A5031811254- Advanced Optimization Algorithms Research
- Advanced Control Systems Optimization
- Process Optimization and Integration
- Advanced Multi-Objective Optimization Algorithms
- Computational Drug Discovery Methods
- Optimization and Variational Analysis
- Reservoir Engineering and Simulation Methods
- Metaheuristic Optimization Algorithms Research
- Scheduling and Optimization Algorithms
- Optimization and Mathematical Programming
- Matrix Theory and Algorithms
- Carbon Dioxide Capture Technologies
- Formal Methods in Verification
- Complexity and Algorithms in Graphs
- Numerical Methods and Algorithms
- Manufacturing Process and Optimization
- Risk and Portfolio Optimization
- Supply Chain and Inventory Management
- Vehicle Routing Optimization Methods
- Protein Structure and Dynamics
- Optimization and Search Problems
- Sparse and Compressive Sensing Techniques
- Fault Detection and Control Systems
- Machine Learning in Materials Science
- CO2 Sequestration and Geologic Interactions
Georgia Institute of Technology
2020-2025
Carnegie Mellon University
2013-2022
Ibero American University
2018
Bayero University Kano
2018
University of Mumbai
2018
United States Department of Energy
2015
National Energy Technology Laboratory
2011-2014
University of Illinois Urbana-Champaign
2001-2010
Inserm
2008
Centre National de la Recherche Scientifique
2008
A central problem in modeling, namely that of learning an algebraic model from data obtained simulations or experiments is addressed. methodology uses a small number to learn models are as accurate and simple possible proposed. The approach begins by building low‐complexity surrogate model. built using best subset technique leverages integer programming formulation allow for the efficient consideration large functional components then improved systematically through use derivative‐free...
This work examines applying deep reinforcement learning to a chemical production scheduling process account for uncertainty and achieve online, dynamic scheduling, benchmarks the results with mixed-integer linear programming (MILP) model that schedules each time interval on receding horizon basis. An industrial example is used as case study comparing differing approaches. Results show method outperforms naive MILP approaches competitive shrinking approach in terms of profitability, inventory...
This paper develops a two-stage stochastic programming approach for process planning under uncertainty. We first extend deterministic mixed-integer linear formulation to account the presence of discrete random parameters. Subsequently, we devise decomposition algorithm solution model. The case continuous variables is handled through same algorithmic framework without requiring any priori discretization their probability space. Computational results are presented problems with up 10...
Abstract Motivation: The Basic Local Alignment Search Tool (BLAST) is one of the most widely used bioinformatics tools. widespread impact BLAST reflected in over 53 000 citations that this software has received past two decades, and use word ‘blast’ as a verb referring to biological sequence comparison. Any improvement execution speed would be great importance practice bioinformatics, facilitate coping with ever increasing sizes biomolecular databases. Results: Using general-purpose graphics...
The need to model uncertainty in process design and operations has long been recognized. A frequently taken approach, the two-stage paradigm, involves partitioning problem variables into two stages: those that have be decided before can after uncertain parameters reveal themselves. resulting stochastic optimization models minimize sum of costs first stage expected cost second stage. potential limitation this approach is it does not account for variability second-stage might lead solutions...
In this paper, we present recent developments in the global optimization software BARON to address problems with integer variables. A primary development was addition of mixed-integer linear programming relaxations BARON's portfolio and nonlinear relaxations, aiming improve dual bounds offer good starting points for primal heuristics. Since such necessitate solution NP-hard problems, their introduction a branch-and-bound algorithm raises many practical issues regarding effective...