Journal of System Safety最新文献

Proposing the Use of Hazard Analysis for Machine Learning Data Sets 建议在机器学习数据集上使用危害分析

Journal of System Safety Pub Date : 2023-06-22 DOI: 10.56094/jss.v58i2.253

H. Carter, Alexander Chan, Christopher Vinegar, J. Rupert

{"title":"Proposing the Use of Hazard Analysis for Machine Learning Data Sets","authors":"H. Carter, Alexander Chan, Christopher Vinegar, J. Rupert","doi":"10.56094/jss.v58i2.253","DOIUrl":"https://doi.org/10.56094/jss.v58i2.253","url":null,"abstract":"There is no debating the importance of data for artificial intelligence. The behavior of data-driven machine learning models is determined by the data set, or as the old adage states: “garbage in, garbage out (GIGO).” While the machine learning community is still debating which techniques are necessary and sufficient to assess the adequacy of data sets, they agree some techniques are necessary. In general, most of the techniques being considered focus on evaluating the volumes of attributes. Those attributes are evaluated with respect to anticipated counts of attributes without considering the safety concerns associated with those attributes. This paper explores those techniques to identify instances of too little data and incorrect attributes. Those techniques are important; however, for safety critical applications, the assurance analyst also needs to understand the safety impact of not having specific attributes present in the machine learning data sets. To provide that information, this paper proposes a new technique the authors call data hazard analysis. The data hazard analysis provides an approach to qualitatively analyze the training data set to reduce the risk associated with the GIGO.","PeriodicalId":250838,"journal":{"name":"Journal of System Safety","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114954885","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Human Reliability Analysis using a Human Factors Hazard Model 基于人因危害模型的人的可靠性分析

Journal of System Safety Pub Date : 2023-06-22 DOI: 10.56094/jss.v58i2.251

Dustin S. Birch, Erika E. Miller, Thomas H. Bradley

{"title":"Human Reliability Analysis using a Human Factors Hazard Model","authors":"Dustin S. Birch, Erika E. Miller, Thomas H. Bradley","doi":"10.56094/jss.v58i2.251","DOIUrl":"https://doi.org/10.56094/jss.v58i2.251","url":null,"abstract":"Human Reliability Analysis (HRA) has found application within a diverse set of engineering domains, but the methods used to apply HRA are often complicated, time-consuming, costly to apply, specific to particular (i.e., nuclear) applications, and are not suitable for direct comparison amongst themselves. \u0000This paper proposes a Human Factors Hazard Model (HFHM), which builds an HRA method from the tools of Fault Tree Analysis (FTA), Event Tree Analysis (ETA), and a novel model of considering serial Human Error Probability (HEP) more relevant to psychomotor-intensive industrial and commercial applications such as manufacturing, teleoperation, and vehicle operation. The HEP approach uses Performance Shaping Factors (PSFs) relevant to human behavior, as well as specific characteristics unique to a system architecture and its corresponding operational behavior. The HFHM tool is intended to establish a common analysis approach, to simplify and automate the modeling of the likelihood of a mishap due to a human-system interaction during a hazard event. \u0000The HFHM is executed commercial software tools (MS Excel and SysML) such that trade and sensitivity studies can be conducted and iterated automatically. The results generated by the HFHM can be used to guide risk assessment, safety requirements generation and management, design options, and safety controls within the system design architecting process. Verification and evaluation of the HFHM through simulation and subject matter expert evaluation illustrate the value of the HFHM as a tool for HRA and system safety analysis in a set of key industrial applications.","PeriodicalId":250838,"journal":{"name":"Journal of System Safety","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131593747","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Review of the Latest Developments in Automotive Safety Standardization for Driving Automation Systems 汽车自动驾驶系统安全标准化研究进展综述

Journal of System Safety Pub Date : 2023-06-22 DOI: 10.56094/jss.v58i2.252

R. Debouk

引用次数: 1

System Safety Bookshelf 系统安全书架

Journal of System Safety Pub Date : 2023-02-22 DOI: 10.56094/jss.v58i1.214

Malcolm Jones

引用次数: 0

Difficulties with Replacing Crew Launch Abort Systems with Designed Reliability 替换具有设计可靠性的机组发射中止系统的困难

Journal of System Safety Pub Date : 2023-02-22 DOI: 10.56094/jss.v58i1.216

S. Ryan

引用次数: 0

Quantification of Benefits for Medical Devices 医疗器械效益的量化

Journal of System Safety Pub Date : 2023-02-22 DOI: 10.56094/jss.v58i1.217

B. Elahi

引用次数: 0

Incremental Assurance Through Eliminative Argumentation 通过排除论证的增量保证

Journal of System Safety Pub Date : 2023-02-22 DOI: 10.56094/jss.v58i1.215

Simon Diemert, John Goodenough, Jeffrey J. Joyce, C. Weinstock

引用次数: 0

Assessing the Software Control Autonomy of System Functions in Safety-Critical Systems 安全关键系统中系统功能的软件控制自主性评估

Journal of System Safety Pub Date : 2022-10-04 DOI: 10.56094/jss.v57i3.206

V. Tran, V. Tran, L. Tran

{"title":"Assessing the Software Control Autonomy of System Functions in Safety-Critical Systems","authors":"V. Tran, V. Tran, L. Tran","doi":"10.56094/jss.v57i3.206","DOIUrl":"https://doi.org/10.56094/jss.v57i3.206","url":null,"abstract":"Software Control Category (SCC) denotes the degree of control autonomy, command and control authority, and redundant fault tolerance software has over hazardous system functions of safety-critical systems. The use of SCC for determining the software contribution to system risks is a unique feature of the MIL-STD-882E System Safety Standard. A lower SCC designation means that the software system has a greater control autonomy over hazardous system functions, whereas SCC 1 means complete autonomous control. Software with greater control autonomy over hazardous system functions require greater effort to assure reliability and safety. Correct assessment of the SCC level of hazardous system functions is crucial for optimizing the safety property of a system developed under budget, schedule, and resource constraints. Beyond the categorical definitions provided by the MIL-STD-882E Standard, there is little information on conducting an SCC assessment. To close this knowledge gap, we present an SCC assessment method. Our paper will describe in detail the process and rules for assessing SCC. For illustration, we apply our method to assess the SCC of several safety-significant functions of an automobile’s brake-assist system.","PeriodicalId":250838,"journal":{"name":"Journal of System Safety","volume":"69 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126680709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Global Warming and System Safety 全球变暖与系统安全

Journal of System Safety Pub Date : 2022-10-04 DOI: 10.56094/jss.v57i3.205

Malcolm Jones

{"title":"Global Warming and System Safety","authors":"Malcolm Jones","doi":"10.56094/jss.v57i3.205","DOIUrl":"https://doi.org/10.56094/jss.v57i3.205","url":null,"abstract":"We are currently confronted with the existential challenge of global warning. Because of its nature it is a challenge that confronts the entire globe both in terms of contributing factors and bearing the consequences. In both aspects there is an inevitable balance of responsibilities and consequences. In the former, some national entities are bigger contributors to the problem than others and in a similar manner some global areas suffer relatively more significant negative consequences. Another major challenge has been that of generating a better scientific understanding of the relationships between greenhouse gas emission, global warming, and the resulting environmental consequences. The remaining challenges that follow are how best to prevent or minimise greenhouse gas emissions, how to store them safety and how to mitigate the potential negative consequences. These are now global level responsibilities. At first sight this appears to be a problem restricted to big science, technology, and engineering alone in terms of finding more acceptable forms of energy production, as a counter to our current dependence on fossil fuels and that it might not be an area where system safety can play a prominent part. However, this is not the case, and this paper explores the system safety application possibilities, because all new developments require to be implemented in a safe manner.","PeriodicalId":250838,"journal":{"name":"Journal of System Safety","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130091092","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Delta Variant 三角洲变体

Journal of System Safety Pub Date : 2022-10-04 DOI: 10.56094/jss.v57i3.204

R. Zito

{"title":"Delta Variant","authors":"R. Zito","doi":"10.56094/jss.v57i3.204","DOIUrl":"https://doi.org/10.56094/jss.v57i3.204","url":null,"abstract":"Nothing is harder than to realize when you are living through history. For most of us, each day is pretty much like another. There is nothing historically remarkable about that. Occasionally, however, our lives are punctuated by events, both natural and man-made, that are apocalyptic and often (but not always) beyond our control – natural disasters, war, pestilence, and famine. These are the events that the historian must recognize. At this time, it is the COVID-19 pandemic that demands to be recorded by “his-story” so that posterity will know what we did right, and what we did wrong. This author has taken up the challenge of producing accurate, unbiased, comprehensive, technical annals of the global coronavirus pandemic that began in 2019. “The Delta Variant” is the third publication in this series. We are now near the end of the third year of the pandemic (summer/fall 2021). As predicted by this author, it has been a draconian year. Last year’s peak in the number of active cases was not a global maximum for the pandemic in the U.S., since this year the number of active cases has already surpassed it. Without knowing where the global maximum lies, no accurate predictions can be made about the magnitude and duration of this modern plague. The “Delta Variant” (δ-variant) of COVID-19 has greatly complicated efforts to combat the virus. The “anti-vaxxer” movement, uncontrolled migration of people into and within the U.S, and the relaxation of safety measures during the late spring and early summer in the U.S. also contributed difficulties. All of these problems were foreseen by the author and were discussed in the second paper (“Vaccine Safety”) of this series on the COVID pandemic. However, our biggest problem in the U.S. was an over confidence born of a natural summertime trough in the daily infection rate. We wanted to believe the infection was past, so we ignored the experience of India, and our administrators fueled our hopes with their words and actions. We believed because we wanted to believe – except for this author. So, what went wrong? What is a δ-variant, and why is it so dangerous? That will be the topic of this publication.","PeriodicalId":250838,"journal":{"name":"Journal of System Safety","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116975156","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 16