Pub Date : 2020-04-01DOI: 10.1109/sieds49339.2020.9106635
{"title":"SIEDS 2020 TOC","authors":"","doi":"10.1109/sieds49339.2020.9106635","DOIUrl":"https://doi.org/10.1109/sieds49339.2020.9106635","url":null,"abstract":"","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114261104","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106642
Menuka Jaiswal, Saad Saleem, Yonghyeon Kweon, Eli J. Draizen, S. Veretnik, C. Mura, P. Bourne
Recent computational advances in the accurate prediction of protein three-dimensional (3D) structures from amino acid sequences now present a unique opportunity to decipher the interrelationships between proteins. This task entails—but is not equivalent to—a problem of 3D structure comparison and classification. Historically, protein domain classification has been a largely manual and subjective activity, relying upon various heuristics. Databases such as CATH represent significant steps towards a more systematic (and automatable) approach, yet there still remains much room for the development of more scalable and quantitative classification methods, grounded in machine learning. We suspect that re-examining these relationships via a Deep Learning (DL) approach may entail a large-scale restructuring of classification schemes, improved with respect to the interpretability of distant relationships between proteins. Here, we describe our training of DL models on protein domain structures (and their associated physicochemical properties) in order to evaluate classification properties at CATH’s “homologous superfamily” (SF) level. To achieve this, we have devised and applied an extension of image-classification methods and image segmentation techniques, utilizing a convolutional autoencoder model architecture. Our DL architecture allows models to learn structural features that, in a sense, ‘define’ different homologous SFs. We evaluate and quantify pairwise ‘distances’ between SFs by building one model per SF and comparing the loss functions of the models. Hierarchical clustering on these distance matrices provides a new view of protein interrelationships—a view that extends beyond simple structural/geometric similarity, and towards the realm of structure/function properties.
{"title":"Deep Learning of Protein Structural Classes: Any Evidence for an ‘Urfold’?","authors":"Menuka Jaiswal, Saad Saleem, Yonghyeon Kweon, Eli J. Draizen, S. Veretnik, C. Mura, P. Bourne","doi":"10.1109/SIEDS49339.2020.9106642","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106642","url":null,"abstract":"Recent computational advances in the accurate prediction of protein three-dimensional (3D) structures from amino acid sequences now present a unique opportunity to decipher the interrelationships between proteins. This task entails—but is not equivalent to—a problem of 3D structure comparison and classification. Historically, protein domain classification has been a largely manual and subjective activity, relying upon various heuristics. Databases such as CATH represent significant steps towards a more systematic (and automatable) approach, yet there still remains much room for the development of more scalable and quantitative classification methods, grounded in machine learning. We suspect that re-examining these relationships via a Deep Learning (DL) approach may entail a large-scale restructuring of classification schemes, improved with respect to the interpretability of distant relationships between proteins. Here, we describe our training of DL models on protein domain structures (and their associated physicochemical properties) in order to evaluate classification properties at CATH’s “homologous superfamily” (SF) level. To achieve this, we have devised and applied an extension of image-classification methods and image segmentation techniques, utilizing a convolutional autoencoder model architecture. Our DL architecture allows models to learn structural features that, in a sense, ‘define’ different homologous SFs. We evaluate and quantify pairwise ‘distances’ between SFs by building one model per SF and comparing the loss functions of the models. Hierarchical clustering on these distance matrices provides a new view of protein interrelationships—a view that extends beyond simple structural/geometric similarity, and towards the realm of structure/function properties.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122047158","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106649
E. Nittinger, G. Arce, Grant Gemici, Valeria Soto
Currently, available flood modeling approaches require High-Performance Computing (HPC) software and high-resolution terrain data. Developing countries with unstable dams face challenges in accessing these technologies and acquiring the required data elements. This project developed a dynamic flood modeling methodology, using established hydrological assumptions, that implemented simulation and optimization models to determine safe evacuation routes by using public datasets, publicly available technical expertise, and common computing capabilities. The Péligre Dam in Haiti was the case study site since data from an HPC model was available for results comparison. QGIS was used to extract the “water flow factors” which are: (i) channel slope, turns, and shape; (ii) major channel obstructions and channel terrain; (iii) floodplain shape; and (iv) major floodplain obstructions and floodplain terrain. A system dynamics model was created to simulate water flow as a function of time using Vensim. This model used the water flow factors as inputs and produced the following key outputs: (i) volumetric flow rate $(mathrm{Q}_{mathrm{i},mathrm{t}})$, (ii) water height over time $(mathrm{h}_{mathrm{i},mathrm{t}})$, (iii) time when actual flooding begins (ChannelMAX), and (iv) time of maximal flooding (FloodplainMAX). The results were plotted, and the root mean square errors were calculated to visualize the extent to which the results from the systems dynamics model compare with the HPC software results. Evacuation routes were modeled with the shortest path algorithm by minimizing the feasible travel distance between at-risk populated areas and safe-high-ground areas with route constraints based on the system dynamics model’s output. The validity of the results demonstrates that the proposed methodology can adequately model inundation and reliable evacuation routes for dam failure scenarios in developing countries.
{"title":"System Dynamics Flood Modeling Framework for Dam Failure Evacuation Planning in Developing Countries","authors":"E. Nittinger, G. Arce, Grant Gemici, Valeria Soto","doi":"10.1109/SIEDS49339.2020.9106649","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106649","url":null,"abstract":"Currently, available flood modeling approaches require High-Performance Computing (HPC) software and high-resolution terrain data. Developing countries with unstable dams face challenges in accessing these technologies and acquiring the required data elements. This project developed a dynamic flood modeling methodology, using established hydrological assumptions, that implemented simulation and optimization models to determine safe evacuation routes by using public datasets, publicly available technical expertise, and common computing capabilities. The Péligre Dam in Haiti was the case study site since data from an HPC model was available for results comparison. QGIS was used to extract the “water flow factors” which are: (i) channel slope, turns, and shape; (ii) major channel obstructions and channel terrain; (iii) floodplain shape; and (iv) major floodplain obstructions and floodplain terrain. A system dynamics model was created to simulate water flow as a function of time using Vensim. This model used the water flow factors as inputs and produced the following key outputs: (i) volumetric flow rate $(mathrm{Q}_{mathrm{i},mathrm{t}})$, (ii) water height over time $(mathrm{h}_{mathrm{i},mathrm{t}})$, (iii) time when actual flooding begins (ChannelMAX), and (iv) time of maximal flooding (FloodplainMAX). The results were plotted, and the root mean square errors were calculated to visualize the extent to which the results from the systems dynamics model compare with the HPC software results. Evacuation routes were modeled with the shortest path algorithm by minimizing the feasible travel distance between at-risk populated areas and safe-high-ground areas with route constraints based on the system dynamics model’s output. The validity of the results demonstrates that the proposed methodology can adequately model inundation and reliable evacuation routes for dam failure scenarios in developing countries.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125598834","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106587
Heidi J. Schellin, Tatiana N. Oberley, Kaitlyn M. Patterson, Boyoung Kim, Kerstin S Haring, Chad C. Tossell, Elizabeth Phillips, E. D. Visser
Commercial robotic dogs, such as Sony’s Aibo, have recently been reimagined. Our goal with this research was to examine factors that influence human-robot dog bonding. We created a 2x2 between-subjects experiment, by framing the Aibo as a puppy or robot, and by adding fur to the Aibo or not. We hypothesized that bonding would be stronger when the robotic dog was framed to participants as a puppy rather than a robot, and it would be stronger when the robotic dog was dressed in a fur suit. Results showed that participants were less positive toward Aibo when framed as a puppy compared to when Aibo was framed as a robot. Adding fur had a positive effect: Aibo was considered less scary compared to having no fur. In addition, behavioral interaction results showed that asking the Aibo to “come here” was the most popular command with respect to the number of completed actions, failures, and social behavior, and the time spent. Our approach could inform design in a way that integrates dogs into the work force to help people relieve boredom, stress, and help them carry out their jobs more effectively and cost efficiently.
{"title":"Man’s New Best Friend? Strengthening Human-Robot Dog Bonding by Enhancing the Doglikeness of Sony’s Aibo","authors":"Heidi J. Schellin, Tatiana N. Oberley, Kaitlyn M. Patterson, Boyoung Kim, Kerstin S Haring, Chad C. Tossell, Elizabeth Phillips, E. D. Visser","doi":"10.1109/SIEDS49339.2020.9106587","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106587","url":null,"abstract":"Commercial robotic dogs, such as Sony’s Aibo, have recently been reimagined. Our goal with this research was to examine factors that influence human-robot dog bonding. We created a 2x2 between-subjects experiment, by framing the Aibo as a puppy or robot, and by adding fur to the Aibo or not. We hypothesized that bonding would be stronger when the robotic dog was framed to participants as a puppy rather than a robot, and it would be stronger when the robotic dog was dressed in a fur suit. Results showed that participants were less positive toward Aibo when framed as a puppy compared to when Aibo was framed as a robot. Adding fur had a positive effect: Aibo was considered less scary compared to having no fur. In addition, behavioral interaction results showed that asking the Aibo to “come here” was the most popular command with respect to the number of completed actions, failures, and social behavior, and the time spent. Our approach could inform design in a way that integrates dogs into the work force to help people relieve boredom, stress, and help them carry out their jobs more effectively and cost efficiently.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116976784","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106678
Mariah Hurt, Arti Patel, Shenghua Wu, G. Learmonth
Numerous studies on corporate leadership show a disproportionately high number of men relative to women serving as Chief Executive Officer (CEO). According to BoardEx, as of the end of 2019 only approximately five percent of Standard and Poor’s 500 Index companies were led by a female CEO. Working with Standard and Poor’s Global Market Intelligence financial data from Compustat along with CEO gender data from Catalyst to investigate when female CEOs are hired and which financial factors contribute to a company having a female CEO. As having a female CEO is only one measure of a company’s dedication to diversity, two additional metrics were used as well, companies whose CEOs have taken the Catalyst Champions for Change Pledge and a list of SSGA Index constituents. The aim of this study is to understand under what financial circumstances a company hires a female CEO as opposed to a male CEO, and to explore if female led firms display financial patterns that are distinct from male led firms. Our approach is twofold, using time series analysis to interrogate trends in share price in the 12 months preceding and 12 months following CEO hire, and using clustering methods to better understand if certain financial metrics characterize companies that hire a female CEO. Our study is focused on Standard and Poor’s 500 Index companies that have hired a new CEO since the year 2000. Financial metrics explored in this study are stock returns, stock volatility, earnings per share, EBITA margin, gross profit margin, cash flow margin, return on assets, return on equity and return on capital. Dimension reduction was used and k-means clustering performed on the first three principal components. Preliminary results indicate that male and female led firms cluster together financially. Due to the inevitably high male to female ratio in corporate leadership, there are many fewer female led companies and these naturally group within the high density clusters.
大量关于企业领导的研究表明,担任首席执行官的男性比例高于女性。根据BoardEx的数据,截至2019年底,标准普尔500指数公司中只有约5%的公司由女性首席执行官领导。结合Compustat提供的标准普尔全球市场情报公司的财务数据,以及Catalyst提供的CEO性别数据,调查女性CEO何时被聘用,以及哪些财务因素促使公司聘请女性CEO。由于拥有女性首席执行官只是衡量公司致力于多元化的一个指标,因此还使用了两个额外的指标,即首席执行官参加Catalyst Champions for Change Pledge的公司和SSGA指数成分股名单。本研究的目的是了解在什么样的财务环境下,一家公司雇佣女性CEO而不是男性CEO,并探讨女性领导的公司是否表现出与男性领导的公司不同的财务模式。我们的方法是双重的,使用时间序列分析来询问CEO聘用之前和之后12个月的股价趋势,并使用聚类方法来更好地了解聘请女性CEO的公司是否具有某些财务指标特征。我们的研究重点是自2000年以来聘请了新CEO的标准普尔500指数公司。本研究探讨的财务指标包括股票收益、股票波动性、每股收益、息税前利润率、毛利率、现金流利润率、资产收益率、股本收益率和资本收益率。采用降维方法对前三个主成分进行k-means聚类。初步结果表明,男性和女性领导的公司在财务上聚集在一起。由于企业领导层中不可避免的高男女比例,女性领导的公司要少得多,这些公司自然会聚集在高密度的集群中。
{"title":"An Exploration and Characterization of Financial Performance of Standard and Poor’s 500 Index Constituents Led By Female CEOs","authors":"Mariah Hurt, Arti Patel, Shenghua Wu, G. Learmonth","doi":"10.1109/SIEDS49339.2020.9106678","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106678","url":null,"abstract":"Numerous studies on corporate leadership show a disproportionately high number of men relative to women serving as Chief Executive Officer (CEO). According to BoardEx, as of the end of 2019 only approximately five percent of Standard and Poor’s 500 Index companies were led by a female CEO. Working with Standard and Poor’s Global Market Intelligence financial data from Compustat along with CEO gender data from Catalyst to investigate when female CEOs are hired and which financial factors contribute to a company having a female CEO. As having a female CEO is only one measure of a company’s dedication to diversity, two additional metrics were used as well, companies whose CEOs have taken the Catalyst Champions for Change Pledge and a list of SSGA Index constituents. The aim of this study is to understand under what financial circumstances a company hires a female CEO as opposed to a male CEO, and to explore if female led firms display financial patterns that are distinct from male led firms. Our approach is twofold, using time series analysis to interrogate trends in share price in the 12 months preceding and 12 months following CEO hire, and using clustering methods to better understand if certain financial metrics characterize companies that hire a female CEO. Our study is focused on Standard and Poor’s 500 Index companies that have hired a new CEO since the year 2000. Financial metrics explored in this study are stock returns, stock volatility, earnings per share, EBITA margin, gross profit margin, cash flow margin, return on assets, return on equity and return on capital. Dimension reduction was used and k-means clustering performed on the first three principal components. Preliminary results indicate that male and female led firms cluster together financially. Due to the inevitably high male to female ratio in corporate leadership, there are many fewer female led companies and these naturally group within the high density clusters.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121988508","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106682
Deepaloke Chattopadhyay, Sania Rasheed, Luyuanyuan Yan, Alfonso A. Lopez, Jay Farmer, Donald E. Brown
All-Electronic Tolling Systems, a global market worth approximately $7B, have made significant contributions in toll collection, commuter convenience, traffic management and highway administration. The current infrastructure however, is multi-tiered and expensive to set up. Alternative ways of vehicle detection can help in significantly lowering costs on new toll infrastructure placement. In this paper, we apply a new perspective to the detection problem by evaluating the applicability of machine learning for detecting vehicle movement through toll gantries in real-time from a novel perpendicular overhead camera angle. We solve this multi-objective problem by incorporating object detection using You Only Look Once (YOLO), more specifically, YOLOv3 and a faster version with less memory requirements, Tiny YOLOv3 to detect vehicles passing through tolls from perpendicular overhead angles in real time and with high accuracy. Additionally, a classification is made between passenger vehicles and trucks/buses of the detected vehicles. Our Experimental results from training YOLOv3 on our data set indicate a recall of 100.0% and a precision of 98.0%. The results of training Tiny YOLOv3 on our data set show a recall of 100.0% and a precision of 98.5%. These results indicate that use of machine learning is not only effective for detecting vehicles in electronic tolling systems in real-time, but that it can be used on cameras positioned at a perpendicular angle despite insufficient annotations.
全电子收费系统的全球市场价值约为70亿美元,在收费、通勤便利、交通管理和高速公路管理方面做出了重大贡献。然而,目前的基础设施是多层的,建立起来很昂贵。车辆检测的替代方法可以帮助显著降低新建收费基础设施的成本。在本文中,我们通过评估机器学习在从新的垂直头顶摄像机角度实时检测通过收费门的车辆运动方面的适用性,为检测问题提供了一个新的视角。我们通过使用You Only Look Once (YOLO)结合目标检测来解决这个多目标问题,更具体地说,YOLOv3和更快的版本,内存要求更少,Tiny YOLOv3可以实时、高精度地从垂直的顶角检测通过收费站的车辆。此外,还对被检测车辆的乘用车和卡车/公共汽车进行分类。在我们的数据集上训练YOLOv3的实验结果表明,召回率为100.0%,准确率为98.0%。在我们的数据集上训练Tiny YOLOv3的结果显示召回率为100.0%,准确率为98.5%。这些结果表明,使用机器学习不仅可以有效地实时检测电子收费系统中的车辆,而且可以在没有足够注释的情况下用于垂直角度的摄像机。
{"title":"Machine Learning for Real-Time Vehicle Detection in All-Electronic Tolling System","authors":"Deepaloke Chattopadhyay, Sania Rasheed, Luyuanyuan Yan, Alfonso A. Lopez, Jay Farmer, Donald E. Brown","doi":"10.1109/SIEDS49339.2020.9106682","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106682","url":null,"abstract":"All-Electronic Tolling Systems, a global market worth approximately $7B, have made significant contributions in toll collection, commuter convenience, traffic management and highway administration. The current infrastructure however, is multi-tiered and expensive to set up. Alternative ways of vehicle detection can help in significantly lowering costs on new toll infrastructure placement. In this paper, we apply a new perspective to the detection problem by evaluating the applicability of machine learning for detecting vehicle movement through toll gantries in real-time from a novel perpendicular overhead camera angle. We solve this multi-objective problem by incorporating object detection using You Only Look Once (YOLO), more specifically, YOLOv3 and a faster version with less memory requirements, Tiny YOLOv3 to detect vehicles passing through tolls from perpendicular overhead angles in real time and with high accuracy. Additionally, a classification is made between passenger vehicles and trucks/buses of the detected vehicles. Our Experimental results from training YOLOv3 on our data set indicate a recall of 100.0% and a precision of 98.0%. The results of training Tiny YOLOv3 on our data set show a recall of 100.0% and a precision of 98.5%. These results indicate that use of machine learning is not only effective for detecting vehicles in electronic tolling systems in real-time, but that it can be used on cameras positioned at a perpendicular angle despite insufficient annotations.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124597815","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106658
Shayne Cassidy, M. Coulter, Thomas Finkelston, Klara Hoherchak, Antonio Mendes, Griffin Ott, Colin Patton, Kaila Stein, B. Etienne, G. Louis, M. Lerdau
The Atlantic hurricane season brings devastation to Small Island Developing States (SIDS) each year. SIDS, as designated by the United Nations, are developing countries with specific vulnerabilities due to their small size and susceptibility to environmental disasters [1]. These disasters have led to episodic food insecurity and disruption of agricultural livelihoods. In order to address this problem, a 3-Cavaliers research group at the University of Virginia partnered with Babylon Micro-Farms (BMF), a company in Charlottesville, VA, to develop a hydroponic crop cultivation (HCC) system for use in the Caribbean as a test case for SIDS. Hydroponics refers to the cultivation of plants through a nutrient rich solution without the need for soil. The research team worked with previous Capstone teams to develop and test a low-cost HCC system in the island of Dominica after Hurricane Maria. The current Capstone project aims to enhance this design by strengthening its resilience to storms and hurricanes. The resulting hydroponics system design for SIDS has four main parts: the plant growth unit, the nutrient water system, the solar power unit, and the structure. The plant growth unit is made up of PVC pipes that are on a top base and covered by an agricultural fabric to protect from insects. This base is angled to initiate the gravitational flow of the nutrient enriched water that is pumped in by the system. The solar power unit provides power to the system of pumps that move the water and nutrients throughout the system. The plant growth unit is supported by a collapsible wooden frame to reduce wind loading on the structure. Once the structure is collapsed, a durable plastic cover can be pulled over the unit to further protect the plants from wind. Both the plant growth unit and the base of the structure are designed to be easily transportable so users can safely store the unit inside during intense storms.Ultimately, we will create blueprints of the structure and design plans for a prototype that will be used and monitored in Abaco, Bahamas. We will also deliver a conditional recommendation on which crops to grow with respect to five specific criteria: harvest period, yield, seed price, market price, and nutritional value. This will allow the farmer to determine his particular crop species based on his own unique goals for hydroponic farming.
{"title":"Hydroponic Crop Cultivation (HCC) for Food Security in Small Island Developing States","authors":"Shayne Cassidy, M. Coulter, Thomas Finkelston, Klara Hoherchak, Antonio Mendes, Griffin Ott, Colin Patton, Kaila Stein, B. Etienne, G. Louis, M. Lerdau","doi":"10.1109/SIEDS49339.2020.9106658","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106658","url":null,"abstract":"The Atlantic hurricane season brings devastation to Small Island Developing States (SIDS) each year. SIDS, as designated by the United Nations, are developing countries with specific vulnerabilities due to their small size and susceptibility to environmental disasters [1]. These disasters have led to episodic food insecurity and disruption of agricultural livelihoods. In order to address this problem, a 3-Cavaliers research group at the University of Virginia partnered with Babylon Micro-Farms (BMF), a company in Charlottesville, VA, to develop a hydroponic crop cultivation (HCC) system for use in the Caribbean as a test case for SIDS. Hydroponics refers to the cultivation of plants through a nutrient rich solution without the need for soil. The research team worked with previous Capstone teams to develop and test a low-cost HCC system in the island of Dominica after Hurricane Maria. The current Capstone project aims to enhance this design by strengthening its resilience to storms and hurricanes. The resulting hydroponics system design for SIDS has four main parts: the plant growth unit, the nutrient water system, the solar power unit, and the structure. The plant growth unit is made up of PVC pipes that are on a top base and covered by an agricultural fabric to protect from insects. This base is angled to initiate the gravitational flow of the nutrient enriched water that is pumped in by the system. The solar power unit provides power to the system of pumps that move the water and nutrients throughout the system. The plant growth unit is supported by a collapsible wooden frame to reduce wind loading on the structure. Once the structure is collapsed, a durable plastic cover can be pulled over the unit to further protect the plants from wind. Both the plant growth unit and the base of the structure are designed to be easily transportable so users can safely store the unit inside during intense storms.Ultimately, we will create blueprints of the structure and design plans for a prototype that will be used and monitored in Abaco, Bahamas. We will also deliver a conditional recommendation on which crops to grow with respect to five specific criteria: harvest period, yield, seed price, market price, and nutritional value. This will allow the farmer to determine his particular crop species based on his own unique goals for hydroponic farming.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124843327","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106675
Boyoung Kim, Micala Bruce, LeSean Brown, E. D. Visser, Elizabeth Phillips
The uncanny valley hypothesis posits that people’s emotional responses to robots are increasingly positive as robots’ resemblance to humans increases. However, when robots closely, but imperfectly resemble humans, people’s responses turn negative, only to revert back once their appearance more closely resembles humans. These sharp emotional transitions (i.e., peaks and valleys in emotional response) from positive to negative, and then back to positive, are collectively referred to as the uncanny valley. In this project, we attempted to validate the uncanny valley with the largest set of real-world robots currently available in open source format (the ABOT Database). Participants saw static images of 251 robots which varied in their degree of human-likeness, and rated them on uncanniness. We found significant empirical support not only for the hypothesized uncanny valley but an additional valley. This unanticipated valley emerged when the robots’ appearance had low to moderate human-likeness. Unique combinations of appearance dimensions of human-like robots may be responsible for the presence of an additional valley for robots that only moderately resemble humans. These findings of uncanny valleys in the existing robots may have important implications for robot design.
{"title":"A Comprehensive Approach to Validating the Uncanny Valley using the Anthropomorphic RoBOT (ABOT) Database","authors":"Boyoung Kim, Micala Bruce, LeSean Brown, E. D. Visser, Elizabeth Phillips","doi":"10.1109/SIEDS49339.2020.9106675","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106675","url":null,"abstract":"The uncanny valley hypothesis posits that people’s emotional responses to robots are increasingly positive as robots’ resemblance to humans increases. However, when robots closely, but imperfectly resemble humans, people’s responses turn negative, only to revert back once their appearance more closely resembles humans. These sharp emotional transitions (i.e., peaks and valleys in emotional response) from positive to negative, and then back to positive, are collectively referred to as the uncanny valley. In this project, we attempted to validate the uncanny valley with the largest set of real-world robots currently available in open source format (the ABOT Database). Participants saw static images of 251 robots which varied in their degree of human-likeness, and rated them on uncanniness. We found significant empirical support not only for the hypothesized uncanny valley but an additional valley. This unanticipated valley emerged when the robots’ appearance had low to moderate human-likeness. Unique combinations of appearance dimensions of human-like robots may be responsible for the presence of an additional valley for robots that only moderately resemble humans. These findings of uncanny valleys in the existing robots may have important implications for robot design.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129653170","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106688
E. Campbell, Emma Chamberlayne, Julie Gawrylowicz, C. Hood, Allison Hudak, Matthew Orlowsky, E. Rivero, M. Porter
With millions of vehicles on the road each day, traffic delays and interstate congestion result in loss of productivity and millions of dollars each year. A majority of these traffic delays are caused by traffic incidents including crashes and disabled vehicles. These incidents are safety hazards and can lead to secondary crashes. Rapid clearance of these events and scene management during an incident can significantly reduce the impact of congestion. To combat hazardous conditions and decrease congestion related delays, the Virginia Department of Transportation (VDOT) has a fleet of Safety Service Patrols (SSP) that monitor highway conditions and assist emergency responders in scene clearance and traffic management. Managers of the SSP program seek to schedule patrollers in a manner that optimizes their influence on safety and congestion. This paper proposes a Genetic Algorithm based route scheduling algorithm that assigns SSP routes with the goal of minimizing the total time vehicles are stranded before an SSP vehicle arrives. The algorithm adapts to different incident rates and response times to produce schedules that vary by time-of-day and day-of-week. To examine the performance of the algorithm, optimal schedules were made for I- 95 in Virginia. A regression model was also developed to estimate the incident rates using a combination of daily traffic counts and historic rates that accounts for the under-counting of incidents in non-patrolled regions. Another model was used to estimate the SSP response times that resolves the inconsistencies with historical response times for incidents that occurred outside of the patrolled roadways. The results indicate that new route schedules based on the day-of-week could lead to a reduction in total time waiting for SSP assistance by an average of 13%, helping VDOT maintain safety, increase impact, and Keep Virginia Moving.
{"title":"Optimization of VDOT Safety Service Patrols to Improve VDOT Response to Incidents","authors":"E. Campbell, Emma Chamberlayne, Julie Gawrylowicz, C. Hood, Allison Hudak, Matthew Orlowsky, E. Rivero, M. Porter","doi":"10.1109/SIEDS49339.2020.9106688","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106688","url":null,"abstract":"With millions of vehicles on the road each day, traffic delays and interstate congestion result in loss of productivity and millions of dollars each year. A majority of these traffic delays are caused by traffic incidents including crashes and disabled vehicles. These incidents are safety hazards and can lead to secondary crashes. Rapid clearance of these events and scene management during an incident can significantly reduce the impact of congestion. To combat hazardous conditions and decrease congestion related delays, the Virginia Department of Transportation (VDOT) has a fleet of Safety Service Patrols (SSP) that monitor highway conditions and assist emergency responders in scene clearance and traffic management. Managers of the SSP program seek to schedule patrollers in a manner that optimizes their influence on safety and congestion. This paper proposes a Genetic Algorithm based route scheduling algorithm that assigns SSP routes with the goal of minimizing the total time vehicles are stranded before an SSP vehicle arrives. The algorithm adapts to different incident rates and response times to produce schedules that vary by time-of-day and day-of-week. To examine the performance of the algorithm, optimal schedules were made for I- 95 in Virginia. A regression model was also developed to estimate the incident rates using a combination of daily traffic counts and historic rates that accounts for the under-counting of incidents in non-patrolled regions. Another model was used to estimate the SSP response times that resolves the inconsistencies with historical response times for incidents that occurred outside of the patrolled roadways. The results indicate that new route schedules based on the day-of-week could lead to a reduction in total time waiting for SSP assistance by an average of 13%, helping VDOT maintain safety, increase impact, and Keep Virginia Moving.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129947336","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 : 2020-04-01DOI: 10.1109/SIEDS49339.2020.9106663
A. Ecelbarger, P. Hamlin, S. McGrath, K. I. Nwanevu, N. Smith, A. E. Stavrinaky, D. Xu, G. J. Gerling, K. Horinek, P. McDermott
The government acquisition process requires a significant amount of research and planning due to its inherent complexities and interdependencies. In particular, in creating a request for proposals (RFP), contract specialists must manage a multitude of tasks and deadlines. The current tools fail to appropriately support their workflow. To create a tool to help better synchronize project planning, we followed an iterative process in designing a novel user experience for use on mobile devices. The design incorporates the three primary phases in generating an RFP, i.e., market research, requirements development, and acquisition strategy and planning. The final design supports a) the retrospective review of project status at high- and low-levels of detail, b) the promotion of personal achievement through goal setting, c) a high level of customizability with numerous filtering options, and d) gamification to engage and guide users. Data visualization indicators were devised to distinguish the completion status of tasks, person-specific goals, interdependencies between actions, and the task completion timeline. Prototype usability walkthroughs with contract specialists evaluated the effectiveness of these design elements.
{"title":"User Experience Design to Synchronize Government Acquisition Strategy and Schedule","authors":"A. Ecelbarger, P. Hamlin, S. McGrath, K. I. Nwanevu, N. Smith, A. E. Stavrinaky, D. Xu, G. J. Gerling, K. Horinek, P. McDermott","doi":"10.1109/SIEDS49339.2020.9106663","DOIUrl":"https://doi.org/10.1109/SIEDS49339.2020.9106663","url":null,"abstract":"The government acquisition process requires a significant amount of research and planning due to its inherent complexities and interdependencies. In particular, in creating a request for proposals (RFP), contract specialists must manage a multitude of tasks and deadlines. The current tools fail to appropriately support their workflow. To create a tool to help better synchronize project planning, we followed an iterative process in designing a novel user experience for use on mobile devices. The design incorporates the three primary phases in generating an RFP, i.e., market research, requirements development, and acquisition strategy and planning. The final design supports a) the retrospective review of project status at high- and low-levels of detail, b) the promotion of personal achievement through goal setting, c) a high level of customizability with numerous filtering options, and d) gamification to engage and guide users. Data visualization indicators were devised to distinguish the completion status of tasks, person-specific goals, interdependencies between actions, and the task completion timeline. Prototype usability walkthroughs with contract specialists evaluated the effectiveness of these design elements.","PeriodicalId":331495,"journal":{"name":"2020 Systems and Information Engineering Design Symposium (SIEDS)","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2020-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131090008","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}