Scott Ferguson

Problem based learning (PBL) has demonstrated promise in providing a range of desirable learning outcomes that are often deficit in engineering undergraduates. However, implementation is challenging for faculty for a variety of reasons. In this research, we will investigate two acute challenges of PBL implementation: generating appropriately ill-defined problems and facilitating students?????????????????? problem solving. Two research questions govern the proposed work: RQ1: How can Jonassen's design theory of problem solving be operationalized to help faculty in developing a range of authentic problem solving opportunities? RQ2: How can Jonassen??????????????????s design theory of problem solving be operationalized to facilitate faculty-student interaction and scaffolding of problem solving? We propose to develop problem solving experiences for an introductory aerospace engineering course at an R1 institution as it transitions to integrate PBL experiences. This transition will be guided by the theoretical foundations of problem typology and problem-solving characteristics as described by Jonassen. Four central research activities are planned: 1) sampling current problems used in introductory aerospace engineering education and mapping them to learning objectives and dimensions of structuredness and complexity; 2) capturing and analyzing the problem generation process for four different problem types at varying levels of difficulty using structuredness and complexity serve as ????????????????controllable factors???????????????; 3) observing and analyzing faculty-student interactions and student problem-solving trajectories with and without an instructional scaffold; and 4) dissemination and discussion of this research with engineering faculty and engineering educators to understand issues impacting extensibility and scalability of findings.

The benefits of rapid build time, geometric control, microstructural control, and novel materials offered by AM processes cannot be fully realized with the existing long certification times associated with the prediction and measurement of defects at different physical scales. The transformative aspects of our proposed research effort are that we will benchmark direct measurement of electron interactions within materials during the operable processing steps of electron beam powder bed fusion (EB-PBF) additive manufacturing (AM) for real-time monitoring, identification, measurement and quantification of AM defects. In short, we have converted our EB-PBF system into a high speed, open architecture, 3D scanning electron microscope (SEM). Because we are depositing material in thin layers, our EB-PBF SEM can, for all intents and purposes, ???????look??????? inside the part as it is being fabricated (with all of the capabilities and limitations expected from an SEM with equivalent beam parameters). We will develop and validate a generalized, EB-PBF machine agnostic, approach to external measurement of the electron beam current, profile, focus and position. These data will be used to calibrate and improve the fidelity of electron data, as well as to identify the detection limits for AM defects, defect morphology and geometric measurement. We will account for the effects of beam jitter, eddy current, magnetic backlash and focus optics variation on the electron signal. We will benchmark the spatial resolution and precision of the electron monitoring across the entire range of spot sizes, beam currents and accelerating potential. we will explore eliminating the need for computationally expensive electron images in favor of structurally terse time-domain data gathered from the electron- material interactions in real time, analyzed using a recurrent neural network with long-short-term memory (LSTM). Ultimately, the goal of this objective is to lay the foundation for a plausible, practical and economically viable approach to real-time defect identification and control in the EB-PBF of metal alloys.

Computer vision and machine learning technology has shown promise as a potential solution for many problems. However, as we rely more heavily on computer vision (CV) and machine learning (ML), the importance of understanding the proper formulation of training data is increasingly important. This project explores fundamental questions about training data construction by understanding how algorithm predictive performance is impacted.

The COVID-19 pandemic has highlighted a lack of protective equipment not only in high-risk hospital settings, but also in non-hospital settings which require close contact with patients such as ophthalmology, optometry, dentistry, and others. The protection factor of N95 masks and plastic face shields are not sufficient in these settings, and a more protective device is required. The goal of this project is to develop a breathable film or fabric that can serve as a platform technology for multiple form factors, such as powered air purifying respirators. This project will focus on improving the protective factor of such devices while keeping costs low enough to support wide deployment in the fields mentioned above. Usability, performance, and cost will be primary factors governing design and manufacturing decisions.

The work will investigate the use of AI for the design and development of padding protection systems. We propose using AI for developing forward models and inverse mapping capabilities.

Accurately mapping wetlands at the road planning stage is critical for timely, cost-effective, and compliant projects. Assessing wetland impacts on proposed road corridors have required trained personnel to identify, measure and evaluate acreage impacts to jurisdictional wetland. NCDOT??????????????????s Wetland Predictive Model (WPM) is a GIS and remote sensing-based tool designed to reduce the need for these costly and time-consuming activities by creating a preliminary map of wetlands without the need for field measurements. To do so, the WPM uses lidar-derived terrain variables such as elevation and slope alongside soil maps and other geospatial data as inputs to a predictive model. WPM inputs such as lidar data are expensive and time-consuming to collect, and so are available only infrequently. Likewise, other geospatial data such as high-resolution satellite imagery and digital soil maps are not always available at the frequency necessary for accurate wetland mapping in potential corridors. Even when up-to-date observations are available, they may lack the spatial or spectral resolution necessary to aid in wetland identification.

This project involves optimization of existing code from Under Armour.

Households in the five counties of the Upper Coastal Plains Council of Government (UCPCOG) spend an average of 35.92% of their total income on energy, compared to 10% for the average US household. Reducing the energy burden of low-income and energy intensive homes is a major challenge for local governments in the UCPCOG region. There are a number of federal, state, local programs that are targeted on improving energy efficiency for low-income households; however, these programs are run independently, leading to little, if any coordination. Our proposed framework allows for integrated planning, combined evaluation of impacts, and tracking of real energy savings from these programs. Our Powering Energy Efficiency & Impacts Framework includes an enterprise geodatabase of all the federal, state, utility, and local programs in the region. The spatial database will allow for a number of data analyses, including segmentation and cluster analysis. The data will be overlayed with various other external data to identify and target low-income households in the region. Our tool will be integrated with Resispeak, a utility energy analysis software, which will help track energy savings associated with the programs. Resispeak will also be used to identify homes with high energy intensity and to provide targeted outreach for applicable energy efficiency programs. This framework will allow UCPCOG and local governments in the region to conduct integrated planning, and informed decision-making, while creating a platform for independently run programs to collaborate, track associated energy savings, and increase program impacts.

The goal of this workshop is to provide PhD students, postdocs, and early career faculty in the engineering design community with a specific forum to learn and discuss best practice when preparing for a NSF CAREER proposal. A second goal of this workshop is to encourage interaction between junior and senior faculty and strengthen the community bonds by providing a forum to discuss career development and planning. Such a workshop is necessary because CAREER proposals are inherently different from general unsolicited proposals to a NSF program, and those preparing to submit a CAREER must consider 1) how their proposed efforts relate to their overall career vision, 2) how such a proposal should be structured, 3) how to make the right connections that support the proposal, and 4) how to incorporate feedback from previous panels. This workshop will be held at the ASME International Design Engineering Technical Conferences & Computers & Information in Engineering Conference (ASME IDETC/CIE) as part of the Design for Manufacturing and the Life Cycle Conference. This workshop will take place on Tuesday, August 23, 2016 from 9:00 am - 4:00 pm at the Charlotte Convention Center in Charlotte, NC and approximately 140 attendees will be expected (subject to room availability).

The research objective of this Faculty Early Career Development (CAREER) proposal is to test the hypothesis that improving the integration of marketing and engineering can lead to customized products better suited to respond to diverse, changing, and unforeseen needs. To achieve this objective, research tasks will advance how engineers leverage heterogeneous consumer preferences, manage system flexibility, and implement reconfigurability. Additionally, a long-term vision is presented that extends the outcomes of this research to significantly advance the area of system-of-systems design. The educational objective of this work is to build on the integration of marketing and engineering to engage in activities that focus on the multidisciplinary, interdisciplinary, and systems-level education of current and future engineers. Realizing this objective involves introducing multidisciplinary design concepts to high school students while simultaneously advancing interdisciplinary and systems-level design education at the collegiate and industrial levels.