Data Science



Ramin Madarshahian
University of California San Diego

François Hemez
Lawrence Livermore National Laboratory

Mostafa Mirshekari
Stanford University

Austin Downey
University of South Carolina


The Data Science (DS) Technical Division is a new emphasis at the Society for Experimental Mechanics that focuses on the application of data analytics in structural and mechanical engineering. While the interest in data analysis is not novel at SEM, this effort is an attempt to organize it somewhat more formally and provide more benefits to SEM members and conference attendees. Machine learning, deep learning, big data, statistics, and related methods define the analytical toolset to process vast volumes of measurements and predictions, analyze complex phenomena, identify trends and relationships, and guide the development of predictive models through data. Advancements in sensing technologies (high-speed cameras, laser sensors, unmanned aerial vehicles, etc.) increasingly require using data management and big data frameworks. High-performance computing and cloud systems are becoming unavoidable to store, classify, interpret, and visualize these data. Statistical and machine learning methods provide fast, resilient, adaptive, scalable engines for the online monitoring of structures and mechanical systems, and to support decision-making and risk analysis. The technical division’s charter is to promote all-of-the-above data science within SEM and the engineering community at-large, educate, and encourage practitioners and researchers to present their work, share points-of-view, and advance the technology.



  • Machine learning in vibration analysis and engineering design
  • Data-driven structural health monitoring
  • Deep learning-based computer vision for classification and prediction
  • Machine learning-based anomaly detection and time series analysis
  • Big data and sensor fusion in structural health monitoring
  • Metamodeling and surrogate modeling
  • Internet of Things (IoT)
  • Natural language processing (NLP)