Pyomo

Pyomo is an open-source optimization modeling language built in Python. It provides a flexible framework for formulating and solving a wide range of mathematical optimization problems using external solvers. In Pyomo, users define models by declaring data structures, variables, an objective, and a set of constraints. Models can be concrete (data embedded in the model) or abstract (data supplied separately), making Pyomo suitable for both single-run and data-driven workflows.

Key components include variables (Var), objectives (Objective), and constraints (Constraint), along with sets and parameters (Set, Param) to represent indices and data. The modeling process is expressed in Python using Pyomo’s object-oriented interface, which supports algebraic expressions and standard mathematical constructs.

Solving a Pyomo model involves selecting a solver with SolverFactory, translating the model into the solver’s compatible formulation, and running the solver to obtain a solution. Pyomo supports a broad range of problem classes, including linear programming (LP), mixed-integer programming (MILP), nonlinear programming (NLP), and mixed-integer nonlinear programming (MINLP). It also provides facilities for stochastic programming (via PySP) and robust optimization, as well as disjunctive programming and other advanced modeling constructs through its extensions.

While Pyomo does not include solvers itself, it interfaces with many external solvers such as CBC, GLPK, CPLEX, Gurobi, Ipopt, and Knitro. The project emphasizes readability, modularity, and extensibility, enabling researchers and practitioners to prototype models, reuse components, and integrate optimization workflows into Python-based data processing and simulation pipelines. Pyomo is widely used in academia and industry for modeling, experimentation, and education.