Search

The objective of this study was to verify the effectiveness of eye-tacking metrics in indicating driver’s mental workload in semi-autonomous driving when the driver is engaged in different non-driving related tasks (NDRTs). A driving simulator was developed for three scenarios (high-, medium-, and low-mental workload presented by SAE (Society of Automotive Engineers) Levels 0, 1, and 2) and three uni-modality secondary tasks. Thirty-six individuals participated in the driving simulation experiment. NASA-TLX (Task Load Index), secondary task performance, and eye-tracking metrics were used as indicators of mental workload. The subjective rating using the NASA-TLX showed a main effect of autonomous level on mental workload in both visual and auditory tasks. Correlation-matrix calculation and principal-component extraction indicated that pupil diameter change, number of saccades, saccade duration, fixation duration, and 3D gaze entropy were effective indicators of a driver’s mental workload in the visual and auditory multi-tasking situations of semi-autonomous driving. The accuracy of predicting the mental-workload level using the K-Nearest Neighbor (KNN) classifier was 88.9% with bootstrapped data. These results can be used to develop an adaptive multi-modal interface that issues efficient and safe takeover requests.

Transfer learning (TL) is a machine learning (ML) method in which knowledge is transferred from the existing models of related problems to the model for solving the problem at hand. Relational TL enables the ML models to transfer the relationship networks from one domain to another. However, it has two critical issues. One is determining the proper way of extracting and expressing relationships among data features in the source domain such that the relationships can be transferred to the target domain. The other is how to do the transfer procedure. Knowledge graphs (KGs) are knowledge bases that use data and logic to graph-structured information; they are helpful tools for dealing with the first issue. The proposed relational feature transfer learning algorithm (RF-TL) embodies an extended structural equation modelling (SEM) as a method for constructing KGs. Additionally, in fields such as medicine, economics, and law related to people’s lives and property safety and security, the knowledge of domain experts is a gold standard. This paper introduces the causal analysis and counterfactual inference in the TL domain that directs the transfer procedure. Different from traditional feature-based TL algorithms like transfer component analysis (TCA) and CORelation Alignment (CORAL), RF-TL not only considers relations between feature items but also utilizes causality knowledge, enabling it to perform well in practical cases. The algorithm was tested on two different healthcare-related datasets — sleep apnea questionnaire study data and COVID-19 case data on ICU admission — and compared its performance with TCA and CORAL. The experimental results show that RF-TL can generate better transferred models that give more accurate predictions with fewer input features.

The safety of socio-technical systems is intractable by traditional approaches, such as a simple cause-effect analysis. It would be no longer effective only to eliminate potential hazards or to explain why things go wrong (i.e., the limitation of the approach called Safety-I). The safety instead could be improved by validating why everyday operations go right and by verifying their resilience, based on an alternative idea of Safety-II. The objective of this research is to develop a tool enabling us to validate and verifying safeties based on the functional resonance analysis method (FRAM). FRAM would be an effective way to analyze the safety of complex socio-technical systems. However, FRAM only provides a conceptual methodology, and further extensions and implementations are needed. This work first presents an extended FRAM model based on a classical idea of cellular automaton, and the result of applying our model to a steel production management problem is provided. Steel production line reveals a quite complicated process consisting of many linked workstations where a production flow might be varied frequently. Wherein flexible delivery decisions are required, including changing the supply chains themselves. There exists empirical know-how to tackle those, but these remain as tacit knowledge, and there has been no way to validate and verify their effectiveness under specific conditions. The results of applying our extended FRAM model to this example provide several insights concerning the characteristics of experienced workers’ operations to handle and manage the process’s complexities, such as harnessing, phase transformation, and identifying critical points attaining resilient operations in terms of the entropy of the process status. We also discuss how these findings contribute to the safety management of the other socio-technical systems based on the findings of the analysis.

FRAM: Functional Resonance Analysis Method is an effective way to know about the safety of complex Socio-Technical Systems. However, it is a method rather than model and required to be extended for its practical usage. According to this, in this paper, a latest version of FRAM simulator based on our developed model is presented. The model simulates a process in which variabilities exiting in a working environment induce variability of FRAM functions that emerge out of the dynamic interactions among the functions as well as with the environment. Moreover, the model simulates a process where a specific context composed of variabilities existing in a working environment "shakes" FRAM functions, while the context is "shaken" by those functions vice versa, which is a typical dynamics specific to complex adaptive systems. This is implemented by integrating FRAM and Fuzzy CREAM which is an extended model of CREAM: Cognitive Reliability and Error Analysis Method with fuzzy reasoning. It enables to parameterize the variabilities, define the context, and formulate their interactions quantitatively, whose result is given as a dynamical change of state in each FRAM function.