Solving Mathematical Programs

Mathematical program components

A mathematical program consists of

  • a set of unknowns to be determined,

  • a collection of constraints that has to be satisfied, and

  • an (optional) objective function to be optimized.

The aim of a mathematical program is to find a solution with the aid of a solver such that the objective function assumes an optimal (i.e. minimal or maximal) value.

Different types

Depending on the characteristics of the variables and constraints, a mathematical program in AIMMS can be classified as one of the following.

  • If the objective function and all constraints contain only linear expressions (in terms of the variables), and all variables can assume continuous values within their ranges, then the program is a linear program.

  • If some of the variables in a linear program can assume only integer values, then the program is a linear mixed integer program.

  • If the objective is a quadratic function in terms of the variables while the constraints are linear, then the program is a quadratic program.

  • If the objective is neither linear nor quadratic, or some of the constraints contain nonlinear expressions, the program is a nonlinear program.

AIMMS will automatically call the appropriate solver to find an (optimal) solution.

This chapter

This chapter first discusses the declaration of a mathematical program, together with auxiliary functions that you can use to specify its set of variables and constraints. The SOLVE execution statement needed to solve any type of mathematical program is presented, and, finally, AIMMS’ capabilities to help resolve infeasibilities in your model are discussed.