Search

This study explores the application of agent transparency in the context of autonomous ships. Four levels of transparency were developed depicting decisions, planned actions, reasoning, and input parameters of a collision and grounding avoidance system in a realistic navigational context. Thirty-four licensed navigators were provided with Human Machine Interface concepts depicting four levels of transparency. Qualitative feedback was obtained through semi-structured interviews about which information they felt is needed to supervise autonomous ships safely and effectively. In addition, the participants’ ranked the HMIs according to their preferences. The results indicate the need for depicting the outcomes of the system’s collision risk analysis for supervisory control. Furthermore, the results illustrate the variations in supervisory strategies and the resulting dilemma for the amount and type of information required to support supervisors. Finally, this study highlights the importance of expert knowledge in the design of safety critical systems.

ABSTRACTPrevious literature has treated software functionalities as a binary choice between informating and automating, based on Zuboff’s definitions. We suggest that the space between the role of a user engaging with a system (informate) and what the system can accomplish for the user (automate) has increasingly become blurred, moving toward a more nuanced continuum. We present a user-system interaction framework that analyzes this continuum with a focus on the user’s perspective. Using cases from archival data, this paper provides an analysis of what goes well and badly in the locus of control handoff between users and systems, highlighting the implications of software that is created in one region but used globally. The paper contributes to the literature by extending Zuboff’s informate vs. automate discussion and identifying design decisions that could help to avoid potentially catastrophic design errors. Using the interaction framework, we argue that it would be beneficial for system designers to move toward the informate end of the continuum, when designing systems for users with a high level of domain knowledge and expertise, and when there are significant differences between the environment in which the system is designed and the one in which the system is used.

Recent advancements in simulation technology facilitated maritime training in various modalities such as full-mission, desktop-based, cloud-based, and virtual reality (VR) simulators. Each of the simulator modality has unique pros and cons considering their technical capabilities, pedagogical opportunities, and different organizational aspects. On the other hand, enhanced training opportunity and diversity of training depends on the proper utilization of simulators. In this context, the absence of an unbiased, transparent, and robust simulator selection process poses a complex decision-making challenge for the maritime instructors and decision-makers at the institutions. In this study, a hybrid multi-criteria decision-making (MCDM) approach is proposed to evaluate four major types of simulator modalities. For the evaluation, a MCDM framework is developed based on 13 key factors (sub-criteria) for simulator selection grouped under three higher-level criteria—technical, instructional, and organizational criteria. Data was collected using a structured best-worst method (BWM) survey from subject matter experts. The Bayesian BWM is used for ranking of the 13 sub-criteria, and the Preference Ranking Organization Method for Enrichment Evaluation (PROMETHEE) is used to evaluate the four simulator modalities utilizing sub-criteria ranking scores from Bayesian BWM. The results reveal that the regulatory compliance of simulators is the most important criterion, while the cost of simulators is considered the least important criterion during the simulator selection process. Overall, full-mission simulators are the most preferred followed by VR simulators, cloud-based and desktop simulators. However, a sensitivity analysis demonstrated context-specific preferences for certain simulator types over others.

This study discusses an approach to support human supervision of autonomous maritime collision avoidance systems by disclosing the system’s perceived information, internal reasoning, decisions, and planned actions as layers of transparency. Information requirements, identified through a cognitive task analysis, were structured using the information processing model by Parasuraman, Sheridan, and Wickens (2000). This model was contextualized to the maritime collision avoidance setting such that the information from the analysis could be structured into unique and distinct layers. A set of minimum information requirements was identified depicting the system’s decisions and planned action, supported by additional layers to reveal its internal reasoning. This approach aims at supporting humans in effectively supervising autonomous collision avoidance systems in their operational context by providing understandability and predictability about what the system is doing, why it is doing it, and what it will do next, i.e., transparency.

The education and training for the maritime industry require renewed focus in the face of technological changes and increasing digitalization. Artificial intelligence presents an avenue for further research that can positively impact efficiency and competence development. Among many applications of artificial intelligence in education, conversational agents or chatbots have gained increased interest in recent years. This paper describes the design and implementation process of “FLOKI”—a chatbot aimed at assisting maritime trainees in learning Collision Avoidance Regulations (COLREGs). For the design of the chatbot, IBM Watson Assistant®—a cognitive computing service—was utilized, which enables the use of Application Programming Interfaces (APIs) in its cloud server. A selected number (n= 18) of 2nd year B.Sc. in Nautical Science students in a Norwegian maritime university interacted with the chatbot for reflecting on their knowledge about COLREGs. In addition to demographic data, the maritime trainees were asked to answer questions related to user experience utilizing the System Usability Scale (SUS). The findings are discussed along with their implications and future research directions involving AI in maritime education and training.

The COVID-19 pandemic has brought unprecedented challenges to the maritime supply chain and called for accelerated adoption of digital technologies in various aspects of maritime operations, including the area of maritime education and training (MET). This paper aims to discuss the current maritime simulator-based training and educational practices that forms an integral part in seafarer training and competency development. The study provides a review of the existing simulators in use in MET, and discusses upon the technological and pedagogical advancement of maritime simulator-based training interventions with predictions regarding the future MET practices with use of virtual reality and cloud-based simulators. This study—by focusing on ship’s bridge operations—highlights the characteristics of various types of simulators and also discusses the role of instructors, challenges, and opportunities involving future simulator-based MET due to accelerated adoption of digital technologies and the need to comply with pandemic-related restrictions for MET institutes. The analysis generated in the paper may contribute to the ongoing discussion regarding the future of simulator-based MET and the fulfillment of the UN Sustainable Development Goal (SDG) 4 in the maritime sector.

Virtual Reality Head Mounted Displays (VR HMDs) offer highly immersive experiences that may alter users’ awareness and perception of the “real-world”. This paper investigates users’ perception of short duration (<5 minutes) time periods during simulated scenarios between two different simulation mediums. Fifty participants (Age: M=29.2 years; SD=10.7 years; 15 Females; 35 Males) were randomly assigned to one of two groups: (i) VR HMD or (ii) desktop computer configuration, where they performed identical searching tasks within a virtual environment three consecutive times. After each virtual session participants were asked to verbally report their perception of time duration. Results reveal that participants across both groups overestimated time duration in virtual environments by an average of 20.3%. As participants gained experience performing the task repeatedly across three consecutive attempts, their task completion time reduced (i.e. learning effect). However, participants’ accuracy of time perception continued to be consistently overestimated across both conditions.

Modern virtual reality technologies increase the immersive experience for users. Increased immersivity may affect how users perceive, internalize, and thus, recall information about the virtual environment and experience compared to traditional mediums, such as desktop simulations. One critical aspect of recalling information is incidental memory, which refers to the ability for people to recall information they were unaware was relevant for evaluation. This paper investigates the differences of incidental memory recall between (i) immersive virtual reality and (ii) traditional desktop systems. Fifty-two participants completed a series of tasks within identical virtual environments of the two systems and completed an incidental memory recall questionnaire testing their recall of elements within the virtual world. Results found no significant difference in incidental memory recall between immersive virtual reality and traditional desktop conditions. However, when controlling for gaming experience and gender effects significant differences between groups were found.

Working at sea places individuals in an inherently dangerous environment for extended periods, exposing them to unique risks not found in land-based industries. Safety-critical socio-technical systems demand an inherent organizational safety culture for reliable and safe operations. Safety climate acts as a mediating factor between the broader organizational climate and safety behavior of individuals and teams. This paper investigates safety climate of individuals working at sea. Two hundred persons (47.3 yrs ±12.9; 175 males, 25 females) working as seafarers (n=132) and onboard service staff (n=68) with Norwegian maritime companies completed an online safety climate questionnaire. Results indicate that maritime workers generally have lower perceptions of safety within their organizations in comparison to other industries. Furthermore, certified seafarers have lower perceptions of safety then onboard service staff, who have lower restrictions to working at sea, and generally less maritime safety education and training.

Transparency is a design principle intended to make the inner workings of autonomous agents visible to end-users such that humans can evaluate the reasoning behind its decisions and actions. To test the effect of agent transparency on situation awareness, mental workload, and task performance, an experiment was performed where 34 nautical navigators were tasked with interpreting the information provided by an autonomous collision and grounding avoidance system. Sixteen traffic situations were created with two levels of complexity. Four levels of transparency varied the amount and type of information in terms of the system’s decisions, planned actions, reasoning, and input parameters. The results show that increased transparency improves SA without increasing mental workload. However, the time to comprehend the system’s decisions and planned actions increased when its reasoning was depicted. Traffic complexity impaired SA, mental workload, and time-to-comprehension regardless of transparency level. However, for level 2 SA, transparency was found to negate the influence of complexity, resulting in improved comprehension of the agent’s reasoning despite high traffic complexity. These outcomes demonstrate the merits of agent transparency as a design principle in supporting human supervision of autonomous agents. However, developers should take care when extending these principles to time-critical applications.

Simulator-based training has witnessed considerable attention in recent years for the training of operators in the maritime domain and is employed by majority of nautical training institutes. IMO has published the model course 6.10 Train the simulator trainer and assessorwhich acts as a guideline and aims to promote uniformity in the simulator-based training for maritime operations. Nevertheless, the model course should not be “implemented blindly,” since one has to acknowledge the institution’s own resources and apply them in an appropriate way. Consequently, a variation in different institutions simulator training can exist. This study aims to discover how such variations exist in the European full mission simulator training institutions. Semi-structured interviews were conducted to clarify each of the participating institution’s simulator training design. The interviews comprised relevant performance indicators—e.g., identical elements, training time, teaching of general principles, students’ needs, feedback, training needs analysis, assessment, and instructors´ qualifications—selected after a detailed literature review. The findings present variations and similarities observed in the European simulator training facilities, in relation to the designated performance indicators. The study demonstrated that although some simulator training in Europe appears to be performed uniformly due to comparable proceedings and understanding of the model course 6.10, the implementations of these aspects on the basis of interpretation and available infrastructure can create dissimilarities.

The maritime industry is considered to be a backbone of the global economy. It is therefore imperative to ensure that maritime operations run safely and efficiently. Assessment of maritime performance is necessary for designers and engineers to be able to pinpoint the weakest links in the system and make impactful system improvements. The current article presents a systematic quantitative literature review of research on performance assessment in the maritime industry with the goal of establishing an understanding of accuracy and consistency in the development of methods used to assess performance. The review focuses on four major segments within the industry—port logistics, ship handling, safety and environmental research—and investigates their uses in developing accurate and consistent performance assessment methods. After the completion of an exclusion process, 62 articles published in a wide range of academic journals were used in the analysis. Two important conclusions were drawn from the analysis. First, performance assessment is generally consistent throughout the maritime industry; most papers used accurate and consistent approaches to develop the methods (n= 43). A subsequent bivariate analysis revealed a call for increased attention to the development of assessment methods within the maritime segment of ship handling. The current study suggests and discusses certain directions with regard to assessment research in the maritime industry.

Training simulators are largely deployed to provide operators working within complex systems to instil adequate skills to handle normal and abnormal situations. Improved technology and higher computation power have significantly increased the use of training simulators in the training programs. The goal of this pilot study is to determine the design of training in order to prepare for complex tasks within the maritime domain. In this experimental study, students are trained to perform docking operations when ocean currents with increasing complexity are introduced in the training program. The effect of the training is evaluated by comparing two groups: one trained with homogenous currents and the other with heterogeneous currents. Objective performance measures are used to analyse the participants’ performance. The results indicate that when the participants were exposed to tasks with gradually increasing complexity, they performed better as compared to those who were exposed to complex tasks too early in the training. The results suggest that even though the technology and computational power provide for new possibilities in training simulators, new features that make the tasks more complicated should not be included too early in the training without sufficient investigation. It is also found that increasing functional fidelity of the simulation during training has resulted in the improved performance of the participants during the complex tasks (docking operations), as compared to those training with the highest fidelity from the beginning.

OCCUPATIONAL APPLICATIONSOperators play a vital role in production and safety in industrial processes. Since the introduction of advanced control techniques, such as model predictive control and real-time optimization, operators’ acquisition of adequate mental models to develop complex cause-and-effect relationship explaining plant behavior has been increasingly challenged. Additionally, distinct challenges have arisen with respect to crew coordination between control room and field operators to orchestrate a coordinated flow of actions to assess situations or choose a course of action. Based on an analysis of training needs, it is argued that traditional training practice, such as the use of operator training simulators, could be advanced by using current training environments, such as virtual reality training simulators. This would allow using modern training technology and its advancements in parallel to the advancements of control techniques to support production and safety at its best.TECHNICAL ABSTRACT Background: Extensive integration of various modern methods in the process industry has changed the tasks of industrial operators. The integration of advanced technology and control algorithms lead to new challenges faced by control room and field operators, from both technical and crew-coordination complexity perspectives. From a technical perspective, couplings, dynamic effects, non-transparency, conflicting goals, comprehension of model predictive control, and real-time optimization challenge the development of an accurate mental model. From a crew-coordination complexity perspective, control room operators and field operators face the challenge to orchestrate their individual actions into a coordination flow of actions to assess a situation and solve problems. Purpose:The purpose of this article is to highlight the cognitive and teamwork requirements of operators and to note the limitations of current training practices compared to the training objectives that need to be achieved individually and as a team. Methods:Evidence is presented from instance-based learning theory and theories addressing the acquisition of mental models, instances, and skills for crew-coordination complexity; this is used to suggest that current training practices match only a subset of the challenging training objectives that are essential to use technology efficiently and safely. Results:Findings from the cognitive training need analysis are linked to training objectives and training methods based on the learning theories presented. Additionally, arguments for using different training environments (operator training simulators, virtual reality training simulators) to achieve the training objectives in an optimal way are presented. Conclusions:It is concluded that advancements in the applications of process control techniques call for a new mindset in the training of operators. Advanced training methods and environments can be one way of helping the operator to improve performance reduce errors and enhance safety.

The significant effect of human errors on the operation of industrial processes calls for improved methodologies and tools to train both control-room and field operators. Common operator-training simulators (OTSs) meet almost exclusively the training needs of control-room operators. As for field operators, little has been carried out to improve their training level. The paper presents and discusses a solution for immersive training of industrial operators that allows experiencing the multifaceted scenarios of real-plant operations. [ABSTRACT FROM AUTHOR]