PSAM 16 - Session T03 Overview

PSAM 16 Conference Session T03 Overview

Session Chair: Josefin Lindström (josefin.lindstrom@risk.lth.se)

Paper 1 UM119

Lead Author: Umemoto Mizuki Co-author(s): Kadono Shunsuke : kadono-shunsuke038@g.ecc.u-tokyo.ac.jp Mitsuhashi Daichi : mitsuhashi@g.ecc.u-tokyo.ac.jp Kanno Taro : tkanno@g.ecc.u-tokyo.ac.jp Kajiyama Kazumi : kazumi@qj8.so-net.ne.jp Sharikura Sachika : sharikura-@cmed.showa-u.ac.jp Ikari Ryoko : wachiryo@gmail.com Yoneyama Masashi : yoneyama-m@cmed.showa-u.ac.jp

Modeling and simulation of in-hospital medical processes in the event of disaster for the evaluation of BCP and training
Background and purpose: Disaster-base hospitals are expected to play a central role in regional disaster medicine and required to provide medical treatment to many injured patients in the event of large-scale disasters. To be well prepared for disasters and improve the response capability, disaster response training based on the well-designed BCPs is demanded for hospital staff. However, traditional live exercise simulation is costly and has many restrictions from the ordinal hospital operations and activities. Also, there is no established method to evaluate the effectiveness and resilience of BCPs. With this background, this study aims to develop a highly realistic computer simulation model of the whole processes of disaster medicine in disaster-based hospitals that can be used for training and BCP evaluation. As the first step of this purpose, we are developing a model of the in-hospital medical processes in the event of disasters. Methods: By analyzing relevant documents including disaster training scenarios, observing an actual disaster response training, and interviewing with the medical staff who are in charge of hospital disaster management, we identified some important medical processes and response tasks, critical resources and decisions making required for each task, movement of staff and patients in the hospital, and communication among different areas in the hospital and its contents. Also, we developed a patient model that describes the detailed information about a patient such as vital signs, condition of injuries, other physical conditions, required medical treatments, and time and quantity of resources required for each treatment. Results: We developed a process-based simulation with the models developed and conducted simulations of disaster medical processes in the target hospital using an actual disaster training scenario considering patients arrivals, resource supply, and staff gathering. Each process is executed when there are patients waiting for the process and there is a necessary quantity of resources for the task required, including staff, medical materials and equipment, and a vacancy in the room or the area to perform the process. This simulation outputs the flow of patients in the hospital and resource consumptions in each area, and we could observe that the entire disaster medical processes including triage, in-hospital transportation of the patients, examination and treatment, operations, hospitalization, and so on. Also, we observed that the simulation could replicate the situation such as the stuck processes due to the lack of resources, for example, the number of patients who are waiting for examinations such as X-ray photography (XP) and computerized tomography (CT) were increasing after the disaster occurrence because the capacities of the rooms for those examinations are too small to accept many patients. In addition, we conducted simulations with modified scenarios, different patient arrivals and different resource allocation policy, to observe the scenario dependency for model verification. The results showed that medical processes were likely to be stuck in the scenario in which many severely injured patients arrive with short intervals. The results also suggest that the entire medical processes proceed efficiently without less stuck when resources are allocated to the patients with triage level of red . Discussion: By comparing the simulation results with different scenarios, it is expected to find bottlenecks as well as potential problems efficiently in the current disaster response rules and resource allocation policy. It is also expected to use this simulation for simulation-based optimization to better design BCPs as well as to find challenging scenarios for training. Furthermore, we intend to extend this simulation into human-in-the-loop simulation and use it for training and exercise. Conclusion: We developed a process-based simulation that can replicate the situation in hospitals in the event of a disaster. In the next step, we will extend the model to include activities in the command post of treatment areas, disaster response headquarters, and normal duties in the hospital, as well as communications and interactions among those different areas and sections in the hospital and develop a precise and comprehensive human-in-the-loop simulation.

Modeling and simulation of in-hospital medical processes in the event of disaster for the evaluation of BCP and training

Background and purpose: Disaster-base hospitals are expected to play a central role in regional disaster medicine and required to provide medical treatment to many injured patients in the event of large-scale disasters. To be well prepared for disasters and improve the response capability, disaster response training based on the well-designed BCPs is demanded for hospital staff. However, traditional live exercise simulation is costly and has many restrictions from the ordinal hospital operations and activities. Also, there is no established method to evaluate the effectiveness and resilience of BCPs. With this background, this study aims to develop a highly realistic computer simulation model of the whole processes of disaster medicine in disaster-based hospitals that can be used for training and BCP evaluation. As the first step of this purpose, we are developing a model of the in-hospital medical processes in the event of disasters. Methods: By analyzing relevant documents including disaster training scenarios, observing an actual disaster response training, and interviewing with the medical staff who are in charge of hospital disaster management, we identified some important medical processes and response tasks, critical resources and decisions making required for each task, movement of staff and patients in the hospital, and communication among different areas in the hospital and its contents. Also, we developed a patient model that describes the detailed information about a patient such as vital signs, condition of injuries, other physical conditions, required medical treatments, and time and quantity of resources required for each treatment. Results: We developed a process-based simulation with the models developed and conducted simulations of disaster medical processes in the target hospital using an actual disaster training scenario considering patients arrivals, resource supply, and staff gathering. Each process is executed when there are patients waiting for the process and there is a necessary quantity of resources for the task required, including staff, medical materials and equipment, and a vacancy in the room or the area to perform the process. This simulation outputs the flow of patients in the hospital and resource consumptions in each area, and we could observe that the entire disaster medical processes including triage, in-hospital transportation of the patients, examination and treatment, operations, hospitalization, and so on. Also, we observed that the simulation could replicate the situation such as the stuck processes due to the lack of resources, for example, the number of patients who are waiting for examinations such as X-ray photography (XP) and computerized tomography (CT) were increasing after the disaster occurrence because the capacities of the rooms for those examinations are too small to accept many patients. In addition, we conducted simulations with modified scenarios, different patient arrivals and different resource allocation policy, to observe the scenario dependency for model verification. The results showed that medical processes were likely to be stuck in the scenario in which many severely injured patients arrive with short intervals. The results also suggest that the entire medical processes proceed efficiently without less stuck when resources are allocated to the patients with triage level of red . Discussion: By comparing the simulation results with different scenarios, it is expected to find bottlenecks as well as potential problems efficiently in the current disaster response rules and resource allocation policy. It is also expected to use this simulation for simulation-based optimization to better design BCPs as well as to find challenging scenarios for training. Furthermore, we intend to extend this simulation into human-in-the-loop simulation and use it for training and exercise. Conclusion: We developed a process-based simulation that can replicate the situation in hospitals in the event of a disaster. In the next step, we will extend the model to include activities in the command post of treatment areas, disaster response headquarters, and normal duties in the hospital, as well as communications and interactions among those different areas and sections in the hospital and develop a precise and comprehensive human-in-the-loop simulation.

Paper UM119 | Download the paper file. | Download the presentation PowerPoint file.

Name: Umemoto Mizuki (umemoto-mizuki9492017@g.ecc.u-tokyo.ac.jp)

Bio: Mizuki Umemoto is the 2nd year of master's degree student in the department of engeneering at the University of Tokyo. His current interests are training and testing preparedness for major accidents and disasters in disaster base hospitals. His graduation thesis addresses a methodology to formulate a business continuity plan from the perspective of resource management(e.g. human, products, money, information, trust)

Country: JPN
Company: The University Of Tokyo
Job Title: Master 2nd

Paper 2 JI5

Lead Author: Yail Kim

Safety of GFRP-Reinforced Concrete Columns Subjected to Sustained Intensity
This research presents the safety of concrete columns reinforced with glass fiber reinforced polymer (GFRP) composite bars under sustained loading over a service period of 100 years. An analytical model is formulated based on force equilibrium and strain compatibility to predict the behavior of the columns and to assess their performance reliability. Upon validation of the model against published experimental data, an extensive parametric study is conducted to elucidate the safety of the columns subjected to the long-term load. The implications of the constituents are examined with a focus on creep, shrinkage, and stress redistributions. The vulnerability of premature failure is discussed for practitioners to consider when implementing GFRP-reinforced concrete columns in the field.

Safety of GFRP-Reinforced Concrete Columns Subjected to Sustained Intensity

This research presents the safety of concrete columns reinforced with glass fiber reinforced polymer (GFRP) composite bars under sustained loading over a service period of 100 years. An analytical model is formulated based on force equilibrium and strain compatibility to predict the behavior of the columns and to assess their performance reliability. Upon validation of the model against published experimental data, an extensive parametric study is conducted to elucidate the safety of the columns subjected to the long-term load. The implications of the constituents are examined with a focus on creep, shrinkage, and stress redistributions. The vulnerability of premature failure is discussed for practitioners to consider when implementing GFRP-reinforced concrete columns in the field.

Paper JI5 | Download the paper file. |

Name: Yail Kim (jimmy.kim@ucdenver.edu)

Bio: Dr. Jimmy Kim is President of the Bridge Engineering Institute, An International Technical Society, and a Professor in the Department of Civil Engineering at the University of Colorado Denver, Colorado, USA. His research interests encompass advanced composite materials for rehabilitation, structural informatics, complex systems, and science-based structural engineering, including statistical, interfacial, and quantum physics.

Country: USA
Company: University of Colorado Denver
Job Title: Professor

Paper 3 KI18

Lead Author: Kim Hintz Co-author(s): Dr.-Ing. Martin Dazer, martin.dazer@ima.uni-stuttgart.de Prof. Dr.-Ing. Bernd Bertsche, bernd.bertsche@ima.uni-stuttgart.de

Availability Analysis of Photovoltaic System Concepts to Derive Reliability Requirements for Inverters within Different Application Scenarios
The energy concept based on fossil fuels is unsustainable in long term due to the ongoing shortage of resources and the associated negative impact on climate change. The remedy is offered by the sustainable, nearly inexhaustible, and almost emission-free sources of renewable energy. Photovoltaics (PV) play a key role, converting the world's solar radiation into electrical energy without emissions and thus making solar energy usable in a decentralized manner. To promote the expansion of PV systems in urban environments, manufacturers must guarantee ecological, safe, but above all economical operation for the customer. Various concepts are used for grid-connected PV systems, which differ in their electronic design and circuit layout. These include concepts based on string or module inverters as well as concepts using power optimizers. The concepts offer several advantages and disadvantages, such as the control of individual PV modules at the optimal operating point, their monitoring, as well as the flexibility in the installation. Furthermore, the profitability of the PV system depends on the specific application scenario. Here, various factors such as the system size, the orientation and angle of irradiation, partial shading as well as individual defects influence the performance of the PV system. These complex interactions make it difficult to compare the different PV system concepts. From the customer's point of view, it is difficult to select an optimal system concept due to his individual application scenario. For him, only the total costs, i.e. the sum of investment and repair costs as well as the energy yield or the feed-in compensation, are decisive for the selection of a suitable concept. The repair costs and the yield of the PV system are directly related to its availability and therefore also the reliability since a defect system cannot feed any electricity into the grid. To achieve reliability and profitability of the PV system, they must be designed for a service life of more than 20 years. In this context, the reliability requirements of the individual components vary greatly due to the different interactions within the PV system concepts and the application scenarios. Depending on the layout of the concepts, the failure of an individual component has a different effect on the overall PV system. The failure either leads to the loss of the power of individual PV modules or to the total failure of the whole PV system. This paper deals with the reliability and availability analysis of different PV system concepts in order to make predictions about their economic efficiency. On this basis, reliability requirements for individual components of the PV system can be derived, with which the economic operation of a PV system can be achieved under the consideration of different operating conditions. The evaluation is carried out using Petri nets, with which a realistic analysis of the reliability and availability of the various PV system concepts can be performed, considering the different repair times and maintenance models. In this way, the availability of the PV system and thus its delivered yield, as well as the installation and repair costs for different setups can be modeled and evaluated within a Monte Carlo simulation. Within the framework of a parameter study, the different failure behavior of individual components as well as the effects of different shading situations on the output of the PV systems are modeled. By specifying a certain application scenario and a desired break-even point, the reliability requirements for individual components of the PV systems can be derived. In this way, it is possible to determine which reliability requirements are necessary for the individual PV system concepts to ensure economic operation for the customer. Important influencing variables and realistic application scenarios such as system size and partial shading situations must be taken into account in this analysis.

Availability Analysis of Photovoltaic System Concepts to Derive Reliability Requirements for Inverters within Different Application Scenarios

The energy concept based on fossil fuels is unsustainable in long term due to the ongoing shortage of resources and the associated negative impact on climate change. The remedy is offered by the sustainable, nearly inexhaustible, and almost emission-free sources of renewable energy. Photovoltaics (PV) play a key role, converting the world's solar radiation into electrical energy without emissions and thus making solar energy usable in a decentralized manner. To promote the expansion of PV systems in urban environments, manufacturers must guarantee ecological, safe, but above all economical operation for the customer. Various concepts are used for grid-connected PV systems, which differ in their electronic design and circuit layout. These include concepts based on string or module inverters as well as concepts using power optimizers. The concepts offer several advantages and disadvantages, such as the control of individual PV modules at the optimal operating point, their monitoring, as well as the flexibility in the installation. Furthermore, the profitability of the PV system depends on the specific application scenario. Here, various factors such as the system size, the orientation and angle of irradiation, partial shading as well as individual defects influence the performance of the PV system. These complex interactions make it difficult to compare the different PV system concepts. From the customer's point of view, it is difficult to select an optimal system concept due to his individual application scenario. For him, only the total costs, i.e. the sum of investment and repair costs as well as the energy yield or the feed-in compensation, are decisive for the selection of a suitable concept. The repair costs and the yield of the PV system are directly related to its availability and therefore also the reliability since a defect system cannot feed any electricity into the grid. To achieve reliability and profitability of the PV system, they must be designed for a service life of more than 20 years. In this context, the reliability requirements of the individual components vary greatly due to the different interactions within the PV system concepts and the application scenarios. Depending on the layout of the concepts, the failure of an individual component has a different effect on the overall PV system. The failure either leads to the loss of the power of individual PV modules or to the total failure of the whole PV system. This paper deals with the reliability and availability analysis of different PV system concepts in order to make predictions about their economic efficiency. On this basis, reliability requirements for individual components of the PV system can be derived, with which the economic operation of a PV system can be achieved under the consideration of different operating conditions. The evaluation is carried out using Petri nets, with which a realistic analysis of the reliability and availability of the various PV system concepts can be performed, considering the different repair times and maintenance models. In this way, the availability of the PV system and thus its delivered yield, as well as the installation and repair costs for different setups can be modeled and evaluated within a Monte Carlo simulation. Within the framework of a parameter study, the different failure behavior of individual components as well as the effects of different shading situations on the output of the PV systems are modeled. By specifying a certain application scenario and a desired break-even point, the reliability requirements for individual components of the PV systems can be derived. In this way, it is possible to determine which reliability requirements are necessary for the individual PV system concepts to ensure economic operation for the customer. Important influencing variables and realistic application scenarios such as system size and partial shading situations must be taken into account in this analysis.

Paper KI18 | Download the paper file. | Download the presentation pdf file.

Name: Kim Hintz (kim.hintz@ima.uni-stuttgart.de)

Bio: Kim Hintz studied Mechanical Engineering at the University of Stuttgart in Germany and received his academic degree Master of Science in 2019. He is working as a research assistant in the field of reliability engineering at the Institute of Machine Components. He is pursuing his Ph.D. studies with a focus on reliability assessment of electronic systems.

Country: DEU
Company: Institute of Machine Components
Job Title: Research Assistant

Paper 4 MB290

Lead Author: Somayeh Mohammadi Co-author(s): Michelle Bensi (mbensi@umd.edu) ZhegangMa (zhegang.ma@inl.gov), and Kaveh Faraji Najarkolaie (kfaraji@terpmail.umd.edu)

Uncertainty in Predicted Tropical Cyclone Path and Landfall Characteristics for Landfalling Storms to Support External Hazard Probabilistic Risk Assessments for Critical Infrastructure – A Preliminary Analysis
Critical infrastructure facilities, such as nuclear power plants, are often located in coastal regions exposed to tropical cyclones (TCs). These facilities may employ permanent protective measures as well as strategies that require manual (human) actions to install temporary features (e.g., flood protection berms and pumps). In addition to the possibility of hardware failures, there is a possibility that actions will be unsuccessful due to delayed organizational decision-making, human errors, and differences between predicted and experienced hazard characteristics. Accurate external hazard probabilistic risk assessments (XHPRAs) must quantify these error probabilities, which depend on factors such as the information available to support decisions, the time available to perform actions, and the environmental conditions under which actions are performed. These factors are subject to uncertainty due to uncertainty in TC forecasts. To support XHPRAs for critical infrastructure facilities, this paper seeks to explore uncertainty in the conditions under which human actions will be performed, with particular emphasis on the time available to execute actions. We analyzed National Oceanic and Atmospheric Administration (NOAA) geographic information system (GIS) datasets related to advisory forecast TC track data for 2012-2020. For each historic storm, we compared advisory forecasted track data (e.g., predicted landfall locations, times, and wind speed) to the observed track to understand errors and uncertainty.

Uncertainty in Predicted Tropical Cyclone Path and Landfall Characteristics for Landfalling Storms to Support External Hazard Probabilistic Risk Assessments for Critical Infrastructure – A Preliminary Analysis

Critical infrastructure facilities, such as nuclear power plants, are often located in coastal regions exposed to tropical cyclones (TCs). These facilities may employ permanent protective measures as well as strategies that require manual (human) actions to install temporary features (e.g., flood protection berms and pumps). In addition to the possibility of hardware failures, there is a possibility that actions will be unsuccessful due to delayed organizational decision-making, human errors, and differences between predicted and experienced hazard characteristics. Accurate external hazard probabilistic risk assessments (XHPRAs) must quantify these error probabilities, which depend on factors such as the information available to support decisions, the time available to perform actions, and the environmental conditions under which actions are performed. These factors are subject to uncertainty due to uncertainty in TC forecasts. To support XHPRAs for critical infrastructure facilities, this paper seeks to explore uncertainty in the conditions under which human actions will be performed, with particular emphasis on the time available to execute actions. We analyzed National Oceanic and Atmospheric Administration (NOAA) geographic information system (GIS) datasets related to advisory forecast TC track data for 2012-2020. For each historic storm, we compared advisory forecasted track data (e.g., predicted landfall locations, times, and wind speed) to the observed track to understand errors and uncertainty.

Paper MB290 | Download the paper file. | Download the presentation pdf file. Download the presentation PowerPoint file.

A PSAM Profile is not yet available for this author.