Computational Intelligence & Design Informatics Laboratory

Kyoung-Yun Kim, professor of industrial and systems engineering at Wayne State University, was awarded as a Fellow of the Society of Design and Process Science (SDPS), per an announcement at the SDPS 2017 Conference in Birmingham, Alabama. Elected by the organization’s board of directors, SDPS Fellows are members that are responsible for significant engineering achievements in design and process improvement. A nomination for promotion to Fellow must be initiated by a Fellow of any other society or a member of SDPS and supported by at least five additional sponsors. Kim’s research focuses on design science and informatics as well as product life-cycle modeling. His work has generated over $10 million in external funding from several U.S. federal agencies and industry leaders. In 2016, Kim led a team that received $1.7 million from the Digital Manufacturing and Design Innovation Institute (DMDII) for a joint project with Ford Motor Company to develop a reliable Resistance Spot Welding (RSW) weldability prediction tool. Kim is the site director of the NSF Center for e-Design and director of the Computational Intelligence and Design Informatics (CInDI) laboratory at Wayne State. He has published over 40 top journal papers and over 50 conference papers in proceedings and numerous technical reports and presentations.

Read original article

Wayne State University has received a $1.7 million grant from the Digital Manufacturing and Design Innovation Institute (DMDII) — an institute of the National Network of Manufacturing Innovation (NNMI) — for a project that will advance Resistance Spot Welding (RSW) weldability qualification environments. The project, VRWP: Virtually Guided RSW Weldability Prediction, will allow original equipment manufacturers (OEMs) and suppliers to rapidly converge to feasible welded assembly designs during the early stages of new product development. According to Kyoung-Yun Kim, Ph.D., associate professor of industrial and systems engineering and site director of the NSF Center for e-Design at Wayne State University, system integrators and OEMs working with products that have metallic structures currently rely on material suppliers and testing service companies to conduct physical testing of materials for new welded assembly designs. Without timely delivery of test results, the optimal selection of new materials, processes, and related design decisions is hindered. “When new materials or combinations of materials are considered for an assembly, industry often requires new physical tests or numerical simulations such as finite element analysis,” said Kim. “Data-driven weldability prediction will improve product design efficiency, but is underutilized because of existing data inconsistences. Resistance spot-welding processes and parameters are complex due to coating conditions and surface roughness and give rise to significant data inconsistences, a well-known reliability issue.” The Wayne State team is partnering with Ford Motor Company to develop a reliable RSW weldability prediction tool. The end result will be a web-based RSW weldability prediction tool that will improve design and engineering efficiency. “This prediction tool will ultimately improve product quality through the utilization of advanced materials, allow users to rapidly assess weldment feasibility, and reduce the amount of physical testing required for new material candidates,” said Kim. “In addition, communication between OEMs and suppliers will be enhanced because of the standardization of RSW test data and material.” The proof of concept system for the RSW weldability qualification, called Virtually Guided RSW Weldability Prediction (VRWP), will initially be applied for qualification of automotive body structure joining/welding at Ford Motor Company. The co-principal investigators of this project from Wayne State University are Shiyong Lu, Ph.D., associate professor of computer science; Jeremy L. Rickli, assistant professor of industrial and systems engineering; Xin Wu, Ph.D., associate professor of mechanical engineering; and Qingyu Yang, associate professor of industrial and systems engineering. The project number for this grant is 15-07-04

Read original article

ESTECO Academy is proud to co-organize with the CInDI Lab. the workshop on Advanced methodologies for effective data-driven optimization to be held on June 9th 2016 at the Wayne State University. The focus of the workshop is to discuss and explore the use of latest numerical optimization techniques to study, design and optimize product models involving big amount of data. Nowadays, the combination of simulation models able to capture the physical phenomena charachterizing complex beahviors with advanced optimization strategies techniques offers enormous potentials for detailed analysis and virtual prototyping. The workshop will be held at the Computational Intelligence and Design Informatics (CInDI) Laboratory of Wayne State University (Engineering Development Building*) with the support of ESTECO Academy. The schedule of the workshop includes, scholar and technical presentations on various optimization cases, given by Wayne State Professors and Researchers, Prof. Enrico Nobile from the University of Trieste (Italy) and the guest speakers from Ford Reasearch and Advanced Engineering.

Read original article

Funded by the Ministry of Knowledge and Economics (MKE), Korea

Semi-annual Industrial Advisory Board (IAB) Meetings are an opportunity for face-to-face collaborations among e-Design members. Industry members, research faculty, and students review the progress of current projects and the potential of proposed projects; discuss the performance of the Center relating to research areas, test beds, education, and technology transfer; and investigate prospective research topics to address current and future industry needs. In this year, this meeting was hosted by Iowa State University.

Center for e-Design IAB Meetings

Center for e-Design Homepage

A Wayne State University researcher has introduced computer technology that makes it easier for people who need wheelchairs to select one that best suits their needs. In "Remote Decision Support for Wheeled Mobility and Seating Devices," recently published online and set to appear in the June edition of Expert Systems with Applications, Kyoung-Yun Kim, Ph.D., associate professor of industrial and systems engineering in WSU's College of Engineering, introduces a Web-based decision support system for remotely selecting wheelchairs. According to the 2010 U.S. Census, 3.3 million people age 15 and older use wheelchairs; 10 million use walking aids, such as canes, crutches or walkers. Eleven million people age 6 and older need personal assistance with everyday activities, including such tasks as getting around inside the home, taking a bath or shower, preparing meals and performing light housework. Many people with disabilities live outside large metropolitan areas and lack access to experienced clinicians who can help them decide what kind of device is best for them. Such help has become more necessary with changes implemented by the U.S. Centers for Medicare and Medicaid Services in the Healthcare Common Procedures Coding System (HCPCS) for wheeled mobility devices. Those changes included expanding the number of device identification codes from four to 64, making it difficult to understand where a product falls within the new structure. "Disabled patients almost always have a unique situation, so for something that looks like a simple device, making an optimal decision is not that simple," Kim said. "It requires doctors' and clinicians' assessments, as well as those of patients and their families. Combined with testing time, these are significant factors that lead to an increasingly expensive selection process." In a study supported by the National Institute on Disability and Rehabilitation Research of the U.S. Department of Education, Kim's team reviewed current research in telerehabilitation, an emerging field that aims to deliver rehabilitation services over telecommunication networks and the Internet, and complements in-person clinical assessment and therapy in underserved areas. His system improves the selection and evaluation processes by enabling remote assessment of appropriate wheelchair alternatives with advanced queries and selection criteria. It also provides a reusable information repository and enables systematic evaluation. HCPCS coding changes have increased the gap in decision-making abilities of less experienced clinicians in underserved areas and their more experienced peers in larger population centers, Kim said. In an effort to minimize that gap, the teleconsultation model gives the former group access to the latter, ultimately allowing clinicians to make better selections. A study based on the Technology Acceptance Model was conducted with three groups of clinicians: just graduated, moderately experienced and senior level, via the Rehabilitation Engineering Research Center. The model is a formal research tool for evaluating technological support of a given task. As a control, Kim's team tested face-to-face patient-clinician interactions. It also set up remote assessments using webcams so that patients and less experienced clinicians in one location could consult with more expert clinicians in another location. Subjects said the remote wheelchair selection system generally was very user friendly and made it easy to find quality information, but they were neutral on whether they wanted to use it in their clinical decision making. Kim and his team plan to work with other medical facilities, such as U.S. Department of Veterans Affairs hospitals, to encourage wider use of the teleconsultation model with this remote wheelchair selection system. "The goal of this study is to create a portal that gives clinicians easy and timely access to the information they need to make the best decisions for their patients," Kim said. "We also aim to reduce the gaps in knowledge and human bias between experienced and inexperienced clinicians. "We believe these improvements can also reduce the time needed to select a wheeled mobility device and eventually reduce the cost of the process as well."

Read original article

This talk is to introduce Wayne State University’s College of Engineering Digital/SMART Manufacturing Demonstration Center (WSU D/SDC) that offers a state-of-the art, industry partnered manufacturing environment for innovative research projects and to educate the next generation workforces in the coming evolution of manufacturing. Students and researchers alike have the opportunity to interact with real- time manufacturing, inspection, design/engineering data integration via Cisco Systems platforms and fully sensored advanced manufacturing processes. The D/SDC houses a variety of advanced manufacturing equipment/software, connected together with secured systems infrastructure with Cisco. This enables research and education not only on advanced equipment, such as collaborative robots, additive manufacturing, computed tomography (CT) scanning, and automated laser scanning, but also on the data management, storage, infrastructure, and security critical for IoT, Industry 4.0, and SMART Manufacturing. Aforementioned equipment and design software including CAD/CAM, and design optimization (including modeFRONTIER) have plug and play capability into the Cisco System digital thread. With D/SDC facilities, the Wayne State team is realizing a connected Resistance Spot Welding (RSW) weldability certification concept. RSW is one of the critical and common joining methods in sheet metal-based industries (e.g., automobile, electronics, and aircraft manufacturing). Certification of RSW weldability is crucial to validate the quality and safety of final products. However, the current RSW certification process has multiple challenges. The first challenge is that it is difficult to predict the weldability of new (or combination of) materials that are constantly required in order to satisfy new product functionalities. The second challenge is that a significant number of physical tests are required to certify a welding process for the new combination of materials. In the auto industry as an example, one weld design can require 300-600 tests to certify weldability. Thus, feasibility decisions can be delayed by 8-10 weeks. With the aid of modeFRONTIER, a connected platform is under development to integrate RSW data driven prediction systems, physical models, in-situ RSW sensors, and RSW weld quality metrics for real-time learning to make decisions per individual RSW sample. In this talk, use cases showing the multiple criteria optimization to estimate the welding control parameters are presented in the context of the connected weldability certification concept.

Read original article