Engineering Commons^™

Explaining Deep Q-Learning Experience Replay With Shapley Additive Explanations, Robert S. Sullivan

Dissertations

Reinforcement Learning (RL) has shown promise in optimizing complex control and decision-making processes but Deep Reinforcement Learning (DRL) lacks interpretability, limiting its adoption in regulated sectors like manufacturing, finance, and healthcare. Difficulties arise from DRL’s opaque decision-making, hindering efficiency and resource use, this issue is amplified with every advancement. While many seek to move from Experience Replay to A3C, the latter demands more resources. Despite efforts to improve Experience Replay selection strategies, there is a tendency to keep capacity high. This dissertation investigates training a Deep Convolutional Q-learning agent across 20 Atari games, in solving a control task, physics task, …

The Use Of Data Balancing Algorithms To Correct For The Under-Representation Of Female Patients In A Cardiovascular Dataset, Sian Miller

Dissertations

Given that women are under-represented in medical datasets, and that machine learning classification algorithms are known to exhibit bias towards the majority class, the growing application of machine learning in the medical field risks resulting in worse medical outcomes for female patients. The Heart Failure Prediction (HFP) dataset is a historical dataset used for the training of models for the prediction of heart disease. This dataset contains significantly fewer female patients than male patients, and as such it is expected that models trained using this data will inherit a gender bias to favour male patients. This dissertation explores the use …

Probability Expressions In Ai Decision Support: Impacts On Human+Ai Team Performance, Elias Spinn

Dissertations

AI decision support systems aim to assist people in highly complex and consequential domains to make efficient, effective, and high-quality decisions. AI alone cannot be guaranteed to be correct in these complex decision tasks, and a human is often needed to ensure decision accuracy. The ambition is for these human+ AI teams to perform better together than either would individually. To realise this, decision makers must trust their AI partners appropriately, knowing when to rely on their recommendations and when to be sceptical. However, research has shown that decision makers often either mistrust and underutilise these systems, or trust them …

Modelling And Mitigating Interlocutor Confusion In Situated Human-Avatar And Human-Robot Interaction, Na Li

Dissertations

Human-Robot Interaction (HRI) is an important but challenging field focused on improving the interaction between humans and robots, to make the interaction more intelligent and effective. However, building a natural conversational HRI is an interdisciplinary challenge for scholars, engineers, and designers. Achieving successful conversational interaction with a social robot necessitates not only observing a user’s active participation in the interaction but also being aware of their emotional and attitudinal states as the interaction progresses. On the topic of attitudinal states, one field that has received little attention to date is monitoring the user for possible confusion states. Confusion is a …

Integrating Life-Cycle Analysis Into Civil Infrastructure Resilience Decision Making: Illustrative Application To Seismic Resilience Modeling Of Us Communities, Milad Roohi, Jiate Li, W. Van De Lindt

Durham School of Architectural Engineering and Construction: Faculty Publications

This paper aims to integrate life-cycle analysis into civil infrastructure resilience modeling and decision-making in seismic-prone communities. To achieve this aim, the authors present a methodology for modeling seismic life-cycle resilience of interdependent buildings and lifeline systems and subsequently informing resilience decisions directly related to the maintenance and retrofit of interdependent infrastructure to enhance a community’s physical, social, and economic systems. The methodology consists of 1) community data collection, 2) seismic hazard analysis, 3) physical damage analysis, and 4) system-level functionality and restoration analysis, 5) socio-economic impact analysis, 6) life-cycle optimization and retrofit decisionmaking. The methodology begins by developing geospatial …

Enhancing Student Veterans' Self-Efficacy And Sense Of Belonging In A Targeted Learning Community: Four Years Of Qualitative Results, Anthony W. Dean, Cynthia Tomovic, Vukica Jovanovic, Kim E. Bullington

Engineering Technology Faculty Publications

Eight semesters of qualitative data, collected over four academic years, are presented from a project that resulted in the development of a student professional learning community of high-achieving, low-income engineering and engineering technology student veterans. In the context of this project, student veterans received academic, professional, and financial support that helped them to be successful in school and to prepare them for a career in the STEM workforce. As adult learners, students in this learning community were a vital part of the curriculum development which resulted in increasing the students’ interest and buy-in. Typically, adult learners have lower levels of …

Future Directions Of Space Education, Kimberly T. Luthi Dr., Andy Aldrin, Keith Wilson, Jim P. Solti

International Journal of Aviation, Aeronautics, and Aerospace

The future of space operations graduate education is reliant on industry leaders’ contributions to help forecast the needs of the industry. The aim of the current study is to build consensus on the future direction of the space industry and generate new knowledge on what the industry expects to occur in the future of space studies education. This study documents the responses of 14 industry experts who currently or previously held highly visible senior leadership positions in a company or organization within the government or the commercial space industry and have extensive experience in a variety of management and leadership …

Using Unmanned Aircraft Systems To Investigate The Detectability Of Burmese Pythons In South Florida, Joseph Cerreta, William Austin, David Thirtyacre, Scott S. Burgess, Peter Miller

Journal of Aviation/Aerospace Education & Research

Burmese pythons are an invasive, non-native species of snake to southern Florida and attempts at eradicating the snakes had yielded mixed results. The current rate of detection had been reported as 0.05%. The purpose of this research project was to determine if a UAS equipped with a near-infrared (NIR) camera could be used to detect pythons at a higher rate when compared to a RGB camera. The approach involved collecting 55 images from RGB and NIR cameras, over carcass pythons at flying heights of 3, 6, 9, 12, and 15 meters. A likelihood ratio consisting of a true positive rate …

Safety In Flight Training - An Analysis Of The Ntsb Data 2014-2018, Michael F. Walach

Journal of Aviation/Aerospace Education & Research

There were 7,500 safety events in the NTSB data sets from 2013-2018. These events were analyzed using Chi-square, Cramer’s V, and the odds ratio. Major findings in the study determined that while pilots crash aircraft for the same reasons whether they are in a training environment or not, student pilots are typically less likely to be killed, or seriously injured. The aircraft that student pilots fly however, do not share the same relative safety in some event types. Students destroy and substantially damage more aircraft than their non-training counterparts in abnormal runway contact events. The top five causes of safety …

Risk Assessment Matrix Of Operational Safety (Ramos): Aviation Safety With A Matlab® Design Toolkit, Haoruo Fu, Chien-Tsung Lu, Zhenglei Ji

Journal of Aviation/Aerospace Education & Research

Safety is the priority of the aviation industry that requires continuous support and improvement. While the Safety Management Systems (SMS) is mandatory for the Federal Aviation Administration (FAA) Federal Aviation Regulation (FAR) Part 121 air carriers and Part 139 airports in the United States, SMS remains optional to General Aviation (GA) due to various reasons including limited budget and manpower associated with technologies. This paper aims to promote the adoption of MATLAB^® to develop a low-cost Risk Assessment Matrix of Operational Safety (RAMOS) (risk calculation and control) for GA operators. A case is presented to demonstrate the application of …

Machine Learning Predictions Of Electricity Capacity, Marcus Harris, Elizabeth Kirby, Ameeta Agrawal, Rhitabrat Pokharel, Francis Puyleart, Martin Zwick

Complex Systems Faculty Publications and Presentations

This research applies machine learning methods to build predictive models of Net Load Imbalance for the Resource Sufficiency Flexible Ramping Requirement in the Western Energy Imbalance Market. Several methods are used in this research, including Reconstructability Analysis, developed in the systems community, and more well-known methods such as Bayesian Networks, Support Vector Regression, and Neural Networks. The aims of the research are to identify predictive variables and obtain a new stand-alone model that improves prediction accuracy and reduces the INC (ability to increase generation) and DEC (ability to decrease generation) Resource Sufficiency Requirements for Western Energy Imbalance Market participants. This …

Multivariate Regression And Variance In Concrete Curing Methods: Strength Prediction With Experiments, Haiyan Sally Xie, Sai Ram Gandla, Owen Shi, Pranshoo Solanki

Faculty Publications – Technology

Because concrete strengths and quality are affected by various factors, multivariate regression models are often used to analyze the differences between predicted and target outputs. However, the variableness of a predicted output and how individual input parameters affect prediction reliabilities are still uncertain in practical applications, especially for the prediction of compressive strengths of concrete. This study aims to develop multivariate models for predicting concrete strengths and providing the variance analysis of prediction results by comparisons with experiment outcomes. First, this paper provides an in-depth examination of established variance analysis methods in the context of commonly used multivariate regression models. …

Chronic Anemia Revealing An Idiopathic Watermelon Stomach: Case Report, Hanane Delsa, Imane Rahmouni, Yassamin Benhayoun Sadafyine, Fatima Belabbes, Anass Nadi, Fedoua Rouibaa

Health Sciences

Watermelon stomach, also known as gastric antral vascular ectasia (GAVE) syndrome, is a rare entity. Patients often present with profound unexplained anemia with or without bleeding externalization. We made the diagnosis during a digestive endoscopy for an etiological assessment of this anemia. Endoscopically, antral erythematous lesions in stripes with a punctate appearance, sometimes hemorrhagic, may suggest a watermelon stomach. The pathophysiology of the watermelon stomach remains complex and unclear. Although several pathologies remain implicated, sometimes, it can be idiopathic. The effectiveness of numerous treatments has been evaluated with relatively satisfactory results. This is a rare case of an idiopathic watermelon …

The Strength Of Motivation For Dental Students, Manal Elhijazi, Ihsane Benyahya

Health Sciences

Context: Motivation plays a key role in academic success. It is a dynamic process that maintains perseverance and tenacity in the face of difficulty. The purpose of our study was to measure and compare the strength of dental students' motivation at the beginning and the end of their training Material and methods: an anonymous self-administered voluntary questionnaire was distributed to 1st and 6th year students at the Faculty of Dentistry in Casablanca This questionnaire consisted of 30 closed questions divided into 3 sections, the first of which provided information on socio-demographic data, the second is devoted to measuring the strength …

Harnessing The Potential Of Fibrous Polyester Composites Meant For Bioactive Medical Devices, Graciela Morales, Heriberto Rodríguez-Tobías, Victoria Padilla-Gainza, Karen Lozano, Daniel Grande

Mechanical Engineering Faculty Publications

Fibrous-based composite polyester mats have gained importance in the biomedical area due to their morphological characteristics and mechanical performance. These materials can be modified by incorporating different inorganic micro- or nano-particles, thus leading to promising potential applications as medical devices with antimicrobial and/or bioactive properties. In this regard, this chapter reports a compilation of different studies related to the development of sub-micron fibers based on two important polyesters for biomedical use, namely poly(3-hydroxybutyrate) and poly(lactic acid). Studies focus on fibers developed through hydrodynamic techniques, namely electrospinning, electrospraying, and centrifugal spinning. The incorporation of zinc oxide or hydroxyapatite on the polyester-based …

Peer-To-Peer Energy Trading In Smart Residential Environment With User Behavioral Modeling, Ashutosh Timilsina

Theses and Dissertations--Computer Science

Electric power systems are transforming from a centralized unidirectional market to a decentralized open market. With this shift, the end-users have the possibility to actively participate in local energy exchanges, with or without the involvement of the main grid. Rapidly reducing prices for Renewable Energy Technologies (RETs), supported by their ease of installation and operation, with the facilitation of Electric Vehicles (EV) and Smart Grid (SG) technologies to make bidirectional flow of energy possible, has contributed to this changing landscape in the distribution side of the traditional power grid.

Trading energy among users in a decentralized fashion has been referred …

Quantitative Response Of Metals Post Heat Treatment, Andrew Elloso

Open Access Master's Theses

This studies the ability to continue using metal assemblies after they have been exposed to significant heat. This is 4130 steel and 17-4 PH stainless steel specifically for industry application.

The metals are being examined in multiple functions. They will be compared in two manners. The first is through dimension and strength pre and post heat exposure on sheet specimen. The second is dynamic response through the Split Hopkinson Pressure Bar (SHPB). The bar samples will be examined at room temperature, post heat treatment, and with induction coil heat exposure.

New Host Records Of Apicomplexan Blood Parasites (Haemogregarinidae And Hepatozoidae) Infecting Two Reptiles (Testudines; Ophidia) From Arkansas, C.T. Mcallister, H.W. Robison

Journal of the Arkansas Academy of Science

Relatively few records of apicomplexan blood parasites from reptiles in Arkansas have been published although the effects of these parasites on reptilian health may be of concern. Using photomicrographs we describe the morphotypes of parasite gamonts found in blood samples from the Midland smooth softshell turtle, Apalone mutica mutica, and a western Milksnake Lampropeltis gentilis from Arkansas. The turtle possessed four distinct morphological gamont forms of a Haemogregarina sp. The snake possessed two morphological forms of gamonts of a Hepatozoon sp. Both infections are new host records and the western milksnake has not been described as a host elsewhere for …

Hemoparasites (Apicomplexa: Hepatozoon; Kinetoplastida: Trypanosoma) Of Two Anurans (Hylidae; Ranidae), From Polk County, Arkansas, C.T. Mcallister, H.W. Robison

Journal of the Arkansas Academy of Science

Arkansas supports 26 species/subspecies of anurans and only one (4%), the green frog, Rana clamitans, has been previously reported with hemoparasites. Here, we collected blood samples from three species of anurans, five American green treefrogs, Dryophytes cinereus, five American bullfrogs, Rana catesbeianus, two southern leopard frogs, Rana sphenocephalus utricularius, and two Fowler’s toads, Anaxyrus fowleri from Polk County and examined each for hemoparasites. American green treefrogs and American bullfrogs harbored hemoparasites, including two (40%) D. cinereus and four (80%) R. catesbeianus with trypanosomes, and one (20%) R. catesbeiana with a Hepatozoon sp. This is the first time these two anurans …

Novel Reproductive Data On Blue Sucker, Cycleptus Elongatus (Cypriniformes: Catostomidae), From Northeastern Arkansas, C.T. Mcallister, D.G. Cloutman, E.M. Leis, H.W. Robison

Journal of the Arkansas Academy of Science

Nothing has been published in the scientific literature concerning the reproductive biology of the Blue Sucker, Cycleptus elongatus in Arkansas. We examined seven female C. elongatus collected in late February 2021 and 2022 and again in early March 2023 from the Black River, Lawrence County. Egg mass (g) averaged 15.8% of the total weight of these gravid females. It appears that this sucker can spawn as early as February in this population. This is the first time information on female reproduction in this species has been published from any population of C. elongatus in the state.

Rare Birds In Arkansas: Historical Observations And New Records, R. Tumlison, R. Kannan

Journal of the Arkansas Academy of Science

Diligent members of birding communities are quick to report rare sightings of birds, which often lead to multiple observations and photographic documentation. Verified reports and images of birds in Arkansas are curated by the Arkansas Audubon Society (AAS), which has led to an appreciation of which species are common versus rare. We gathered historic (literature) and more recent unpublished records of rare birds (those with fewer than 10 reports, per AAS) to document the historic and current state of knowledge of those species. Currently, there are a total of 425 species of birds reported in Arkansas, of which 54 are …

Journal Acknowledgements And Editorial Board, Academy Editors

Journal of the Arkansas Academy of Science

No abstract provided.

Table Of Contents, Academy Editors

Journal of the Arkansas Academy of Science

No abstract provided.

Characterizing Variability In The Transient Storage Zones Of Miller Run In Lewisburg, Pa, Sabrina Savidge

Master’s Theses

The transient storage zone processes are investigated in a small second order stream with a 2.2 square kilometer watershed. The presence of transient storage zones in small streams impacts the available flow paths for water and results in a wider range of residence times for water and dissolved chemicals than would be predicted by considering only the main channel flow path. Residence times can be used to quantify the health of a stream as several biogeochemical and ecological processes occur in water slowed by transient storage.

Sections of the studied stream are impacted by varying types of stream restoration practices …

Moving Towards Sustainable Mining Communities: The Case Of The Community-Led Non-Cyanide Non-Mercury Cold Extraction Method (Clinn-Cem) In Compostela Valley, Philippines, Rowee Joy S. Decena

Journal of Public Affairs and Development

The small-scale mining industry and communities are left behind in terms of development. They are confronted with economic, environmental, and social problems. Thus, to address these problems and move towards sustainable development, this social development research explores how CLINN-GEM, as a technological innovation, contributes to achieving sustainable mining communities. The researcher employs a single case study approach and adopts qualitative methods in data gathering and analysis. Data gathering methods include document reviews, guided tour, key informant interviews with seven informants, focus group discussions with the operations group and the members of the project staff, and a structured learning exercise with …

Genetic Algorithm For Solving A Just-In-Time Inventory Model With Imperfect Rework Implemented In A Serial Multi-Echelon System, Hsien-Chung Tsao, Cheng-Chi Chung, Hsuan-Shih Lee, Chih-Ping Lin, Yan-Yun Tu, Ssu-Chi Lin

Journal of Marine Science and Technology–Taiwan

As global industrial competition intensifies, enterprises can achieve substantial competitive advantages in the supply chain management environment by promptly meeting customer demands and efficiently reducing both supply and demand costs. This paper proposes an inventory model for supply chain optimization that considers uncertain delivery lead times and defective products. Solving the model requires solving a nonlinear mixed-integer problem, which traditionally requires considerable time. Solutions to nondeterministic polynomial-time hard problems with high complexity and difficulty are often obtained using heuristic algorithms. Among these algorithms, genetic algorithms have high efficiency and quality. Therefore, we employed a genetic algorithm to solve the proposed …

An Mrf Model-Based Study On The Effect Of Fishes Upon The Characteristics Of Flow Field Of An Aquaculture Tank, Xian-Ying Shi, Zheng-Zheng Huang, Meng Li, Yinxin Zhou, Xiao-Zhong Ren, Hang-Fei Liu, Shu-Peng Du

Journal of Marine Science and Technology–Taiwan

The study aims to investigate the effect of fish movement on the flow field in the aquaculture tank of a recirculating water aquaculture system. Herein, based on the Navier-Stokes equations and the RNG k-ε turbulence model, the flow field in the aquaculture tank with fish movement was numerically simulated using the multiple reference frame (MRF) model and compared with the numerical simulation results of the fishless aquaculture tank. The results revealed that the overall mean flow velocity in the tank decreased significantly when the fish swam counter-currently with a fixed trajectory, and the overall mean flow velocity increased slightly when …

Occurrence Of Predation Of Juvenile Chum Salmon (Oncorhynchus Keta) By Chub Mackerel (Scomber Japonicus) And Spotted Mackerel (Scomber Australasicus) In Miyako Bay, Iwate Prefecture, Japan, Kei Sasaki, Miwa Yatsuya, Daisuke Shimizu, Daisuke Ojima, Shinji Komatsu

Journal of Marine Science and Technology–Taiwan

In recent years, chub mackerel (Scomber japonicus) and spotted mackerel (Scomber australasicus) have started migrating to the coastal areas off Iwate Prefecture earlier than they did previously, resulting in a geographic overlap between hatchery-bred juvenile chum salmon (Oncorhynchus keta) released at sea and the occurrence of predatory mackerel. To clarify whether the mackerels prey on the juvenile chum salmon, we captured adult mackerels by angling during the period when these species overlap in Miyako Bay, Iwate Prefecture. After capture, the stomach contents of the mackerel were examined using a combination of visual examinations and DNA metabarcoding analysis. As a result, …

Numerical Analysis Of Impact Vibration Based On A Bifurcation Diagram With Two Varying Control Parameters, June-Yule Lee

Journal of Marine Science and Technology–Taiwan

This study investigates impact vibration by using a bifurcation diagram with one or two varying control parameters. Sticking, periodic, and chaotic motions are explored using the Poincaré mapping and impact mapping techniques. A full range of bifurcation diagrams with varying driving frequencies is obtained, and the statistical index of the standard deviation (Std) is used to calculate the impact series in the bifurcation diagrams. The results indicate strong agreement between impact conditions and their corresponding statistical index values for the Std. Subsequently, the Std values for two control parameters of (w, m), (w, c), (w, k), (w, f), and (w, …

Efficient Maritime Object Detection And Validation For Enhancing Safety Of Uncrewed Marine Systems, Ahmed Saglam, Yiannis Papelis

VMASC Publications

Safe operation of uncrewed maritime systems is a major concern in the presence of other vehicles or obstacles. Typically, perception algorithms utilize sensor data to identify obstacles that must be avoided, and AI algorithms are used to interpret raw sensor data for use in navigation and object avoidance algorithms. However, perception algorithms are typically computationally expensive. In this paper, we present an efficient method for detecting obstacles using raw lidar data in the form of range or Point Cloud, employing computationally efficient techniques that do not depend on trained models or AI matching. The approach
converts the sensor readings into …