Xingquan Guan

Xingquan Guan

Lead Data Scientist

ZEST AI



Biography

Xingquan Guan is a Lead Data Scientist at ZEST AI. He has received a Ph.D. in Structural Engineering with a Minor in Statistics and Computer Science at the University of California, Los Angeles. His background broadly aligns with leveraging experimental tests, physics-based simulation, causal inference, and artificial intelligence to evaluate civil infrastructural systems’ resistance, response, and resilience.

Interests
  • Multi-Hazard Risk Assessment
  • AI-Informed Decision Support
  • ML-Powered Design & Analysis
  • Infrastructural Resilience
Education
  • Ph.D. in Structural Engineering (Minor in Statistics and Computer Science), 2021

    University of California, Los Angeles

  • M.Sc. in Earthquake Engineering, 2020

    University of California, Los Angeles

  • M.Sc. in Structural Engineering, 2016

    Huazhong University of Science and Technology

  • B.Sc. in Civil Engineering, 2013

    Huazhong University of Science and Technology

Research Themes

Physics-Based Simulation
Experimental Test
Automation
HPC & Big Data
Artificial Intelligence
Statistical Modeling

Experience

 
 
 
 
 
ZEST AI
Lead Data Scientist
March 2024 – Present Burbank, California, USA
  • Migrated the machine learning modeling infrastructure from the file system to Amazon S3.
  • Established risk profile ranking models to evaluate the financial reliability of credit applicants.
 
 
 
 
 
ZEST AI
Senior Data Scientist
March 2022 – March 2024 Burbank, California, USA
  • Created 100+ AI-powered decision-support systems that have enabled 30+ credit unions to make more informed and efficient credit lending decisions.
  • Contributed to the design and implementation of a modeling platform that automates the data processing, machine learning modeling, and perform analysis process.
  • Optimized the modeling strategy tailored to the unique data characteristics (e.g., data size, duration, and feature distribution) of each credit union.
  • Proposed and implemented a novel modeling strategy by combining foundation model and client-tailored model.
 
 
 
 
 
University of California, Los Angeles
Postdoctoral Scholar
May 2021 – March 2022 Los Angeles, California, USA
  • Performanced probabilistic risk assessment on gas pipeline systems in Southern California, equipping stakeholders the ability to make informed decision on day-to-day maintenance and post-disaster repair activities.
  • Created and implemented a novel collaborative filtering-based collapse fragility assessment approach inspired by recommender systems, which increased evaluation efficiency by 50% while maintaining high accuracy levels.

Projects

*
Probabilistic Risk Assessment of Gas/Water Transmission Systems in Southern California (2021 - 2022)
The gas/water transmission pipeline system underpins a critical service that is foundational to a community’s well-being. Given its overall importance to social and economic development, it is essential to establish a more robust and reliable model that quantifies the seismic vulnerability of the pipeline system.
Performance-Based Analytics-Driven Seismic Design of Steel Moment Frame Buildings (2018 - 2021)
This project aims to propose an AI-informed seismic design method: performance-based analytics-driven seismic design, which is applied to steel moment resisting frame (SMRF) buildings.
Fire Resistance Behavior of Welded Tubular Structures (2013 - 2016)
This project aims to establish a systematic understanding of the fire resistance behavior of steel tubular joints and provide a set of guidelines to inform design.

Recent Publications

Quickly discover relevant content by filtering publications.
If you don’t have access to the listed papers, feel free to request from ResearchGate.
(2022). StEER 2022 Mw 5.6 Indonesia Earthquake Preliminary Virtual Reconnaissance Report (PVRR).

PDF Dataset

(2022). Collaborative filtering-based collapse fragility assessment. In 12th National Conference on Earthquake Engineering.

Cite

(2021). A comparative assessment of mechanistic and data-driven models to estimate building responses. In 17th World Conference on Earthquake Engineering.

Cite

Contact

Feel free to reach out to me if would like to request copies of my published papers or if you would like to discuss any matter further.