Explore ARCExplore ARC

Xun Huan

By |

Xun (Ryan) Huan is an Assistant Professor in the Department of Mechanical Engineering. His research broadly revolves around uncertainty quantification, data-driven modeling, and numerical optimization. His expertise focuses on bridging models and data: optimal experimental design, Bayesian methods for statistical inference, uncertainty propagation in high-dimensional settings, and methods that are robust to model misspecification. He seeks to develop efficient computational methods that integrate realistic models with big data, and combine uncertainty quantification with machine learning to enable robust prediction, design, and decision-making. He is interested in collaborative opportunities in various applications that can benefit from a better understanding of uncertainty and modeling. Current research activities include assessing uncertainty in deep neural networks, and developing sequential experimental design methods for improving autonomy.

Optimal experimental design seeks to identify experiments that produce the most valuable data, and can lead to substantial resource savings. For example, in the design of a shock-tube combustion experiment, design condition A maximizes the expected information gain. When Bayesian inference is performed on data from this experiment, we obtain “tighter” posteriors (with less uncertainty) compared to those obtained from suboptimal design conditions B and C.

Brian Umberger

By |

Brian Umberger is a professor of Movement Science in the School of Kinesiology. Professor Umberger’s research is focused on the biomechanics, energetics, and control of locomotion in humans and other bipeds. A major emphasis of his group is developing computational models of muscle and the musculoskeletal system, and using these models to study bipedal locomotion. Applications have ranged from fundamental studies of locomotion energetics, to the restoration of mobility in gait disorders, and the evolutionary basis for human bipedalism. His research often involves solving large-scale optimal control problems, which present a number of computational challenges. Past work has focused on topics such as parallel global optimization and efficient numerical evaluation of large, sparse Jacobian matrices. Current interests include bi-level and multi-objective optimization approaches, and stochastic methods for evaluating simulation results. The research is often cross-disciplinary in nature, involving teams of scientists, engineers and clinicians.

Musculoskeletal model of a person with lower limb amputation for optimizing prosthesis design

Yafeng Yin

By |

Yafeng Yin is a Professor of Civil and Environmental Engineering. He investigates critical issues associated with the design, operations, regulation, and management of innovative mobility services and systems. The goal is to support them in becoming integral components of transportation systems, improving system connections and integration, yielding efficient and multimodal mobility of people and goods, and enhancing rural underserved communities’ access to employment, education and other lifeline opportunities. He is focus on understanding the interaction between travelers, transportation modes and infrastructure, and then modeling the consequence of the interaction. With the model established, he then investigates how to optimize the design and operations of transportation systems. In his work, he often needs to solve large-scale optimization models.

Raed Al Kontar

By |

Rael Al Kontar is an Assistant Professor in the department of Industrial & Operations Engineering. His research broadly focuses on developing data analytics and decision-making methodologies specifically tailored for Internet of Things (IoT) enabled smart and connected products/systems. He envisions that most (if not all) engineering systems will eventually become connected systems in the future. Therefore, his key focus is on developing next-generation data analytics, machine learning, individualized informatics and graphical and network modeling tools to truly realize the competitive advantages that are promised by smart and connected products/systems.

Towards smart and connected systems

Alex Gorodetsky

By |

Alex Gorodetsky is an Assistant Professor in the Department of Aerospace Engineering. His research includes using applied mathematics and computational science to enhance autonomous decision making under uncertainty. His research has an emphasis on controlling systems that must act in complex environments that are often represented through expensive computational simulations. His research uses tools from wide ranging areas including uncertainty quantification, statistical inference, machine learning, numerical analysis, function approximation, control, and optimization. Several of the key areas he focuses on are: optimal planning by solving large scale Markov decision processes, fast Bayesian estimation for nonlinear dynamical systems, high-dimensional compression and approximation of physical quantities of interest, and fusion of information from varying simulation fidelities and data through multi-fidelity modeling.

Steven Skerlos

By |

Steven Skerlos is an Arthur F. Thurnau Professor of Mechanical Engineering and a Professor of Civil and Environmental Engineering. He is the director of the U-M program in Sustainable Engineering and co-director of the Engineering Sustainable Systems Program. His research focus is on the design of technology systems to reduce environmental impact while advancing economic and societal objectives. His group works on environmental and sustainable technology systems, life cycle product design optimization and sustainable water and wastewater systems, among other topics. From designing humanitarian technologies to purifying water using anaerobic membrane reactors, Prof. Skerlos research addresses challenges in the fields of systems design, technology selection, manufacturing, and water.

Sustainable Technology Policy Maximizing the cost-effectiveness of pollution elimination eastlab.org

Sustainable Technology Policy
Maximizing the cost-effectiveness of pollution elimination (eastlab.org)

Ming Lin

By |

Ming Lin’s research focuses on  high dimensional high order statistics and the related applications in real world machine learning problems. His recent research topics includes symmetric matrix sensing, Positive Unlabeled learning, One-bit Active learning and nonconvex tensor machine.

Brian Denton

By |

Brian Denton is a Professor in the Department of Industrial & Operations Engineering, and a member of the Institute for Healthcare Policy and Innovation. His primary research interests are in optimization under uncertainty with applications to medical decision-making. He uses stochastic programming, simulation-optimization and Markov decision processes to optimize decisions regarding detection, treatment, and prevention of chronic diseases, including cancer, diabetes and heart disease.

Seth Guikema

By |

Seth Guikema is an Associate Professor of Industrial & Operations Engineering and Civil and Environmental Engineering. Prof. Guikema’s research is focused on and grounded in risk analysis, statistical learning theory, Bayesian probability, stochastic simulation, decision analysis, and agent-based modeling. The issues studied are related to climate adaptation and the sustainability of cities and infrastructure, disaster risk analysis, critical infrastructure modeling, natural hazards, and terrorism risk. Current projects include large-scale agent-based simulation models of evolution of regions in response to repeated climate-related events under different policy scenarios, data-driven evaluation of urban renewal and sustainability, and data-driven predictive modeling of the impacts of storms on power systems.

Cong Shi

By |

Cong Shi is an Assistant Professor in the Department of Industrial and Operations Engineering.

Professor Shi’s current research is focused on the design and performance analysis of efficient algorithms for stochastic optimization models, arising in the context of inventory and supply chain management, revenue management, as well as logistics. These stochastic optimization problems involve sequential decision-making under highly evolving or poorly understood environments, which are typically hard to solve to optimality. He constructs efficient heuristics that admit worst-case or average-case performance guarantees, and in doing so he develops novel analytical and computational techniques that are applicable to a broad class of models.