The advancements in sensing technology and data acquisition present profound opportunities to collect various types of data from systems. To efficiently utilize data, We will research constructing a unified framework supporting efficient data collection/processing, real-time data analytics, and decision-making processes.

Real-time data collection/processing framework. To efficiently deal with a large volume of streaming data, raw sensor data should be collected and processed effectively in real time. I will research ways to organize data in a structured format, to design and extract features that are relevant to characterizing a target system, and to manage the data using database and cloud computing techniques. Specifically, I will work on the standardization of the data collection and representation so that monitoring data can uncover the mutual interactions among subsystems that comprise a whole system.

Real-time data collection and learning

Real-time adaptive learning algorithms for distributed and networked systems. To efficiently extract knowledge from data, I will explore the theoretical and practical aspects of machine learning algorithms, particularly ones designed for a large-scale, networked system. A majority of engineering systems are often composed of subsystems that are interacting each other to achieve global objectives. Furthermore, their interaction pattern changes depending on environmental contexts. To effectively characterize such networked systems under various contexts, I will investigate how to (1) incorporate the contextual information into the data-driven model, (2) model/extract the interaction patterns (i.e., causal relationships) among the subsystems, and (3) incorporate, simultaneously, the interactions among subsystems and the interaction between the target system and the environmental context.

Decision-making strategies in an uncertain environment. Most engineering systems continuously interact with a stochastic environment. The behavior of such systems in uncertain environments can be learned from the data, and the enhanced understanding can be exploited to make optimum decisions regarding the maintenance and operation of the systems. I will explore various sequential decision-making procedures (such as the bandit problem, the Markov decision process, Reinforcement Learning, and Bayesian Optimization) in which learning about the target systems and optimizing the system response occur simultaneously. I will focus on the conceptualization of such algorithms and the implementation of these algorithms to control physical target systems that interact with stochastic environments.

BO procedure


 By using the developed framework for data collection, learning, and decision-making, I will strive to solve various problems with a goal of making our world more sustainable, reliable, and productive. The target applications of interest are as follows.

Optimum operations for renewable energy systems. The performance of renewable energy systems, such as wind farms and solar plants, strongly depends on their immediate operational strategies under varying environmental conditions. In addition, the performance of renewable energy systems degrades continuously over time. The key question, then, is how to operate and manage such systems to increase the instantaneous energy-producing efficiency and, at the same time, how to maintain that performance level over the long-term. The tasks planned are:

  • For control, we are developing and validating various sequential decision-making algorithms (i.e., data-driven control algorithms) to increase the performance of the wind farm under stochastically varying wind conditions.


  • For optimum maintenance, we are developing a condition-based maintenance scheme that can optimally schedule maintenance based on the measured performance of the system. For example, solar panels’ repair schedule can be determined by investigating the variation in the power production performance under various environmental conditions. I will develop a data-driven diagnostic/prognostic model for solar panels and will validate that through actual implementation.
  • For a long-term goal, we are researching about the optimum operation and management of renewable energy systems that are connected to an energy grid, in which the intermittency of the energy source and the stochastic variation of energy load need to be well-balanced.



Smart manufacturing systems. An enormous amount of data is currently being collected from modern construction and manufacturing processes, providing great opportunities to use the data to better understand these complex processes.

smart manufacturing


  • We employ a data-driven approach to understanding better not only the individual operation but also the complex interactions among multiple operations, in an attempt to achieve an efficient use of resources (i.e., scheduling), effective maintenance (i.e., condition-based monitoring), and process control to improve the quality and productivity of the process.

condition based monitoring

  • We are modeling and analyzing human activities in various industrial environments in order to improve the productivity and safety of human workers. Specifically, I intend to research developing a work environment where human workers can effectively and safely collaborate with intelligent machine tools (i.e., autonomous robots) by integrating sensing, control, and artificial intelligence technologies. 


Smart cities. Population density in urban environments is rapidly increasing and expected to be doubled by 2050. I am highly interested in finding causal relationships among human activities in cities and the performances of various urban systems, such as the traffic network, energy grid, and public transportation, by applying spatiotemporal analysis to urban big data. The main goal is to understand through urban data how urban systems perform under different spatial and temporal contexts while interacting with humans.

smart city

  • Various spatiotemporal analysis tools will be explored and developed, and the cyber-infrastructure for managing contextualized urban sensing data will be developed with extensive collaborations. The enhanced relationship model can then be used to optimally operate the urban systems in the short-term and to make long-term planning and policy decisions.
  • We are highly interested in analyzing the spatiotemporal variations of energy usage by using power meter data and to optimally schedule energy generation and design a (cooperative) load shifting control scheme for the optimum utilizations of the energy source and infrastructures.