PSAM 16 - Session M24 Overview

PSAM 16 Conference Session M24 Overview

Session Chair: Sai Zhang (sai.zhang@inl.gov)

Paper 1 WA49

Lead Author: Wasin Vechgama Co-author(s): Mr. Watcha Sasawattakul, Department of Computer Engineering, Faculty of Engineering, Chulalongkorn University, 17th floor, Engineering 4 Building (Charoenvidsavakham), Phayathai Road, Wang Mai, Pathumwan, Bangkok, 10330, Thailand, Email: watcha.sasawattakul@gmail.com.

Development of Classification Model for Public Perception of Nuclear Energy in Social Media Platform using Machine Learning: Facebook Platform in Thailand
Due to the nuclear consequences of the severe accidents of Nuclear Power Plants (NPPs) of Fukushima Daiichi in 2011, the public acceptance of nuclear energy has been decreased significantly in many countries including Thailand. Since 2011, the Thailand Government had continuously postponed the NPP project until in 2018 the NPP project was not contained in the latest Power Development Plan. Apart from the concerns of the safety of NPPs, public apprehension was the important reason affecting the nuclear energy plan of Thailand. In the past, public acceptance surveys have been conducted by using questionnaires to reflect the people's opinion about nuclear energy in Thailand, especially after the Fukushima disaster. However, the surveys using the questionnaire had the limitation of people access, and the high cost and time consumption. Since, nowadays, the key role of computational code and social media is influential to people around the world significantly including Thailand, data collections from the direct and indirect surveys in various fields have been evaluated through social media platforms. The objective of this study was to develop classification models for public perception of nuclear energy of Thai people in social media platforms using machine learning focusing on the Facebook platform. The comment data from the Facebook Pages having nuclear energy news and information in their posts were extracted by web scraping. Then the extracted data were classified and prepared for proper machine learning model inputs consisting of train data and test data. Train data were used to generate machine learning models to classify the public perception of nuclear energy of Thai people to understand their positive and negative perceptions. The results of generating machine learning models were validated by test data to suggest appropriate models for the public perception of nuclear energy of Thai people. Facebook was selected in the study because now there are Thailand Facebook users with more than 50 million accounts that account for around 70 percent of the total people in Thailand. Besides, machine learning was applied to the study since it allows to generate the specific models to classify the expected data from complex sentences including spoken and written languages in social media platforms. The developed classification model of this study is expected to widely understand the public perception of nuclear energy of Thai people in social platforms in order to provide approaches and activities to increase the public acceptance of nuclear energy in Thailand in the future.

Development of Classification Model for Public Perception of Nuclear Energy in Social Media Platform using Machine Learning: Facebook Platform in Thailand

Due to the nuclear consequences of the severe accidents of Nuclear Power Plants (NPPs) of Fukushima Daiichi in 2011, the public acceptance of nuclear energy has been decreased significantly in many countries including Thailand. Since 2011, the Thailand Government had continuously postponed the NPP project until in 2018 the NPP project was not contained in the latest Power Development Plan. Apart from the concerns of the safety of NPPs, public apprehension was the important reason affecting the nuclear energy plan of Thailand. In the past, public acceptance surveys have been conducted by using questionnaires to reflect the people's opinion about nuclear energy in Thailand, especially after the Fukushima disaster. However, the surveys using the questionnaire had the limitation of people access, and the high cost and time consumption. Since, nowadays, the key role of computational code and social media is influential to people around the world significantly including Thailand, data collections from the direct and indirect surveys in various fields have been evaluated through social media platforms. The objective of this study was to develop classification models for public perception of nuclear energy of Thai people in social media platforms using machine learning focusing on the Facebook platform. The comment data from the Facebook Pages having nuclear energy news and information in their posts were extracted by web scraping. Then the extracted data were classified and prepared for proper machine learning model inputs consisting of train data and test data. Train data were used to generate machine learning models to classify the public perception of nuclear energy of Thai people to understand their positive and negative perceptions. The results of generating machine learning models were validated by test data to suggest appropriate models for the public perception of nuclear energy of Thai people. Facebook was selected in the study because now there are Thailand Facebook users with more than 50 million accounts that account for around 70 percent of the total people in Thailand. Besides, machine learning was applied to the study since it allows to generate the specific models to classify the expected data from complex sentences including spoken and written languages in social media platforms. The developed classification model of this study is expected to widely understand the public perception of nuclear energy of Thai people in social platforms in order to provide approaches and activities to increase the public acceptance of nuclear energy in Thailand in the future.

Paper WA49 | Download the paper file. | Download the presentation pdf file.

Name: Wasin Vechgama (wasinvechgama@gmail.com)

Bio: I graduated Master's Degree in Nuclear Engineering from Chulalongkorn University, Thailand, in 2017. I have worked as a nuclear engineer in the Nuclear Technology Research and Development Center, Thailand Institute of Nuclear Technology for 5 years. Currently, I am a Ph.D. student in Risk Assessment at the Risk Assessment and Management Team, Korea Atomic Energy Research Institute (KAERI School), University of Science and Technology, South Korea.

Country: KOR
Company: 1) Korea Atomic Energy Research Institute (KAERI School) - University of Science and Technology (South Korea), 2) Thailand Institute of Nuclear Techno
Job Title: Ph.D. Student

Paper 2 GA187

Lead Author: Garill Coles Co-author(s): Steve M. Short, steve.short@pnnl.gov Steve J. Maheras, steven.maheras@pnnl.gov Harold E. Adkins, harold.adkins@pnnl.gov

Risk-Informed Approach for Regulatory Approval of Microreactor Transport
Pacific Northwest National Laboratory (PNNL) was tasked to develop and evaluate regulatory options for transport of microreactors. The work was funded by the National Reactor Innovation Center a National Department of Energy program within the Office of Nuclear Energy research and development which supports demonstration of microreactor technology. PNNL developed a risk-informed regulatory framework for the licensing of the transportation of microreactors, focused on the most challenging case which is the transportation of irradiated nuclear fuel that is assumed to be an integral component of the microreactor transportation package. The framework lays out a viable regulatory pathway, including decision points for regulatory options and the supporting technical evaluations for those options in phases from near to long term. Microreactors are very small nuclear reactors of 20 megawatts electric (MWe) or less, designed to be factory-built, and may be transportable. The microreactor designs being considered in this evaluation are tristructural isotropic (TRISO) fueled using high-assay low-enriched uranium with enrichments of 5 percent to 20 percent. A microreactor and its unirradiated or irradiated fuel contents will likely not be able to meet all the regulatory requirements as a Type B or fissile material transportation package under 10 CFR Part 71. Therefore, the objective of developing a risk-informed regulatory framework for the transportation of a microreactor and its fuel content is to provide viable licensing options that are safe and feasible. Risk assessment such as probabilistic risk assessment (PRA) can be used to show comparable safety to that provided by a Type B or fissile material package for surface transport. The framework includes guidance on applicable regulations and consideration of historical precedence in using risk information for transportation licensing as well as previous transportation risk assessments. The framework includes guidance for development of safety goals and for performing an offsite transportation PRA for shipping a TRISO fueled microreactor by truck as well as consideration of defense in depth and safety margin concepts. Key advantages of using the approach are (1) increasing the likelihood of successfully obtaining regulatory transportation package approval, (2) informing the design about the relative significance of microreactor containment and shielding, and (3) informing the need for transportation compensatory measures (or confirming there is no need). This paper presents the framework but focuses primarily on describing the development of a transportation PRA for microreactor packages and the risk acceptance guidelines that will be needed to access the results of the PRA for regulatory decision-making.

Risk-Informed Approach for Regulatory Approval of Microreactor Transport

Pacific Northwest National Laboratory (PNNL) was tasked to develop and evaluate regulatory options for transport of microreactors. The work was funded by the National Reactor Innovation Center a National Department of Energy program within the Office of Nuclear Energy research and development which supports demonstration of microreactor technology. PNNL developed a risk-informed regulatory framework for the licensing of the transportation of microreactors, focused on the most challenging case which is the transportation of irradiated nuclear fuel that is assumed to be an integral component of the microreactor transportation package. The framework lays out a viable regulatory pathway, including decision points for regulatory options and the supporting technical evaluations for those options in phases from near to long term. Microreactors are very small nuclear reactors of 20 megawatts electric (MWe) or less, designed to be factory-built, and may be transportable. The microreactor designs being considered in this evaluation are tristructural isotropic (TRISO) fueled using high-assay low-enriched uranium with enrichments of 5 percent to 20 percent. A microreactor and its unirradiated or irradiated fuel contents will likely not be able to meet all the regulatory requirements as a Type B or fissile material transportation package under 10 CFR Part 71. Therefore, the objective of developing a risk-informed regulatory framework for the transportation of a microreactor and its fuel content is to provide viable licensing options that are safe and feasible. Risk assessment such as probabilistic risk assessment (PRA) can be used to show comparable safety to that provided by a Type B or fissile material package for surface transport. The framework includes guidance on applicable regulations and consideration of historical precedence in using risk information for transportation licensing as well as previous transportation risk assessments. The framework includes guidance for development of safety goals and for performing an offsite transportation PRA for shipping a TRISO fueled microreactor by truck as well as consideration of defense in depth and safety margin concepts. Key advantages of using the approach are (1) increasing the likelihood of successfully obtaining regulatory transportation package approval, (2) informing the design about the relative significance of microreactor containment and shielding, and (3) informing the need for transportation compensatory measures (or confirming there is no need). This paper presents the framework but focuses primarily on describing the development of a transportation PRA for microreactor packages and the risk acceptance guidelines that will be needed to access the results of the PRA for regulatory decision-making.

Paper GA187 | Download the paper file. | Download the presentation pdf file. Download the presentation PowerPoint file.

Name: Garill Coles (garill.coles@pnnl.gov)

Bio: Mr. Coles has made a career in safety and risk evaluations for domestic and international nuclear reactors, nonreactor nuclear facilities, and critical infrastructure. He has led research in nuclear proliferation risk, facility safeguard-ability, and cyber security was an author of NUREG/CR-6847 for nuclear power plants. He is NRC’s technical reviewer for risk-informed Licensing Amendment Requests based on PRA, SAMA analysis, and Seismic and High Wind PRAs. He performed research for NRC of the impact of environmental factors on operator actions and currently supports NRC in their deployment of a new HRA methodology, IDHEAS. He is the lead author of NUREG/CR-7265 addressing low-power shutdown risk. Garill was a primary expert to a risk-informed transportation safety evaluation for air borne shipment of radiological material and a primary developer of a framework for risk informed licensing of the transportation of microreactors for the DOE.

Country: USA
Company: Pacific Northwest National Laboratory
Job Title: Nuclear Engineer - Risk Assessment

Paper 3 YI289

Lead Author: Ben Chen Co-author(s): Bruce Hamilton, bhamilton@anl.gov Dave Grabaskas, dgrabaskas@anl.gov Mark Cunningham, mark.cunningham@anl.gov Sinem Perk, sperk@anl.gov

Success Path Analysis as a Recommended Practice for enhanced Quality in High Reliability Organizations
Since 2012, Argonne National Laboratory has worked with the U.S. Bureau of Safety and Environmental Enforcement (BSEE) to develop and implement tools that support risk-informed decision making for the oil and gas industry. The Success Path Analysis Method that was developed helped visualize risk in an easy-to-understand way, provided a common language and systematic process for understanding and managing high-risk activities and equipment, enabled operational risk to be quantified, and proved to be an effective tool to facilitate communication and prioritize discussion topics among operators and BSEE with a focus on improving safety. A Success Path begins with a diagram of the hardware, software, and human actions needed to ensure safe operation of a system or component. Success Paths provide a "chain of causality" illustrating what (hardware, software, and human actions) must go right to ensure safe operations. Visualizing what must go right helps us understand, manage, and respond to what can fail. The benefits of the application of the Success Path Method in the work with BSEE has shown promise that this methodology may have broad applicability in improving safety and risk communication. Usages of which includes FMECA (Failure mode, effects, and criticality analysis), and HRO (high reliability organization) initiatives in high-risk industries (e.g., Oil & Gas, Health Care, Defense, Transportation). This project aims to generalize the Success Path Analysis methodology that was used in the BSEE work, and demonstrate how the approach satisfies the requirements associated with ISO9001 (Quality Management Systems). In addition, the work identifies the benefits and challenges of the Success Path Analysis methodology, and reviews some examples of the methodology in action.

Success Path Analysis as a Recommended Practice for enhanced Quality in High Reliability Organizations

Since 2012, Argonne National Laboratory has worked with the U.S. Bureau of Safety and Environmental Enforcement (BSEE) to develop and implement tools that support risk-informed decision making for the oil and gas industry. The Success Path Analysis Method that was developed helped visualize risk in an easy-to-understand way, provided a common language and systematic process for understanding and managing high-risk activities and equipment, enabled operational risk to be quantified, and proved to be an effective tool to facilitate communication and prioritize discussion topics among operators and BSEE with a focus on improving safety. A Success Path begins with a diagram of the hardware, software, and human actions needed to ensure safe operation of a system or component. Success Paths provide a "chain of causality" illustrating what (hardware, software, and human actions) must go right to ensure safe operations. Visualizing what must go right helps us understand, manage, and respond to what can fail. The benefits of the application of the Success Path Method in the work with BSEE has shown promise that this methodology may have broad applicability in improving safety and risk communication. Usages of which includes FMECA (Failure mode, effects, and criticality analysis), and HRO (high reliability organization) initiatives in high-risk industries (e.g., Oil & Gas, Health Care, Defense, Transportation). This project aims to generalize the Success Path Analysis methodology that was used in the BSEE work, and demonstrate how the approach satisfies the requirements associated with ISO9001 (Quality Management Systems). In addition, the work identifies the benefits and challenges of the Success Path Analysis methodology, and reviews some examples of the methodology in action.

Paper YI289 | | Download the presentation pdf file.

Name: Ben Chen (yichao.chen@anl.gov)

Bio: Ben has experience working with multiple LWR fleets developing risk-informed applications across the US, working with PRA models of various plant sites, and more recently has been involved in a number of projects regarding the developing field of advanced reactor licensing and regulations under the Safety and Risk Assessment Group at Argonne National Laboratory. Ben is a member of the ANS/ASME Joint Committee on Nuclear Risk Management (JCNRM), and holds a Bachleor's Degree in Nuclear Engineering from the University of Wisconsin-Madison.

Country: USA
Company: Argonne National Lab
Job Title: Nuclear Engineer

Paper 4 JE282

Lead Author: Jeeyea Ahn Co-author(s): Wooseok Jo, cws5528@unist.ac.kr; Byung Joo Min, bjmin135@unist.ac.kr; Seung Jun Lee, sjlee420@unist.ac.kr

A methodology for measuring the difficulty of nuclear safety culture and safety management factors
Since a strong safety culture is required for safety, the evaluation of safety culture and safety management must also meet the requirements of safety evaluation. However, at present, there is no suitable method to evaluate safety culture or safety management index in this systematic way. Most existing safety culture evaluation methods mainly focus on evaluating the maturity level of the target organization's safety culture, and do not take into account the concept of a graded approach and generally do not deviate from the framework of discovery and removal of vulnerable elements. Moreover, there is currently no unified safety culture model, so there is a difficulty in smooth communication due to differences in understanding of the same factors between organizations during safety culture-related exchanges. In this regard, this research has a purpose to develop an in-depth analysis tool for safety culture for detailed analysis by safety culture attribute and to propose a methodology to promote mutual understanding of safety culture by institution. This method can be utilized in the distribution of appropriate resources when establishing a response strategy for the improvement of the organization's safety culture weaknesses, which can be obtained from the results of other safety culture evaluations in the future. The weight of each factor contributing to the difficulty of safety culture reflects the respondents' bias in perception of the difficulty of safety culture, meaningful results can be derived by comparing them. For example, it can be used to derive the difference in the perception of safety culture that exists between the regulatory body and the operating organization, or to derive the difference in the perception of safety culture according to the position within the organization. The concept of difficulty and quantification method of safety culture characteristic elements proposed in this study were used in F-D matrix analysis for in-depth analysis of safety culture. In this paper, the concept of difficulty in safety culture characteristic elements will be defined and a quantification method will be introduced.

A methodology for measuring the difficulty of nuclear safety culture and safety management factors

Since a strong safety culture is required for safety, the evaluation of safety culture and safety management must also meet the requirements of safety evaluation. However, at present, there is no suitable method to evaluate safety culture or safety management index in this systematic way. Most existing safety culture evaluation methods mainly focus on evaluating the maturity level of the target organization's safety culture, and do not take into account the concept of a graded approach and generally do not deviate from the framework of discovery and removal of vulnerable elements. Moreover, there is currently no unified safety culture model, so there is a difficulty in smooth communication due to differences in understanding of the same factors between organizations during safety culture-related exchanges. In this regard, this research has a purpose to develop an in-depth analysis tool for safety culture for detailed analysis by safety culture attribute and to propose a methodology to promote mutual understanding of safety culture by institution. This method can be utilized in the distribution of appropriate resources when establishing a response strategy for the improvement of the organization's safety culture weaknesses, which can be obtained from the results of other safety culture evaluations in the future. The weight of each factor contributing to the difficulty of safety culture reflects the respondents' bias in perception of the difficulty of safety culture, meaningful results can be derived by comparing them. For example, it can be used to derive the difference in the perception of safety culture that exists between the regulatory body and the operating organization, or to derive the difference in the perception of safety culture according to the position within the organization. The concept of difficulty and quantification method of safety culture characteristic elements proposed in this study were used in F-D matrix analysis for in-depth analysis of safety culture. In this paper, the concept of difficulty in safety culture characteristic elements will be defined and a quantification method will be introduced.

Paper JE282 | Download the paper file. |

Name: Jeeyea Ahn (jeeya@unist.ac.kr)

Bio: I am a graduate student of the nuclear safety and HMI evolution lab (NuSAPHE: pronounced as nu-safe) at UNIST. I am interested in nuclear safety and human/organizational factors, and I am also interested in research in various fields to prevent accidents caused by human error in nuclear power plants. I am currently studying nuclear safety culture, which has been considered one of the important organizational factors.

Country: KOR
Company: Ulsan National Institute of Science and Technology
Job Title: Combined m.s.-ph.d. student (ph.d candidate)