Heri Purnawan

Affiliation Department of Data Science Faculty of Mathematics and Natural Sciences Universitas Negeri Surabaya Office Building E2 Campus 1, 1st floor Ketintang street Gayungan 60231, Surabaya, Indonesia heripurnawan(at)unesa(dot)ac(dot)id Links [LinkedIn] [Google Scholar]

I am a lecturer at the Universitas Negeri Surabaya, where I work in the Department of Data Science.
For my biography, see this page and for more about me, please take a look at my CV [updated in November, 2025].
Also feel free to browse through my publications and presentations.

News

Research Interests

My research focuses on data-driven optimization and decision-making for complex dynamical and socio-technical systems under uncertainty. I develop mathematically grounded and computationally efficient methods that integrate learning-based models, uncertainty quantification, and optimization to support robust, fair, and interpretable decisions.

The research is motivated by real-world challenges in engineering systems and data-driven resource allocation, with applications spanning autonomous systems, manufacturing processes, and data-informed public policy, including agricultural subsidy distribution and supply chain management.

• Data-Driven & Physics-Informed Modeling (New Research Branch in Artificial Intelligence)

  Learning interpretable representations of dynamical systems by combining data-driven methods with physical and structural constraints.

  • System identification and time-series modeling
  • Physics-informed and hybrid models
  • Sparse and interpretable modeling (e.g., SINDy)
  • Surrogate models for complex processes

• Uncertainty Quantification & Probabilistic Inference

  Characterizing, propagating, and interpreting uncertainty arising from data, models, and environments.

  • Probabilistic and Bayesian modeling
  • Parameter and model uncertainty
  • Uncertainty propagation and stochastic processes
  • Risk and reliability metrics

• Optimization, Control, and Decision-Making

  Designing optimization and control strategies that leverage learned models and quantified uncertainty for robust and fair decision-making.

  • Deterministic, stochastic, and robust optimization
  • Optimization-based control (e.g., Model Predictive Control)
  • Risk-aware and chance-constrained optimization
  • Multi-objective optimization
  • Decision-support systems for policy and engineering applications

Current Research Focus

Data-driven optimization for fair and robust allocation of subsidized resources under uncertainty, with a focus on agricultural input distribution using real-world data.

A complete list of my publications and preprints can be found here or on my Google Scholar page