Human Factors and Ergonomics in Manufacturing & Service Industries最新文献

Anthropomorphic avatars serve as a key tool to humanizing chatbots. This study aimed to investigate the effects of four specific anthropomorphic visual design—human avatars, caricatured avatars, zoomorphic avatars, and functional avatars—on users' first impressions of customer service chatbots. A combination of subjective evaluation and eye-tracking methods were employed to capture users' initial responses. The results revealed that the anthropomorphic nature of chatbot avatars influences users' perceptions of whether the chatbot is human or artificial intelligence, with more human-like designs increasing the likelihood of being categorized as human. Both human and caricatured chatbot avatars enchanced perceptions of warmth, competence, social presence and trust, whlie also generating higher fixation counts and longer fixation durations compared to the other two avatar types. In conclusion, anthropomorphic avatar play a critial role in shaping users' subjective impressions and visual engagement with customer-service chatbots. A key implication of this study is that chatbot avatars should incorporate a high degree of anthropomorphism to foster positive first impressions among users. However, given the privacy concernss associated with using real photos, it is not necessary to use images of actual individuals. Instead, caricatured avatars can effectively enhance users' first impressions while mitigating potential drawbacks of hyper-realistic designs, such as uncanny valley effects or mismatched user expectations. Furthermore, this study demonsterates the feasibility of using eye-tracking tools to evaluate avatar designs in customer-service chatbots.

Considering the COVID-19 pandemic, understanding the behavior of social groups on crosswalks has become increasingly important. This field study aims to investigate the dynamic behavior of pedestrian groups crossing streets during the pandemic. By analyzing the movement patterns and behaviors of group under different conditions, we aim to provide insights into how crosswalk infrastructures can be improved for a safer and more efficient pedestrian experience during the pandemic. An observation experiment was conducted at a crosswalk near Shinjuku Station in Japan in 2021. Trajectories of 296 groups were analyzed with up to five members, examining their velocity characteristics and spatial relations. Our findings reveal several important insights. First, the group size affects the average speed and speed difference of group members, and that the width of the crosswalk is positively correlated with the average speed of group members. Second, the group size is uncorrelated with mean inter-distance of members, regardless of group size. However, the farther members are from the leader, the greater the average distance between adjacent members. Third, distinct group shapes for groups of different sizes. Groups of three people form a V shape, while groups of four people form a U shape, and groups of five people form a trapezoid shape. Finally, the group size is correlated with the average offset angle of group members in the northbound scenario. Larger groups tend to have smaller average offset angles, making them more likely to choose the shortest route. Overall, this study is a crucial step in developing safety-oriented walking modeling tools for pedestrians at intersections. It also has important implications for predicting pedestrian crossing behavior and show how human factors and ergonomics methods can be used to improve urban public infrastructure systems in densely populated countries, such as Japan.

Seated sleep during flight reduces the frequency of postural adjustment, which increases low limb fatigue and discomfort. The purpose of this study was to explore the mechanism by which leg postural adjustment relieves low limb fatigue and discomfort during sedentary sitting. Four postural tasks were set for the mechanism study (each lasted 40 min). Task 1: Bending the knee all through, quickly extending the knee once at the 20th minute. Task 2: Extending the knee all through, quickly flexing the knee once at the 20th minute. Task 3: Flexing the knee for 20 min, then switching to extension until the end. Task 4: Extending the knee for 20 min, then switching to flexion until the end. During tasks, electromyography, muscle oxygen, and subjective discomfort scores were recorded to assess the development of fatigue and discomfort in the lower limbs. The data from the first and latter halves of each task, and 4 min after the switch, were extracted, for a total of three periods of data. The results showed that continuous maintenance of one posture gradually increased fatigue and discomfort in the lower limbs, whereas posture changes relieved discomfort and fatigue. Under different leg postures, four leg muscles are in alternating passive stretch and fatigue development. Moreover, instantaneous switching behavior has a positive effect on reducing subjective discomfort. This study helps elucidate the mechanism by which leg posture adjustment relieves fatigue and subjective discomfort.

Drilling accidents often occur due to inadequate situational awareness of drilling workers. The prolonged and tedious monitoring of drilling parameters may result in drilling workers experiencing mind-wandering. To explore this problem, this study employed wearable eye tracking technology to assess the mind-wandering status of drilling workers during work finishing (WF) and investigate the influence of mind-wandering on the situational awareness of these workers. A simulation of monitoring drilling parameters was conducted in the laboratory by 18 drillers, who were asked to fill out situational awareness questionnaires using the global assessment of situational awareness (SAGAT) and had their behavioral and eye movement data collected during the task. The eye movement results revealed that drilling workers experienced mind-wandering during the WF, as evidenced by a significant increase in fixation duration and count, along with a decrease in pupil diameter. This led to a decrease in situational awareness scores and a notable increase in reaction times, indicating that mind-wandering had an impact on the situational awareness of drilling workers. In addition, the eye movement areas of interest (AOIs) were further analyzed to explore drilling worker attention allocation. This study showcases the potential of using wearable eye-tracking technology to identify mind-wandering in drilling workers who are responsible for monitoring drilling parameters. By gaining a deeper insight into the eye movement patterns associated with mind-wandering in drilling workers, we can establish a strong basis for creating interventions that enhance their situational awareness.

Reliable assessment of team performance based on teamwork characteristics is crucial for optimizing maritime teamwork and ensuring maritime safety. However, the relationship between teamwork characteristics and team performance remains unclear. This study explored teamwork characteristics that contribute to good team performance, and constructed a team performance assessment model based on teamwork characteristics. A total of 84 qualified participants were recruited and divided into 21 teams to perform underwater collaborative tasks on a maritime simulator. Eye movement, communication, scale, and team performance data were collected and divided into input features and outcome features. Based on input features, important teamwork characteristics were obtained, showing that teams with good team performance presented adequate situation awareness for judgment and decision, excellent task management, appropriate decision making, and efficient communication. In addition, machine learning algorithms were applied to train classification models, and the support vector machine (SVM) exhibited the best results, with an accuracy of 83.3%. The team performance assessment model offers an effective auxiliary tool for maritime training to evaluate training effects as well as create more efficient training programs, and is beneficial for optimizing teamwork skills, improving team performance, and ensuring maritime safety.

Pilot errors account for the majority of flight accidents, many of which are influenced by mental workload. This study introduces a data-driven psychophysiological model, utilizing machine learning algorithms, to evaluate pilots' mental workload. We conducted a flight-simulation experiment involving twenty commercial airline pilots, assessing their mental workload under varying levels of task demand and visibility conditions. Psychophysiological responses were recorded, and machine learning algorithms were employed to analyze the data. To evaluate model performance, we used a leave-one-subject-out cross-validation method and calculated area under the curve values. Our findings indicate that psychophysiological metrics vary in their sensitivity to changes in pilots' mental workload. Notably, the Gradient Boosting Decision Tree algorithm demonstrated the highest classification performance under high-visibility conditions, while the Gaussian Naive Bayes algorithm excelled under low-visibility conditions. These results suggest pilots' mental workload can be effectively identified through psychophysiological metrics combined with machine learning algorithms. Furthermore, visibility conditions may influence the model's classification performance. This model offers a complementary approach to the subjective evaluation currently used by flight instructors to assess pilots' mental workload management capabilities during flight training and certification. It also provides a data-driven tool aligned with evidence-based training principles, enhancing the evaluation of pilots' mental workload management capabilities in flight scenarios.