharing repositioning considering future usage and workload balance

Dockless bike sharing (DBS) has the free-floating feature and large-scale fleet, which poses significant challenges to repositioning. This paper finds the Matthew mechanism of bike floating based on real-world trip data in Nanjing, China, and then proposes an integrated method for the large-scale DBS repositioning problem. Firstly, trip origins are analyzed by the first layer clustering to identify thousands of virtual stations. Secondly, the novel index of bike density and turnover rate are used in target inventory estimation for higher future usage. Lastly, the second-level clustering algorithm assigns workload-balanced tasks to truck drivers, and an advanced neighborhood search algorithm is applied to design the truck route. Results show that there are about three hundred pick-up stations and about three thousand drop-off stations for Nanjing DBS repositioning. And the proposed method is demonstrated to provide a better solution for increasing future usage and improving driver workload balance.

GAMBARAN BEBAN KERJA MENTAL DENGAN PRODUKTIVITAS KERJA PADA DRIVER OJEK ONLINE DI KOTA MEDAN

Online motorcycle taxis are one of the public transportations that are currently being used more and more because they are considered effective, besides that online motorcycle taxi drivers must always be ready under any circumstances and conditions to deliver and serve consumers so that it can cause work stress both physically and psychologically (mentally). A high workload will affect the performance and productivity of online motorcycle taxis, many factors cause the high mental workload experienced by online motorcycle taxis. The purpose of this study is to describe the mental workload of online motorcycle taxi drivers in Medan City. This research is a type of descriptive research using a qualitative approach in the form of a interviews. The sample was taken, namely some online motorcycle taxi drivers in the city of Medan. The results of the study show an overview of the workload and work productivity of online motorcycle taxi drivers in Medan City which is relatively low with increasing employee productivity. It was concluded that the description of Workload with Work Productivity on Gojek Online Drivers in Medan City is relatively low so that it supports increasing work productivity for online Gojek drivers because of the motivation from the family for drivers to continue to be enthusiastic at work.c

Decision Support System Assessment Of Truck Driver Work Mental Load in Giwangan Market Area, Yogyakarta Using NASA-TLX

Traditional markets in Indonesia have experienced a decline in performance as from 2002 to 2013. Sales value in 2002 was 74.8%, in 2005 the sales value of traditional markets was 67.6%, then in 2011 it was 55.8%. Giwangan Market is the largest traditional market in the area of Yogyakarta Province. However, many of the traditional markets have inadequate infrastructure both in terms of cleanliness and tidiness of the market location which is detrimental to truck drivers in these markets. Research shows that driver fatigue is the cause of road accidents by 30%. Regarding the measurement of mental workload, subjective measures of workload are easy to provide and have high assessment ability because the measurement is independent of the task. The Decision Support System Model can provide input to Giwangan Market managers to show the mental workload scale of truck drivers using the NASA-TLX Scale (Task Load Index) approach, the most widely used subjective scale by asking participants to rank separately on the mental command subscales. demand, physical demand, temporal demand, own performance, effort, and frustration level. The results of the system show that the mental workload of truck drivers in Giwangan Yogyakarta Market has a very high workload, as many as 4 drivers, 5 drivers have a high workload and 1 driver has a fairly high workload interpretation.

Detection of the Driver’s Mental Workload Level in Smart and Autonomous Systems Using Physiological Signals

With the continuous advancement of automation technology, autonomous driving assistance systems are gradually sharing the tasks during driving, but the driver still assumes the main driving tasks. In addition to driving activities, the advent of numerous new functions will have an indirect impact on the driver’s mental effort. However, determining the driver’s mental effort remains a difficult issue. In this paper, a method is proposed to assess the mental workload of drivers, combining real driver’s physiological data with the speed of his/her vehicle. The correlation coefficient and significance level are obtained by analyzing the correlation between physiological data and road types. The relevant data is then preprocessed to determine the characteristic index, with the mental workload as the input index. The driver’s mental workload is classified and the mental workload prediction model is constructed on the basis of the combination of the Fuzzy Pattern Recognition Algorithm and Genetic Algorithm. At the same time, the suggested approach is compared to the J48 Classification Algorithm and the Simulated Annealing Optimization Algorithm. The results demonstrate that the proposed method in this paper’s effectiveness for identifying the driver’s mental workload level is evidently better than other algorithms, which provides new theoretical support for assessing the L3+ driver’s mental workload level under the background of the safety of the intended functionality when they take over the control of the drive.

Evaluation of drivers' mental workload based on multi-modal physiological signals

Accurately assessing the driver's mental workload is of great significance to reduce the traffic accidents caused by the driver's mental overload. This study aims to evaluate drivers' mental workload in simulated typical driving scenarios, with <italic>N</italic>-back cognitive tasks used to manipulate varied levels of task difficulty. We collect data on multi-modal physiological signals including electroencephalogram (EEG), electrocardiogram (ECG), and electrodermal activity (EDA) signals, and subjective mental load of the National Aeronautics and Space Administration task load index (NASA_TLX) during the task completion process of the driver in the experiment, and propose a series of mental workload classification models based on feature analysis and pattern recognition of the multi-modal physiological signals. These classification models are verified by machine learning algorithms of random forest, decision tree and <italic>k</italic>-nearest neighbor models. The results show that the accuracy of classification models varies with different modalities of physiological signals. EEG-based classification models yield the highest accuracy among single-modal classification models, followed by EDA-based and ECG-based models. Multi-modal-based classification models generally perform better than single-modal classification models. The random forest classification algorithm based on three-modal physiological signals of EEG, ECG and EDA has the highest classification accuracy.

Physiological indicators of driver workload during car-following scenarios and takeovers in highly automated driving

This driving simulator study, conducted as a part of Horizon2020-funded L3Pilot project, investigated how different car-following situations affected driver workload, within the context of vehicle automation. Electrocardiogram (ECG) and electrodermal activity (EDA)-based physiological metrics were used as objective indicators of workload, along with self-reported workload ratings. A total of 32 drivers were divided into two equal groups, based on whether they engaged in a non-driving related task (NDRT) during automation (SAE Level 3) or monitored the drive (SAE Level 2). Drivers in both groups were exposed to two counterbalanced experimental drives, lasting ∼ 18 min each, of Short (0.5 s) and Long (1.5 s) Time Headway conditions during automated car-following (ACF), which was followed by a takeover that happened with or without a lead vehicle. Results showed that driver workload due to the NDRT was significantly higher than both monitoring the drive during ACF and manual car-following (MCF). Furthermore, the results indicated that a lead vehicle maintain a shorter THW can significantly increase driver workload during takeover scenarios, potentially affecting driver safety. This warrants further research into understanding safe time headway thresholds to be maintained by automated vehicles, without placing additional cognitive or attentional demands on the driver. Our results indicated that ECG and EDA signals are sensitive to variations in workload, which warrants further investigation on the value of combining these two signals to assess driver workload in real-time, to help future driver monitoring systems respond appropriately to the limitations of the driver, and predict their performance in the driving task, if and when they have to resume manual control of the vehicle after a period of automated driving.

배송 기사의 근무환경 개선을 위한 권역 설정 기법 연구

Purpose Recently, customer consumption patterns are changing due to the influence of COVID-19, which is increasing the amount of delivery and the increased workload of delivery drivers. The purpose of this study is to propose a regional setting technique to improve the poor working environment of delivery drivers and secure convenience. Methods In the delivery stage, after Clustering, the Routing proceeds. The scope of this study is the stage of setting the area before Routing. After securing the convenience of the delivery driver using Hungarian Algorithm, the density improvement was performed by applying the Geometric Median method. Results It was confirmed that it was possible to set the area closely to the Depot for each delivery driver, and it was confirmed that the density of the area was improved compared to the application of the K-Mean Clustering technique. Conclusion Through biased research, research is actively being conducted to secure efficiency in terms of logistics operation, such as reducing delivery distance and manpower. However, additional research needs to be conducted at the same time to improve the poor working environment of delivery drivers.

Estimating the Workload of Driving Using Video Clips as Anchors

&lt;div class="section abstract"&gt;&lt;div class="htmlview paragraph"&gt;As new technology is added to vehicles and traffic congestion increases, there is a concern that drivers will be overloaded. As a result, there has been considerable interest in measuring driver workload. This can be achieved using many methods, with subjective assessments such as the NASA Task Loading Index (TLX) being most popular. Unfortunately, the TLX is unanchored, so there is no way to compare TLX values between studies, thus limiting the value of those evaluations. In response, a method was created to anchor overall workload ratings. To develop this method, 24 subjects rated the workload of clips of forward scenes collected while driving on rural, urban, and limited-access roads in relation to 2 looped anchor clips. Those clips corresponded to Level of Service (LOS) A and E (light and heavy traffic) and were assigned values of 2 and 6 respectively. Subjects said if they would perform any of 3 tasks—dialing a phone, manually tuning a radio, or entering a destination—while they were driving in each test scene. Finally, subjects provided other ratings for a wider range of situations (not shown in clips) to further examine the effects of road geometry, traffic, and other factors on workload. For each clip, vehicle dynamics data were available for the driven vehicle and the relationship of that vehicle to those in the scene. Using regression analysis, the rated workload (averaged across subjects) was estimated to be 8.87 - 3.01(LogMeanRange)+ 0.48(MeanTrafficCount)+ 2.05(MeanLongitudinalAcceleration), which accounted for 87% of the ratings, an extremely high value. In that expression, range is the distance from the subject’s front bumper to the lead vehicle rear bumper, mean traffic count is the mean number of vehicles visible in the road scene to radar, and the mean acceleration is that of the subject vehicle.&lt;/div&gt;&lt;/div&gt;

Subjective assessment for an advanced driver assistance system: a case study in China

Purpose This study aims to propose a novel subjective assessment (SA) method for level 2 or level 2+ advanced driver assistance system (ADAS) with a customized case study in China. Design/methodology/approach The proposed SA method contains six dimensions, including perception, driveability and stability, riding comfort, human–machine interaction, driver workload and trustworthiness and exceptional operating case, respectively. And each dimension subordinates several subsections, which describe the corresponding details under this dimension. Findings Based on the proposed SA, a case study in China is conducted. Six drivers with different driving experiences are invited to give their subjective ratings for each subsection according to a predefined rating standard. The rating results show that the ADAS from Tesla outperforms the upcoming electric vehicle in most cases. Originality/value The proposed SA method is beneficial for the original equipment manufacturers developing related technologies in the future.

Analysis on hazard perception ability of drivers in plateau areas: by different elevations

The low-pressure and low-oxygen environment of plateaus results in a greater workload for drivers, which contributes to the serious road safety situation in plateau areas in China. This study conducted four hazard perception experiments in Nanjing (50 m above sea level (asl)), Nyingchi (2995 m asl), Lhasa (3650 m asl), Nagqu (4460 m asl), and Yanghu Scenic Spot (4998 m asl) using UC-WIN/ROAD driving simulation software. A total of 31 drivers (23 males, 8 females) were recruited in this study, with a mean age of 28.0 years old and mean driving experience of 6.5 years. A bivariate correlation test was adopted to analyze the impacts of the elevation, age, acclimation period, gender, and driving experience on the perception–reaction time. Then, the drivers were divided into three groups (Good, Medium, and Poor) using K-means clustering. Finally, the marginal effects of the linear regression models for the three groups were calculated to augment the comparison. As expected, the elevation showed a positive correlation with the perception–reaction time but indicated a variance in influence effects in the three groups. Consistent with previous evidence, experienced drivers had better perception ability in the plateau. These findings could help in the design of high-elevation roads and guide the definition of suitability to drive in the plateau, thereby improving driving safety.

Analyzing the Influencing Factors and Workload Variation of Takeover Behavior in Semi-Autonomous Vehicles.

There are many factors that will influence the workload of drivers during autonomous driving. To examine the correlation between different factors and the workload of drivers, the influence of different factors on the workload variations is investigated from subjective and objective viewpoints. Thirty-seven drivers were recruited to participant the semi-autonomous driving experiments, and the drivers were required to complete different NDRTs (Non-Driving-Related Tasks): mistake finding, chatting, texting, and monitoring when the vehicle is in autonomous mode. Then, we introduced collision warning to signal there is risk ahead, and the warning signal was triggered at different TB (Time Budget)s before the risk, at which time the driver had to take over the driving task. During driving, the NASA-TLX-scale data were obtained to analyze the variation of the driver’s subjective workload. The driver’s pupil-diameter data acquired by the eye tracker from 100 s before the TOR (Take-Over Request) to 19 s after the takeover were obtained as well. The sliding time window was set to process the pupil-diameter data, and the 119-s normalized average pupil-diameter data under different NDRTs were fitted and modeled to analyze the variation of the driver’s objective workload. The results show that the total subjective workload score under the influence of different factors is as follows: obstacle-avoidance scene > lane-keeping scene; TB = 7 s and TB = 3 s have no significant difference; and mistake finding > chatting > texting > monitoring. The results of pupil-diameter data under different factors are as follows: obstacle-avoidance scene > lane-keeping scene; TB = 7 s > TB = 3 s; and monitoring type (chatting and monitoring) > texting type (mistake finding and texting). The research results can provide a reference for takeover safety prediction modeling based on workload.

Off-Site Construction Three-Echelon Supply Chain Management with Stochastic Constraints: A Modelling Approach

Off-site construction is becoming more popular as more companies recognise the benefits of shifting the construction process away from the construction site and into a controlled manufacturing environment. However, challenges associated with the component supply chain have not been fully addressed. As a result, this study proposes a model for three-echelon supply chain supply management in off-site construction with stochastic constraints. In this paper, multiple off-site factories produce various types of components and ship them to supplier warehouses to meet the needs of the construction sites. Each construction site is directly served by a supplier warehouse. The service level for each supplier warehouse is assumed to be different based on regional conditions. Because of the unpredictable nature of construction projects, demand at each construction site is stochastic, so each supplier warehouse should stock a certain number of components. The inventory control policy is reviewed regularly and is in (R, s, S) form. Two objectives are considered: minimising total cost while achieving the desired delivery time for construction sites due to their demands and balancing driver workloads during the routeing stage. A grasshopper optimisation algorithm (GOA) and an exact method are used to solve this NP-hard problem. The findings of this study contribute new theoretical and practical insights to a growing body of knowledge about supply chain management strategies in off-site construction and have implications for project planners and suppliers, policymakers, and managers, particularly in companies where an unplanned supply chain exacerbates project delays and overrun costs.

Relevant Physiological Indicators for Assessing Workload in Conditionally Automated Driving, Through Three-Class Classification and Regression

In future conditionally automated driving, drivers may be asked to take over control of the car while it is driving autonomously. Performing a non-driving-related task could degrade their takeover performance, which could be detected by continuous assessment of drivers' mental load. In this regard, three physiological signals from 80 subjects were collected during 1 h of conditionally automated driving in a simulator. Participants were asked to perform a non-driving cognitive task (N-back) for 90 s, 15 times during driving. The modality and difficulty of the task were experimentally manipulated. The experiment yielded a dataset of drivers' physiological indicators during the task sequences, which was used to predict drivers' workload. This was done by classifying task difficulty (three classes) and regressing participants' reported level of subjective workload after each task (on a 0–20 scale). Classification of task modality was also studied. For each task, the effect of sensor fusion and task performance were studied. The implemented pipeline consisted of a repeated cross validation approach with grid search applied to three machine learning algorithms. The results showed that three different levels of mental load could be classified with a f1-score of 0.713 using the skin conductance and respiration signals as inputs of a random forest classifier. The best regression model predicted the subjective level of workload with a mean absolute error of 3.195 using the three signals. The accuracy of the model increased with participants' task performance. However, classification of task modality (visual or auditory) was not successful. Some physiological indicators such as estimates of respiratory sinus arrhythmia, respiratory amplitude, and temporal indices of heart rate variability were found to be relevant measures of mental workload. Their use should be preferred for ongoing assessment of driver workload in automated driving.

Modeling and analysis of driver behaviour under shared control through weighted visual and haptic guidance

Driver–automation shared control through haptic steering is developed to reduce the workload of drivers while keeping them in the control loop. For the optimum design of a haptic steering system, an understanding of driver behaviour based on measurements and modeling is crucial early in the development process. This paper aims to propose a driver model using a weighting process of visual guidance from the road ahead and haptic guidance from a steering system for a lane-following task. The proposed weighting process describes the interaction and reliance of drivers in relation to haptic guidance steering. A driving simulator experiment is conducted to identify the model parameters for driving manually and with haptic guidance. The proposed driver model matched the driver input torque with reasonable fitness among 14 participants after considering individual differences. The validation results reveal that the output trajectory from the model effectively followed the driving course by matching the measured trajectory from the drivers. The model evaluation results reveal the potential of the proposed driver model to be applied to the design and evaluation of haptic guidance systems.

Effect of Non-urban Two Lane Highway Geometry on Car and Bus Drivers – A Physiological Study

Life is a complex phenomenon, so is the task of driving. Mostly human performance is controlled by sympathetic and parasympathetic systems functioning in human body. It is quite dynamic in response to numerous stimuli the driver acquires from the human-vehicle-environmental ensemble. Geometry is one of the fundamental stimulus for a driver, driving on a non-urban highway, especially when there is minimal interference from other vehicles. Consequently, it is quite natural that drivers adopt higher speeds. His/her mental workload will be based on the input he/she gets from the above system and the rate of reception of input will be more with speed. Any inability to cope with the workload may lead to occurrence of crashes. Hence, it is very important to take into account the influence of geometry on driver workload while designing highways. One way to assess the workload is to capture the physiological responses of a driver like heart rate, galvanic skin resistance and rate of eye blinking, during driving.The study included driving experiments done on 114 horizontal curves of gradient less than 2 percentage, each curve being driven over by 30 car and bus drivers. The effect of geometry on workload of car and bus drivers were further analysed. It was found that driver physical characteristics like, age, driving experience, depth perception, vision acuity and reaction time were not found to significantly influence the workload. But workloads of car and bus drivers are significantly different. Sight distance and width of shoulder affect the workload of car drivers. Sight distance was the significant variable affecting bus driver workload. Workload of a driver is found to increase with increase in visibility as increased visual sensations actuates more physiological responses on a human body.

Driver State Modeling Through Latent Variable State Space Framework in the Wild

Analyzing the impact of the environment on drivers’ stress level and workload is of high importance for designing human-centered driver-vehicle interaction systems and to ultimately help build a safer driving experience. However, driver’s state, including stress level and workload, are latent variables that cannot be measured on their own and should be estimated through sensor measurements such as psychophysiological measures. We propose using a latent-variable state-space modeling framework for driver state analysis. By using latent-variable state-space models, we model drivers’ workload and stress levels as latent variables estimated through multimodal human sensing data, under the perturbations of the environment in a state-space format and in a holistic manner. Through using a case study of multimodal driving data collected from 11 participants, we first estimate the latent stress level and workload of drivers from their heart rate, gaze measures, and intensity of facial action units. We then show that external contextual elements such as the number of vehicles as a proxy for traffic density and secondary task demands may be associated with changes in driver’s stress levels and workload. We also show that different drivers may be impacted differently by the aforementioned perturbations. We found out that drivers’ latent states at previous timesteps are highly associated with their current states. Additionally, we discuss the utility of state-space models in analyzing the possible lag between the two latent variables of stress level and workload, which might be indicative of information transmission between the different parts of the driver’s psychophysiology in the wild.