SIGMA Research Group emblemSIGMA Lab

About us

About the SIGMA research group

Artificial intelligence for safe and sustainable urban infrastructure

Context and rationale

Rapid urbanisation in Ho Chi Minh City and the Mekong Delta is creating severe geotechnical and infrastructure challenges: widespread land subsidence, settlement of structures on soft soils, slope and riverbank instability, deterioration and overloading of technical infrastructure, and schedule, cost and safety risk in construction investment projects. Survey, monitoring, remote sensing and infrastructure data are growing rapidly but remain fragmented and under-exploited, while conventional analysis methods struggle with multi-source, non-linear and large-scale data.

Research field

Interdisciplinary application of artificial intelligence, machine learning and deep learning to geotechnical engineering, urban infrastructure management, project management and geographic information systems. Machine and deep learning is the core method across every project of the group.

Overall objective

To grow into a strong research group that develops interdisciplinary machine and deep learning methods for identifying, predicting and managing geotechnical and urban infrastructure risk on GIS and BIM foundations, and that delivers internationally competitive models, tools and technology products linked to education and technology transfer for smart, sustainable cities.

Specific objectives

  1. Objective 1

    At least two WoS/Scopus papers per year aligned with the group's research directions

    Target: ≥ 6 papers over 3 years

  2. Objective 2

    Publish at least one monograph, reference book or textbook within three years

    Target: 1 book

  3. Objective 3

    Successfully supervise at least two master's students within three years

    Target: ≥ 2 master's graduates

  4. Objective 4

    Obtain at least one accepted intellectual property application, or successfully transfer one science and technology product

    Target: 1 IP filing or technology transfer

  5. Objective 5

    Build standardised training datasets, ML/DL models and decision support tools for geotechnical and urban infrastructure management

    Target: Dataset, models and DSS

Novelty and interdisciplinarity

The group's distinctive approach uses machine and deep learning as a unifying method with an integrated interdisciplinary perspective: linking below-ground geotechnical data with above-ground infrastructure and project management data on a single GIS and BIM platform; combining physics-based and data-driven models (physics-informed and hybrid ML); and applying geospatial deep learning together with explainable AI.

Three-year roadmap

  1. Year 1 (2026-2027)

    Main activities

    Consolidate the group; build the methodological framework; collect, label, standardise and integrate geotechnical, infrastructure, remote sensing and field imagery data; build the training dataset and baseline ML/DL models.

    Expected outputs

    • ≥ 2 WoS/Scopus papers
    • Training dataset and spatial database v1
    • Baseline ML/DL models
    • Book outline
    • Institutional research project proposal

  2. Year 2 (2027-2028)

    Main activities

    Develop and optimise deep learning models for geotechnical hazard prediction and infrastructure defect detection; integrate models into a WebGIS tool; apply ML to project risk analysis; validate in the field.

    Expected outputs

    • ≥ 2 WoS/Scopus papers
    • Optimised DL models
    • AI-enabled WebGIS tool (pilot)
    • 1 IP or utility solution filing
    • 1 master's graduate

  3. Year 3 (2028-2029)

    Main activities

    Complete the XAI-based decision support system; validate, deploy through MLOps and pilot technology transfer; consolidate results and publish.

    Expected outputs

    • ≥ 2 WoS/Scopus papers
    • 1 published book
    • 1 master's graduate
    • DSS/AI product for pilot transfer and commercialisation

Legal basis and recognition decision

The research group was established and recognised on the basis of the following documents.

  • Law on Science, Technology and Innovation No. 93/2025/QH15
  • Law on Higher Education No. 125/2025/QH15
  • Decree No. 109/2022/ND-CP
  • Politburo Resolution No. 57-NQ/TW dated 22 December 2024
  • Decision No. 329/QD-TDHTPHCM dated 10 March 2026 of Ho Chi Minh City University of Natural Resources and Environment
  • Recognition Decision No. 1053/QD-TDHTPHCM dated 23 July 2026 of Ho Chi Minh City University of Natural Resources and Environment