AIMMS white papers:
A Nonlinear Presolve Algorithm in AIMMS
This paper describes the AIMMS presolve algorithm for nonlinear problems. This presolve algorithm uses standard techniques like removing singleton rows, deleting fixed variables and redundant constraints, and tightening variable bounds by using linear constraints. Our algorithm also uses the expression tree of nonlinear constraints to tighten variable bounds.
The AIMMS Outer Approximation Algorithm for MINLP (using GMP functionality)
This document describes how to use the GMP variant of the AIMMS Outer Approximation (AOA) algorithm for solving MINLP problems. We show how the AOA algorithm can be combined with the nonlinear presolver and the multi-start algorithm.
Solving convex MINLP problems with AIMMS
This document describes the Quesada and Grossman algorithm that is implemented in AIMMS to solve convex MINLP problems. We benchmark this algorithm against AOA which implements the classic outer approximation algorithm.