GAMS

GAMS
Name	GAMS
Developer	GAMS Development Corporation
Initial release	1976
Latest release	2024
Operating system	Microsoft Windows, Linux, macOS
Genre	Mathematical programming, optimization, modeling
License	Proprietary, academic licensing

GAMS GAMS is a high-level modeling system for mathematical optimization designed to represent complex models and connect them to a wide range of solvers. It provides a declarative modeling language and an integrated environment used in academia, industry, and government research for fields such as energy, transportation, finance, and agriculture. The system interfaces with numerous commercial and open-source solvers and is used in projects involving organizations like World Bank, International Monetary Fund, United Nations, and corporations such as Shell, Siemens, BASF.

Overview

GAMS is a domain-specific language and platform that enables modelers to formulate large-scale linear, nonlinear, mixed-integer, and complementarity problems. It was created to facilitate transparent model representation for collaborations among institutions such as Massachusetts Institute of Technology, Princeton University, University of Cambridge, and Stanford University. The environment supports advanced data management and scenario analysis for stakeholders including European Commission, U.S. Department of Energy, International Energy Agency, and consulting firms like McKinsey & Company and Booz Allen Hamilton.

History and Development

Development of the system began in the 1970s at World Bank-affiliated research groups and by the GAMS Development Corporation, with early contributions from researchers associated with Cornell University, University of California, Berkeley, and Argonne National Laboratory. Over successive decades, releases expanded solver interfaces to include engines developed by companies and labs such as IBM (with CPLEX), FICO (with Xpress), Gurobi Optimization (with Gurobi), and open-source projects like COIN-OR and Ipopt. Major milestones involved support for mixed-integer nonlinear programming, stochastic programming features used in collaborations with Los Alamos National Laboratory and Oak Ridge National Laboratory, and incorporation of standards influenced by groups like IEEE and INFORMS.

Language and Modeling Features

The modeling language emphasizes algebraic representation of sets, parameters, variables, equations, and models, facilitating translation of domain models from teams at Harvard University, Yale University, Columbia University, and research centers such as RAND Corporation. It supports constructs for indexing and mapping akin to techniques used in large-scale models developed by National Renewable Energy Laboratory and Lawrence Berkeley National Laboratory. Key features include support for mixed-integer linear programming used by practitioners familiar with Knapsack problem instances, nonlinear programming relevant to models like Rosenbrock function testing, and complementarity formulations linked to equilibrium models studied by researchers at MIT and London School of Economics.

Solvers and Integration

GAMS integrates with a wide portfolio of solvers: commercial solvers such as CPLEX, Gurobi, Xpress, KNITRO, BARON, and CONOPT; and open-source solvers like CBC, GLPK, Ipopt, and Couenne. It provides APIs and links to computational platforms and languages including Python (programming language), MATLAB, R (programming language), and workflow tools used in environments like Jupyter Notebook and Microsoft Excel. Integration supports high-performance computing centers such as National Energy Research Scientific Computing Center and cluster schedulers like SLURM in projects run by institutions including Princeton Plasma Physics Laboratory and Sandia National Laboratories.

Applications and Use Cases

GAMS is employed in energy system modeling for scenarios developed by International Energy Agency and projects at Argonne National Laboratory (e.g., regionally disaggregated energy market models). It underpins transportation planning studies used by agencies like Federal Highway Administration and optimization in supply chain problems tackled by firms such as Procter & Gamble and Walmart. In finance, GAMS models support portfolio optimization efforts associated with researchers at London Business School and Wharton School. Agricultural policy and land-use models using GAMS have been used by Food and Agriculture Organization and research groups at CIMMYT and IFPRI.

Licensing and Distribution

GAMS is distributed by GAMS Development Corporation under proprietary licensing, offering academic, commercial, and government license tiers. Academic licenses are common at universities including University of Oxford, University of Michigan, University of California, Davis, and research institutes such as Potsdam Institute for Climate Impact Research. Commercial licenses provide solver bundling options involving vendors like IBM, Gurobi Optimization, and FICO/Xpress while cloud and enterprise deployments collaborate with providers such as Amazon Web Services, Microsoft Azure, and Google Cloud Platform.

Category:Optimization software