Data-driven Spatial Branch-and-bound Algorithm for Box-constrained Simulation-based Optimization

Archive 2020

In this paper, we present a novel approach that uses convex underestimators of data and a branch-and-bound procedure to obtain globally optimal solutions of simulation-based problems.

Managing Uncertainty in Data-Driven Simulation-Based Optimization

Computers & Chemical Engineering 2019

Gordon Hullen Jianyuan Zhai Sun Hye Kim Anshuman Sinha Matthew Realff Fani Boukouvala

Optimization using data from complex simulations has become an attractive decision-making option, due to ability to embed high-fidelity, non-linear understanding of processes within the search for optimal values. Due to lack of tractable algebraic equations, the link between simulations and…

Machine learning-based surrogate modeling for data-driven optimization: a comparison of subset selection for regression techniques

Optimization Letters 2019

Optimization of simulation-based or data-driven systems is a challenging task, which has attracted significant attention in the recent literature. A very efficient approach for optimizing systems without analytical expressions is through fitting surrogate models. Due to their increased flexibility, nonlinear…

Nonlinear Variable Selection Algorithms for Surrogate Modeling

AIChE Journal 2019

Jianyuan Zhai Fani Boukouvala

Having the ability to analyze, simulate and optimize complex systems is becoming more important in all engineering disciplines. Decision-making using complex systems usually leads to nonlinear optimization problems, which rely on computationally expensive simulations. Therefore, it is often challenging to…

Optimization of black-box problems using Smolyak grids and polynomial approximations

Journal of Global Optimization 2018

C.A. Kieslich F. Boukouvala C.A. Floudas

A surrogate-based optimization method is presented, which aims to locate the global optimum of box-constrained problems using input–output data. The method starts with a global search of the n-dimensional space, using a Smolyak (Sparse) grid which is constructed using Chebyshev extrema…

ARGONAUT: AlgoRithms for Global Optimization of coNstrAined grey-box compUTational problems

Optimization Letters 2016

F. Boukouvala C.A. Floudas

The algorithmic framework ARGONAUT is presented for the global optimization of general constrained grey-box problems. ARGONAUT incorporates variable selection, bounds tightening and constrained sampling techniques, in order to develop accurate surrogate representations of unknown equations, which are globally optimized. ARGONAUT…

Global Optimization Advances in Mixed-Integer Nonlinear Programming, MINLP, and Constrained Derivative-Free Optimization, CDFO

European Journal of Operational Research 2015

F. Boukouvala R. Misener C.A. Floudas

This manuscript reviews recent advances in deterministic global optimization for Mixed-Integer Nonlinear Programming (MINLP), as well as Constrained Derivative-Free Optimization (CDFO). This work provides a comprehensive and detailed literature review in terms of significant theoretical contributions, algorithmic developments, software implementations…