Learning from accidents is decisive for establishing barriers to prevent future accidents and ensuring safety in socio-technical systems. Farmed fish escape is a major concern in the aquaculture industry and there is limited understanding of the accident scenarios involving fish escape. Hence, the systematic analysis of information and data about these types of near accidents and accidents may reveal contributing causes. The objective of this article is to present data and scenarios based on the investigation of 745 fish escape accidents and near accidents reported to the Norwegian Directorate of Fisheries from January 2006 to August 2019. The aim is to derive the direct and underlying causes and develop generic scenarios for fish escape, which can be used for improved reporting of near accidents and accidents. The paper discusses the limitations of current reporting systems, suggests a new categorization scheme for the Directorate of Fisheries to enhance the industry and authorities’ learning from fish escape events, and incorporates new types of fish farms emerging in the industry. Further, the study results can support barrier management and emergency planning, since they help understand the causes of fish escapes. This may be used for analyzing whether existing accident prevention strategies and barriers are sufficient or if additional barriers need to be implemented by fish farm operators.
3,0( !(&+01.1*9 138(*,$0 "0,7(34,59 1) &,(0&( $0' !(&+01.1*9 !" !310'+(,/ 138$9 /$,.,0*3,' %
Current decision making regarding whether to abort a high-risk aquaculture operation in a Norwegian fish farm is mainly experience-driven. The on-site personnel decides whether to start/delay/abort operations primarily based on their subjective judgement about whether they can handle the situation. The risk is considered implicitly as "gut feelings". There are no explicit operational limits nor a structured process to derive these for high-risk operations. In this research, a predefine safety-critical attributes have been identified from major accident scenarios to guide machine learning process to define operational limits based on multi-source data. Bayesian network, Tree Augmented Naïve Bayes (TAN) search algorithms were selected to build up prediction model so that operational limits upon a given condition can be decided. The paper concludes that machine learning techniques have great potential to be used to support safe decision-making in high-risk aquaculture operation, and the risk-based operational limits facilitates better understanding of operational context, and comprehension of the meaning of several deviations which may indicate a dangerous situation.
Enabling higher levels of autonomy requires an increased ability to identify and handle internal faults and unforeseen changes in the environment. This work presents an approach to improve this ability for a robotic system executing a series of independent tasks, such as inspection, sampling, or intervention, at different locations. A dynamic decision network (DDN) is used to infer the presence of internal faults and the state of the environment by fusing information over time. This knowledge is used to make risk-informed decisions enabling the system to proactively avoid failure and to minimize the consequence of faults. Past states are evaluated with new information to identify and counteract previous sub-optimal actions. A case study on an inspection drone tasked with contact-based ultrasound inspection is presented. The case study successfully demonstrates the proposed capabilities while minimizing time use and maximizing mission completion.
No abstract is provided for this article.
In this chapter, definitions of important terms related to online probabilistic risk assessmentOnline probabilistic risk assessment are clarified (Sect. 2.1). In Sect. 2.2, dynamic, real-time, and online risk assessmentsOnline risk assessment are defined. In Sect. 2.3, methods for online risk assessmentOnline risk assessment such as hazard identificationHazard identification, event sequence diagramsEvent sequence diagram (ESDs), fault tree analysisFault tree analysis (FTA), and hybrid methodsHybrid method are presented. In addition, in this subsection, the online risk assessmentOnline risk assessment methodology for autonomousAutonomous systems is more clarified by applying the method in a case studyCase study. In Sect. 2.4, methods for dynamic probabilistic risk assessmentDynamic probabilistic risk assessment of complex systemsComplex systems including dynamic event sequence diagramsDynamic event sequence diagrams (DESDs), Markov chainsMarkov chain, dynamic fault treesDynamic fault tree, and dynamic Bayesian networksDynamic bayesian network are presented and discussed. Finally, in Sect. 2.5, the challengesChallenges of available methods are discussed; and possible solutions to overcome these challengesChallenges are proposed. Integrating predictive and optimizationOptimization algorithmsAlgorithm, as well as simulation, with risk assessmentRisk assessment may result in more powerful risk assessmentRisk assessment methods for complex systemsComplex systems.
No abstract is provided for this article.
Failures of critical infrastructures can represent a threat both to people, economy and societal functions and to national security. So, thorough risk analyses of infrastructures are required to reduce the probability and mitigate the consequences of failures. The interdependencies between infrastructures can be strong, but are seldom accounted for in current analyses. This chapter presents a method for assessing these interdependencies and also provides an example. The analysis is part of an overall cross-sector risk and vulnerability analysis (RVA), see Chap. 3.
This chapter presents three case studies in the data leakage domain and the methods proposed and evaluated for mitigating the threat of data leakage. The case studies are: detecting an insider attempting to misuse and leak data stored in a database system; using honeytokens to detect insider threats; and detecting leakage through email.
No abstract is provided for this article.
<p>Enabling higher levels of autonomy requires an increased ability to identify and handle internal faults and unforeseen changes in the environment. This article presents an approach to improve this ability for a robotic system executing a series of independent tasks by using a dynamic decision network (DDN). A case study of an industrial inspection drone performing contact-based inspection is used to demonstrate the capabilities of the resulting system. The case study demonstrates that the system is able to infer the presence of internal faults and the state of the environment by fusing information over time. This information is used to make risk-informed decisions enabling the system to proactively avoid failure and to minimize the consequence of faults. Lastly, the case study demonstrates that evaluating past states with new information enables the system to identify and counteract previous sub-optimal actions. </p>
No abstract is provided for this article.
Technological innovations and new areas of application introduce new challenges related to safety and control of risk in the maritime industry. Dynamically positioned systems are increasingly used, contributing to a higher level of autonomy and complexity aboard maritime vessels. Currently, risk assessment and verification of dynamically positioned systems are focused on technical reliability, and the main effort is centered on design and demonstration of redundancy in order to protect against component failures. In this article, we argue that factors, such as software-requirement errors, human errors, including unsafe or too late decision-making, and inadequate coordination between decision makers, also should be considered in the risk assessments. Hence, we investigate the feasibility of using a systems approach to analyzing risk in dynamically positioned systems and present an adapted version of the system-theoretic process analysis. A case study where the system-theoretic process analysis is applied to a dynamically positioned system is conducted to assess whether this method significantly expands the current view on safety of dynamically positioned systems. The results indicate that the reliability-centered approaches, such as the failure mode and effect analysis, sea trials, and hardware-in-the-loop testing, are insufficient and that their view on safety is too narrow. This article shows that safety constraints can be violated in a number of manners other than component failures for dynamically positioned systems, and hence, system-theoretic process analysis complements the currently applied methods.
Maritime accident statistics are used as a key part of the IMO's formal safety assessment (FSA), a risk assessment methodology to guide policy decisions in the maritime industry.Under-reporting of maritime accidents can inhibit the accuracy of results derived from the FSA, therefore having a direct influence on maritime policy.The objective of this work is to perform comparisons between accident databases, and to investigate the degree to which underreporting is biased by factors including the type of accident, degree of severity, and ship type.This study analyzes databases of reported maritime casualties from 1) IMO GISIS, 2) IHS Fairplay, and 3) the United States Coast Guard CGMIX.The databases are subset to an eight-year period and for commercial ships greater than 100 gross tonnage (GT) to enable a direct comparison.The reporting rates for the GISIS and IHS databases are calculated for accident type, accident severity, and ship type.Results indicate that the GISIS and IHS databases contain significantly fewer non-serious accidents than serious accidents.Further biases were observed by accident and ship types.Founderings, fires / explosions, and strandings are more likely to be reported than other accident modes.Hull / machinery damage is the accident mode with the lowest reporting rate.
Autonomous systems, including airborne, land-based, marine, and underwater vehicles, are increasingly present in the world. One important aspect of autonomy is the capability to process information and to make independent decisions for achieving a mission goal. Information on the level of risk related to the operation may improve the decision-making process of autonomous systems. This article describes the integration of risk analysis methods with the control system of autonomous and highly automated systems that are evaluated during operation. Four main areas of implementation are identified; (i) risk models used to directly make decisions, (ii) use of the output of risk models as input to decision-making and optimization algorithms, (iii) the output of risk models may be used as a constraint in or modifying constraints of algorithms, and (iv) the output of risk models may be used to inform representations or maps of the environment to be used in path planning. A case study on a dynamic positioning controller of an offshore supply vessel exemplifies the concepts described in this article. In addition, it demonstrates how risk model output may be used within a hybrid controller.
