Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735606
G. Lorenz, Jacob S. Ehrenstrom, Tyler B. Ullmann, Ryan C. Palmer, N. Tenhundfeld, E. D. de Visser, Bianca Donadio, Chad C. Tossell
The present study aims to enhance the design of future cockpits by supporting pilots in their monitoring and control of autonomous wingmen. In our scenario, autonomous wingmen are F-16 aircraft that can fly autonomously; a real capability currently under development by industry partners. These autonomous fighter aircraft exist to enhance mission effectiveness by reducing the risk of harm to humans, increasing sensor coverage, providing access to more environments, and decreasing cost. However, the method and systems used for interaction between pilots and the autonomous systems are still under development and an active area of research. In collaboration with industry partners, we are assessing different control input devices for supervisory control of autonomous fighter aircraft in a flight simulator. In this simulator, 60 participants will fly various mission types with autonomous wingmen using one of three different types of controllers: Microsoft Sidewinder, F-35 Hands on Throttle and Stick (HOTAS), and a wrist-mounted computer gaming keypad. Pilots will interact with their autonomous wingmen for brief periods of time through calling “plays” that initiate the unmanned aircraft to conduct a task in combat scenarios. Physiological data will be collected via electroencephalogram (EEG), electrocardiogram (ECG), and galvanic skin response (GSR). Eye movements and behavioral measures of reliance on the autonomous wingmen and usage of each of the aforementioned controllers will also be collected. These data, in addition to survey responses, will inform design recommendations for optimal interaction with autonomous wingmen.
{"title":"Assessing Control Devices for the Supervisory Control of Autonomous Wingmen","authors":"G. Lorenz, Jacob S. Ehrenstrom, Tyler B. Ullmann, Ryan C. Palmer, N. Tenhundfeld, E. D. de Visser, Bianca Donadio, Chad C. Tossell","doi":"10.1109/SIEDS.2019.8735606","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735606","url":null,"abstract":"The present study aims to enhance the design of future cockpits by supporting pilots in their monitoring and control of autonomous wingmen. In our scenario, autonomous wingmen are F-16 aircraft that can fly autonomously; a real capability currently under development by industry partners. These autonomous fighter aircraft exist to enhance mission effectiveness by reducing the risk of harm to humans, increasing sensor coverage, providing access to more environments, and decreasing cost. However, the method and systems used for interaction between pilots and the autonomous systems are still under development and an active area of research. In collaboration with industry partners, we are assessing different control input devices for supervisory control of autonomous fighter aircraft in a flight simulator. In this simulator, 60 participants will fly various mission types with autonomous wingmen using one of three different types of controllers: Microsoft Sidewinder, F-35 Hands on Throttle and Stick (HOTAS), and a wrist-mounted computer gaming keypad. Pilots will interact with their autonomous wingmen for brief periods of time through calling “plays” that initiate the unmanned aircraft to conduct a task in combat scenarios. Physiological data will be collected via electroencephalogram (EEG), electrocardiogram (ECG), and galvanic skin response (GSR). Eye movements and behavioral measures of reliance on the autonomous wingmen and usage of each of the aforementioned controllers will also be collected. These data, in addition to survey responses, will inform design recommendations for optimal interaction with autonomous wingmen.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130125646","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735594
Peter Kromkowski, Shaoran Li, Wenxi Zhao, Brendan Abraham, Austin Osborne, Donald E. Brown
Large organizations may have hundreds or thousands of applications running simultaneously to support their operations. To maintain high levels of efficiency, they need to quickly detect outages or anomalies in order to quickly fix the problem and reduce costs. This paper describes the analytical framework for a network traffic data anomaly-detection method to reduce application downtime and the need for human involvement in detecting or reporting anomalous application behavior. We use the described framework to compare the performances of a Seasonal Autoregressive Integrated Moving Average (SARIMA) times series model and Long Short-Term Memory (LSTM) Autoencoder model at anomaly detection. We evaluated these models using false positive rates and accuracy, with a requirement of being able to give timely alerts, and saw that even though both models were accurate, their false positive rates were very high. We then improved overall detection performance by ensembling the SARIMA and LSTM autoencoder. Our results demonstrate a possible new method of anomaly detection in network traffic flow using time series and autoencoders.
{"title":"Evaluating Statistical Models for Network Traffic Anomaly Detection","authors":"Peter Kromkowski, Shaoran Li, Wenxi Zhao, Brendan Abraham, Austin Osborne, Donald E. Brown","doi":"10.1109/SIEDS.2019.8735594","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735594","url":null,"abstract":"Large organizations may have hundreds or thousands of applications running simultaneously to support their operations. To maintain high levels of efficiency, they need to quickly detect outages or anomalies in order to quickly fix the problem and reduce costs. This paper describes the analytical framework for a network traffic data anomaly-detection method to reduce application downtime and the need for human involvement in detecting or reporting anomalous application behavior. We use the described framework to compare the performances of a Seasonal Autoregressive Integrated Moving Average (SARIMA) times series model and Long Short-Term Memory (LSTM) Autoencoder model at anomaly detection. We evaluated these models using false positive rates and accuracy, with a requirement of being able to give timely alerts, and saw that even though both models were accurate, their false positive rates were very high. We then improved overall detection performance by ensembling the SARIMA and LSTM autoencoder. Our results demonstrate a possible new method of anomaly detection in network traffic flow using time series and autoencoders.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114797337","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735620
Madhur Behl, Jackson DuBro, Taylor Flynt, Imaan Hameed, G. Lang, Felix Park
Electric vehicle (EV) adoption has surpassed the growth of charging infrastructure. As the demand for charging stations surpass the supply, expanding charging infrastructure for consumers is crucial to improving the experience of owning and maintaining an EV. One solution is to simply provide more charging stations; however, this requires significant upfront hardware and space cost. In addition, parking spots allocated for EVs should only be used by EVs, forcing manufacturers to make a decision on the number of EV and non-EV parking spots. Current charging stations also have their own problems. When an EV is finished charging, any additional time it spends in the charging location is time that another EV could be using to charge itself. Innovative new products are necessary to create an adequate charging network. In this work, a mobile autonomous robot which charges parked EVs at any location with its own battery is presented. We created a proof-of-concept autonomous charging robot to demonstrate feasibility and motivate future work. The goal is to provide three main decoupled functionalities: parking lot navigation, EV plug guidance, and robot battery swapping. The current iteration meets these functionalities using a TurtleBot to navigate a mock parking lot, new designs and prototypes for swapping batteries, and a robotic arm paired with a computer vision algorithm to guide a 3D printed plug. Ongoing challenges for future iterations involve integrating the main functionalities and dealing with a wider range of less common use cases.
{"title":"Autonomous Electric Vehicle Charging System","authors":"Madhur Behl, Jackson DuBro, Taylor Flynt, Imaan Hameed, G. Lang, Felix Park","doi":"10.1109/SIEDS.2019.8735620","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735620","url":null,"abstract":"Electric vehicle (EV) adoption has surpassed the growth of charging infrastructure. As the demand for charging stations surpass the supply, expanding charging infrastructure for consumers is crucial to improving the experience of owning and maintaining an EV. One solution is to simply provide more charging stations; however, this requires significant upfront hardware and space cost. In addition, parking spots allocated for EVs should only be used by EVs, forcing manufacturers to make a decision on the number of EV and non-EV parking spots. Current charging stations also have their own problems. When an EV is finished charging, any additional time it spends in the charging location is time that another EV could be using to charge itself. Innovative new products are necessary to create an adequate charging network. In this work, a mobile autonomous robot which charges parked EVs at any location with its own battery is presented. We created a proof-of-concept autonomous charging robot to demonstrate feasibility and motivate future work. The goal is to provide three main decoupled functionalities: parking lot navigation, EV plug guidance, and robot battery swapping. The current iteration meets these functionalities using a TurtleBot to navigate a mock parking lot, new designs and prototypes for swapping batteries, and a robotic arm paired with a computer vision algorithm to guide a 3D printed plug. Ongoing challenges for future iterations involve integrating the main functionalities and dealing with a wider range of less common use cases.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128314573","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735589
Niharika R Bollumpally, Andrew C Evans, Scott W Gleave, Alexander R Gromadzki, G. Learmonth
Our client, S&P Global, is a leading provider of cross-industry data products, whose success is largely dependent on the timeliness and quality of its data. The company relies heavily on manual search across a variety of public documents to update internal records, making workflow prioritization an important component to the timeliness of its value proposition. Given the broad scope of prioritizing a highly granular workflow, our team aimed to leverage operational metadata at the lowest level: information extraction. Rather than parsing documents themselves, we aimed to preserve parsimony in developing a model capable of providing actionable insight towards workflow optimization. The selected model was trained using gradient decision tree-boosting with a logistic output, predicting the probability of task success. By combining a number of previously unused features, we were able to classify tasks that resulted in an update to any of our client's expansive datasets. The classification accuracy was measured with a ROC-AUC and the recall for the positive outcome class. Given the 98% F1 score achieved predicting at this level, we constructed a priority score, at a higher level of granularity, where the implementation of a rating system is of more practical use to our client in scheduling. The model was trained on our client's financial domain data from 2018, with hopes of generalizing our findings to other domains in the future.
{"title":"A Machine Learning Approach to Workflow Prioritization","authors":"Niharika R Bollumpally, Andrew C Evans, Scott W Gleave, Alexander R Gromadzki, G. Learmonth","doi":"10.1109/SIEDS.2019.8735589","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735589","url":null,"abstract":"Our client, S&P Global, is a leading provider of cross-industry data products, whose success is largely dependent on the timeliness and quality of its data. The company relies heavily on manual search across a variety of public documents to update internal records, making workflow prioritization an important component to the timeliness of its value proposition. Given the broad scope of prioritizing a highly granular workflow, our team aimed to leverage operational metadata at the lowest level: information extraction. Rather than parsing documents themselves, we aimed to preserve parsimony in developing a model capable of providing actionable insight towards workflow optimization. The selected model was trained using gradient decision tree-boosting with a logistic output, predicting the probability of task success. By combining a number of previously unused features, we were able to classify tasks that resulted in an update to any of our client's expansive datasets. The classification accuracy was measured with a ROC-AUC and the recall for the positive outcome class. Given the 98% F1 score achieved predicting at this level, we constructed a priority score, at a higher level of granularity, where the implementation of a rating system is of more practical use to our client in scheduling. The model was trained on our client's financial domain data from 2018, with hopes of generalizing our findings to other domains in the future.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125052563","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735603
Shubham Patel, James Mekavibul, Ja-Yeon Park, Anchit Kolla, Ryan French, Zachary Kersey, G. Lewin
Additive manufacturing (AM) - also known as 3D printing - promises a new approach to creating parts in a manufacturing environment; the process allows more design freedom and the production of parts with more complex features, compared to traditional manufacturing processes. The laser powder bed fusion (L-PBF) printer operates by building a part layer by layer in an iterative process of spreading metal powder and melting the desired shape. One particular feature is an overhang (material being melted onto the part over loose un-melted parts). However, some of the un-melted powder from the process could become melted to the overhanging feature - which is known as dross. Overhangs tend to form dross, but the extent of dross created at these features is not fully understood. Due to this unpredictable nature of dross formation, the build process exhibits variability in build quality, deterring industry-wide adoption. The conducted research aims to develop a system that analyzes cross-sectional image data captured from each layer of the print in order to identify dross with a certain level of confidence. Using machine learning techniques, images are used in a model that identifies pixels as a region that contains dross. These images are first labeled with bounding boxes (a coordinate system that identifies features/objects as existing within its boundaries) to train a neural network. The result is an adaptive model that autonomously detects dross in image scans of the part, pointing out these impurities to the printers' users, especially in regions difficult to inspect like interior surfaces of parts. The model aims to further understand L-PBF processing by location regions of excessive dross to relate dross formation with specific design features.
{"title":"Using Machine Learning to Analyze Image Data from Advanced Manufacturing Processes","authors":"Shubham Patel, James Mekavibul, Ja-Yeon Park, Anchit Kolla, Ryan French, Zachary Kersey, G. Lewin","doi":"10.1109/SIEDS.2019.8735603","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735603","url":null,"abstract":"Additive manufacturing (AM) - also known as 3D printing - promises a new approach to creating parts in a manufacturing environment; the process allows more design freedom and the production of parts with more complex features, compared to traditional manufacturing processes. The laser powder bed fusion (L-PBF) printer operates by building a part layer by layer in an iterative process of spreading metal powder and melting the desired shape. One particular feature is an overhang (material being melted onto the part over loose un-melted parts). However, some of the un-melted powder from the process could become melted to the overhanging feature - which is known as dross. Overhangs tend to form dross, but the extent of dross created at these features is not fully understood. Due to this unpredictable nature of dross formation, the build process exhibits variability in build quality, deterring industry-wide adoption. The conducted research aims to develop a system that analyzes cross-sectional image data captured from each layer of the print in order to identify dross with a certain level of confidence. Using machine learning techniques, images are used in a model that identifies pixels as a region that contains dross. These images are first labeled with bounding boxes (a coordinate system that identifies features/objects as existing within its boundaries) to train a neural network. The result is an adaptive model that autonomously detects dross in image scans of the part, pointing out these impurities to the printers' users, especially in regions difficult to inspect like interior surfaces of parts. The model aims to further understand L-PBF processing by location regions of excessive dross to relate dross formation with specific design features.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127091995","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735609
Zachary A. Marshall, Christian J. Venzlauskas, John H. Mott
While nearly 500 airports across the United States have staffed control towers, the remainder of the more than 19,000 airports nationwide lack the resources necessary to record and store operations data. These smaller airports, though forming the backbone of America's general aviation industry, face growing safety risks, as they are somewhat disadvantaged when applying for improvement funding through the Federal Aviation Administration's Airport Improvement Program. Airport data, such as fleet mix, takeoff and landing counts, and weather conditions, can be collected from various commercial sensors monitoring airfield operations and analyzed to identify risk factors and improve corresponding safety measures. However, rural airports must transmit this data wirelessly from antennas mounted at low elevations to network access points located potentially at considerable distances from those points in a power-efficient and cost-effective manner. An antenna system was designed, manufactured, and tested within the height, power, and cost constraints of these smaller airfields to explore the economic viability and technical feasibility of facilitating a data-driven safety improvement program. This system intends to mitigate the multipath interference that confounds data transmitted over long ranges at low altitudes, enabling compilation of accurate rural aviation operations information. Empowering airport managers with a reliable and efficient Internet connection to collect the data that influences federal grant allocations, this system would directly enhance the safety of America's aging general aviation infrastructure and stimulate America's depleted rural community economies.
{"title":"Revitalizing Rural Communities Through Enhanced Aviation Microwave Data Transmission Systems","authors":"Zachary A. Marshall, Christian J. Venzlauskas, John H. Mott","doi":"10.1109/SIEDS.2019.8735609","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735609","url":null,"abstract":"While nearly 500 airports across the United States have staffed control towers, the remainder of the more than 19,000 airports nationwide lack the resources necessary to record and store operations data. These smaller airports, though forming the backbone of America's general aviation industry, face growing safety risks, as they are somewhat disadvantaged when applying for improvement funding through the Federal Aviation Administration's Airport Improvement Program. Airport data, such as fleet mix, takeoff and landing counts, and weather conditions, can be collected from various commercial sensors monitoring airfield operations and analyzed to identify risk factors and improve corresponding safety measures. However, rural airports must transmit this data wirelessly from antennas mounted at low elevations to network access points located potentially at considerable distances from those points in a power-efficient and cost-effective manner. An antenna system was designed, manufactured, and tested within the height, power, and cost constraints of these smaller airfields to explore the economic viability and technical feasibility of facilitating a data-driven safety improvement program. This system intends to mitigate the multipath interference that confounds data transmitted over long ranges at low altitudes, enabling compilation of accurate rural aviation operations information. Empowering airport managers with a reliable and efficient Internet connection to collect the data that influences federal grant allocations, this system would directly enhance the safety of America's aging general aviation infrastructure and stimulate America's depleted rural community economies.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127189231","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735604
Brayden Beardsley, A. Brewer, Matthew Gummersbach, Zachary Houck, S. Humbert, Edward J. O'Rourke, Nicholas Verham, Benjamin J. Lobo, Donald Brown
After undergoing cardiac surgery, a significant number of patients develop Acute Kidney Injury (AKI), a condition that contributes to higher mortality and morbidity rates. Current methods of diagnosing AKI are largely reactionary, as kidney damage can only be assessed after creatinine levels in the blood rise, a process that occurs 24-48 hours after initial injury. During this time period, doctors make medical decisions that may add extra stress to kidney function, unknowingly contributing to further kidney damage. The University of Virginia (UVa) Health System is interested in improving its ability to predict AKI following cardiac surgery in order to more quickly and accurately identify at-risk patients. Currently, the UVa Health System uses the Society of Thoracic Surgeons (STS) preoperative AKI Risk Score to assess each patient's risk of kidney injury prior to surgery. Hoping to improve predictive performance, the Health System desires a new risk model that also incorporates risk factors from the intraoperative period. The final dataset ($mathrm{n}=335$ surgeries) includes both preoperative and intraoperative factors compiled from the UVa Health System EMR database. Machine learning models were utilized to predict each patient's change in creatinine level, the metric used to assign AKI classifications. Specific focus was given to incorporating intraoperative time series factors. Changepoint analysis, estimated entropy, and heteroscedastic modeling were employed to analyze the time series readings from lab, anesthesiology, and medication records taken during cardiac surgery. Several of these intraoperative time series features were significant variables in all of the highest performing L1 Linear Regression, L1 Logistic Regression, Random Forest, Neural Net, and Extreme Gradient Boost models.
在接受心脏手术后，相当多的患者会出现急性肾损伤(AKI)，这种情况会导致更高的死亡率和发病率。目前诊断AKI的方法在很大程度上是反应性的，因为只有在血液中肌酐水平升高后才能评估肾脏损害，这一过程发生在初始损伤后24-48小时。在这段时间里，医生做出的医疗决定可能会给肾脏功能增加额外的压力，在不知不觉中导致肾脏进一步受损。弗吉尼亚大学(UVa)卫生系统有意提高其预测心脏手术后AKI的能力，以便更快、更准确地识别高危患者。目前，UVa健康系统使用胸外科学会(STS)术前AKI风险评分来评估每位患者在手术前肾损伤的风险。为了提高预测性能，卫生系统需要一种新的风险模型，该模型还包括术中期的风险因素。最终数据集($ mathm {n}=335$ surgery)包括从UVa Health System EMR数据库编译的术前和术中因素。使用机器学习模型来预测每位患者肌酐水平的变化，肌酐水平是用于分配AKI分类的指标。特别关注纳入术中时间序列因素。采用变点分析、估计熵和异方差模型分析来自实验室、麻醉学和心脏手术期间用药记录的时间序列读数。在所有性能最高的L1线性回归、L1逻辑回归、随机森林、神经网络和极端梯度Boost模型中，这些术中时间序列特征中的几个是重要的变量。
{"title":"Using Intraoperative Variables to Predict Acute Kidney Injury Following Cardiac Surgery","authors":"Brayden Beardsley, A. Brewer, Matthew Gummersbach, Zachary Houck, S. Humbert, Edward J. O'Rourke, Nicholas Verham, Benjamin J. Lobo, Donald Brown","doi":"10.1109/SIEDS.2019.8735604","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735604","url":null,"abstract":"After undergoing cardiac surgery, a significant number of patients develop Acute Kidney Injury (AKI), a condition that contributes to higher mortality and morbidity rates. Current methods of diagnosing AKI are largely reactionary, as kidney damage can only be assessed after creatinine levels in the blood rise, a process that occurs 24-48 hours after initial injury. During this time period, doctors make medical decisions that may add extra stress to kidney function, unknowingly contributing to further kidney damage. The University of Virginia (UVa) Health System is interested in improving its ability to predict AKI following cardiac surgery in order to more quickly and accurately identify at-risk patients. Currently, the UVa Health System uses the Society of Thoracic Surgeons (STS) preoperative AKI Risk Score to assess each patient's risk of kidney injury prior to surgery. Hoping to improve predictive performance, the Health System desires a new risk model that also incorporates risk factors from the intraoperative period. The final dataset ($mathrm{n}=335$ surgeries) includes both preoperative and intraoperative factors compiled from the UVa Health System EMR database. Machine learning models were utilized to predict each patient's change in creatinine level, the metric used to assign AKI classifications. Specific focus was given to incorporating intraoperative time series factors. Changepoint analysis, estimated entropy, and heteroscedastic modeling were employed to analyze the time series readings from lab, anesthesiology, and medication records taken during cardiac surgery. Several of these intraoperative time series features were significant variables in all of the highest performing L1 Linear Regression, L1 Logistic Regression, Random Forest, Neural Net, and Extreme Gradient Boost models.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127313464","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735630
Hollie P. Coleman, Rajan D. Jani, Victoria G. Lum, Kelly L. Norfleet, William J. Rimer, Louis G. Tanous, Matthew R. Wajsgras, D. J. Andrews, Thomas L. Polmateer, Daniel C. Hendrickson, J. Lambert
Recent trends in markets, technologies, demographics, regulations, organizations, energy, and environments require adaptations for operations of container ports. This paper describes resilience of ports to emergent conditions in three areas: Energy and fuels, controls and automation, and logistics transformation. First, the sustainability of energy and fuels have become a priority for ports due to growing concerns of carbon emissions. Liquified natural gas (LNG) is both a cost-effective and safe alternative fuel for reducing emissions of container ships. The feasibility and several strategies for adopting LNG bunkering at ports are described. Second, ports must leverage technological development to drive greater efficiency and operative resiliency. Recent developments in industrial automation and electrification are described. Lastly, the impact of corporate development to the region of the port is discussed. An analysis of enterprise capacity expansion includes both risks and benefits to increase competitiveness. This paper uses methodology of systems and data analysis, mathematical simulation, and risk cost-benefit optimization. The key results include identifying operational and security risks for container ports and providing strategies for risk mitigation and resilience. The recommendations are discussed in context of a USD $750 million ten-year strategic plan of a major container port on the Atlantic Coast of the United States.
{"title":"Enterprise Resilience and Sustainability for Operations of Maritime Container Ports","authors":"Hollie P. Coleman, Rajan D. Jani, Victoria G. Lum, Kelly L. Norfleet, William J. Rimer, Louis G. Tanous, Matthew R. Wajsgras, D. J. Andrews, Thomas L. Polmateer, Daniel C. Hendrickson, J. Lambert","doi":"10.1109/SIEDS.2019.8735630","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735630","url":null,"abstract":"Recent trends in markets, technologies, demographics, regulations, organizations, energy, and environments require adaptations for operations of container ports. This paper describes resilience of ports to emergent conditions in three areas: Energy and fuels, controls and automation, and logistics transformation. First, the sustainability of energy and fuels have become a priority for ports due to growing concerns of carbon emissions. Liquified natural gas (LNG) is both a cost-effective and safe alternative fuel for reducing emissions of container ships. The feasibility and several strategies for adopting LNG bunkering at ports are described. Second, ports must leverage technological development to drive greater efficiency and operative resiliency. Recent developments in industrial automation and electrification are described. Lastly, the impact of corporate development to the region of the port is discussed. An analysis of enterprise capacity expansion includes both risks and benefits to increase competitiveness. This paper uses methodology of systems and data analysis, mathematical simulation, and risk cost-benefit optimization. The key results include identifying operational and security risks for container ports and providing strategies for risk mitigation and resilience. The recommendations are discussed in context of a USD $750 million ten-year strategic plan of a major container port on the Atlantic Coast of the United States.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129974130","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735625
Katie Carlson, Ashif Chowdhury, A. Kepley, E. Somerville, Kevin Warshaw, J. Goodall
There is evidence that flooding events are becoming more frequent and intense as a result of climate change. This problem is especially prevalent in Norfolk, VA which has the second highest rate of sea level rise on the east coast. Model and sensing innovations are needed to produce high-resolution flood warnings in real-time to improve public safety. New sensing approaches are also needed to accurately measure the extent of flooding during storm events so this data can be used to calibrate models. Our methodology creates an end-to-end modeling system for Norfolk, VA to provide real-time flood forecast information to users. Our process begins with data collection through our group's water level sensor. This device relies on an ultrasonic sensor to measure how its distance from the ground changes as water levels rise. Readings are then filtered before they are transmitted to a persistent database. The data from this sensor, combined with historical flood data, are stored in a locally-hosted relational SQLite database and a cloud-hosted InfluxDB database. The locally-hosted database can be used for further development of flood prediction models. The cloud-hosted database can store data as it is collected for real time analysis. Currently, the sensor has accurately recorded changes in distances of up to ten feet in the lab and successfully transmitted these readings. For future testing, measurements will be sent to a static URL hosted on Heroku. A Python function has been written that reads the URL in JSON format and transmits the data to the Influx database. Another Python function has been written that reads a csv containing historical data and transforms it to the proper format, then inserts it into SQLite.
{"title":"Smart Cities Solutions for More Flood Resilient Communities","authors":"Katie Carlson, Ashif Chowdhury, A. Kepley, E. Somerville, Kevin Warshaw, J. Goodall","doi":"10.1109/SIEDS.2019.8735625","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735625","url":null,"abstract":"There is evidence that flooding events are becoming more frequent and intense as a result of climate change. This problem is especially prevalent in Norfolk, VA which has the second highest rate of sea level rise on the east coast. Model and sensing innovations are needed to produce high-resolution flood warnings in real-time to improve public safety. New sensing approaches are also needed to accurately measure the extent of flooding during storm events so this data can be used to calibrate models. Our methodology creates an end-to-end modeling system for Norfolk, VA to provide real-time flood forecast information to users. Our process begins with data collection through our group's water level sensor. This device relies on an ultrasonic sensor to measure how its distance from the ground changes as water levels rise. Readings are then filtered before they are transmitted to a persistent database. The data from this sensor, combined with historical flood data, are stored in a locally-hosted relational SQLite database and a cloud-hosted InfluxDB database. The locally-hosted database can be used for further development of flood prediction models. The cloud-hosted database can store data as it is collected for real time analysis. Currently, the sensor has accurately recorded changes in distances of up to ten feet in the lab and successfully transmitted these readings. For future testing, measurements will be sent to a static URL hosted on Heroku. A Python function has been written that reads the URL in JSON format and transmits the data to the Influx database. Another Python function has been written that reads a csv containing historical data and transforms it to the proper format, then inserts it into SQLite.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127837634","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}
Pub Date : 2019-04-01DOI: 10.1109/SIEDS.2019.8735636
Allison Lee, Greg Connelly, Raewyn L. Haines, A. Lyons, Timothy R. Eddy, Y. Haimes
Founded with a vision for well-rounded education, the University of Virginia is a Complex System of Systems with subsystems that interface to fulfill the University's core purposes. The interdependence and interconnectedness of subsystems, including schools and departments, provide insight into the University's capability to bridge disciplinary boundaries in offering holistic learning opportunities to faculty and students. Interscholastic courses, classes offered in two or more schools; and interdisciplinary courses, classes combining two or more disciplines, are examples of such opportunities. Thus, the team focused its analytical and modeling efforts on identifying obstacles to these courses and to other programs that could cultivate both a knowledge of disciplinary perspectives and skills in disciplinary integration at the undergraduate level. Interviews with administrators and professors as well as a survey distributed to professors of interscholastic courses were used to gather qualitative data about experiences in forming and administering courses listed in both the School of Engineering and Applied Science and the College of Arts and Sciences. Students in both schools were also surveyed about their awareness and interest in interdisciplinary and interscholastic classes. This systems analysis utilized both quantitative data and, primarily, qualitative insights regarding personal motivations and attitudes in understanding the intricacies of the University as a Complex System of Systems and identifying contradictory objectives, key limiting resources, and relevant cultural factors. This research highlights existing impediments to interdisciplinary and interscholastic collaboration within the University, as well as recommendations on how these barriers can be lowered.
{"title":"Lowering Barriers to Interscholastic Undergraduate Initiatives at the University of Virginia","authors":"Allison Lee, Greg Connelly, Raewyn L. Haines, A. Lyons, Timothy R. Eddy, Y. Haimes","doi":"10.1109/SIEDS.2019.8735636","DOIUrl":"https://doi.org/10.1109/SIEDS.2019.8735636","url":null,"abstract":"Founded with a vision for well-rounded education, the University of Virginia is a Complex System of Systems with subsystems that interface to fulfill the University's core purposes. The interdependence and interconnectedness of subsystems, including schools and departments, provide insight into the University's capability to bridge disciplinary boundaries in offering holistic learning opportunities to faculty and students. Interscholastic courses, classes offered in two or more schools; and interdisciplinary courses, classes combining two or more disciplines, are examples of such opportunities. Thus, the team focused its analytical and modeling efforts on identifying obstacles to these courses and to other programs that could cultivate both a knowledge of disciplinary perspectives and skills in disciplinary integration at the undergraduate level. Interviews with administrators and professors as well as a survey distributed to professors of interscholastic courses were used to gather qualitative data about experiences in forming and administering courses listed in both the School of Engineering and Applied Science and the College of Arts and Sciences. Students in both schools were also surveyed about their awareness and interest in interdisciplinary and interscholastic classes. This systems analysis utilized both quantitative data and, primarily, qualitative insights regarding personal motivations and attitudes in understanding the intricacies of the University as a Complex System of Systems and identifying contradictory objectives, key limiting resources, and relevant cultural factors. This research highlights existing impediments to interdisciplinary and interscholastic collaboration within the University, as well as recommendations on how these barriers can be lowered.","PeriodicalId":265421,"journal":{"name":"2019 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2019-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124435431","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：