Increasing complexity and interdependency of infrastructure networks has presented a unique challenge for ensuring smooth functioning of these connected networks. At the same time, the growing use of digital technologies in managing current infrastructure systems has created some opportunities for Industrial and Systems Engineering (ISE) professional to develop more innovative tools in building reliable and resilient infrastructure systems. My talk highlights a few of these emerging challenges and opportunities that ISE researchers are attempting to address. One of the focus areas of my research has been assessing risk and reliability of these critical infrastructure networks. Given advancement in digital computing tools, estimating the all-terminal network reliability (ATNR) by using artificial neural networks (ANNs) has emerged as a promissory alternative to classical exact NP-hard algorithms. Approaches based on traditional ANNs have usually considered the network reliability upper bound as part of the inputs, which implies additional time-consuming calculations during both training and testing phases. This work briefly reviews the results of our recent work on advanced neural networks for ATNR, which dispense with upper bound input need and offer improved performance. In first case, we used CNN for all terminal network reliability assessment. Although CNNs have been successful in image classification, a CNN based approach is proposed with an appropriate format to convert the network features such as adjacency matrix and topological attributes to an image-like matrix. In the second example, we bring more complexity in reliability assessment by assuming both links and nodes are imperfect. To manage this complexity, an integration of DNN and Monte Carlo (MC) is proposed to allow accurate ATNR of networks. Finally, since most of these networks start degrading before actual failure occurs, our research proposes a framework to account for degradation. Different from previous works, our proposed framework considers the reliability of links, nodes, and network as functions of time. To this end Bayesian methods (BM) are proposed to estimate reliability of links and nodes as functions of time considering degradation data. Due to the complexity of the all-terminal reliability problem, and to get fast estimations of network reliability, an integration of Monte Carlo (MC) and DNNs is proposed.
Bio:
Om Prakash Yadav is a Professor and Chair of Industrial and Systems Engineering Department at North Carolina A&T State University Greensboro. He also served as Professor and Duin Endowed Fellow in the Department of Industrial and Manufacturing Engineering at North Dakota State University—Fargo. He received his PhD degree in Industrial Engineering from Wayne State University, MS in Industrial Engineering from National Institute of Industrial Engineering Mumbai (India) and BS in Mechanical Engineering from Malaviya National Institute of Technology, Jaipur (India). His research interests include reliability modeling and analysis, risk assessment, design optimization and robust design, complex engineered system modeling, and manufacturing systems analysis. The research work of his group has been published in Reliability Engineering and Systems Safety, Journal of Risk and Reliability, Quality and reliability Engineering International, International Journal of Production Research, Engineering Management Journal, and IEEE Transaction of Systems, Man, and Cybernetics: Systems. He served as Editor-in-Chief and Associate Editor of International Journal of Reliability and Safety and on the editorial board of several international journals. Dr. Yadav is a recipient of 2015, 2018, and 2021 IISE William A.J. Golomski best paper award and 2021 and 2022 SRE Best paper award. He has published over 150 research papers in the area of reliability, risk assessment, design optimization, and operations management. His research grants of over $10.0 Million funded by NSF, DOE, NASA, and various companies over the period of last 10 years. He is currently a member of IISE, ASQ, SRE, and INFORMS. He is a Fellow of the Institute of Industrial and Systems Engineers (IISE).
