An algorithm for linearly constrained programs with a partly linear objective function (Q1200847)

The author solves the optimization problem `minimize \((F(x)+\mu^ T y)\)' subject to \(Ax+By=f\) and \(y\geq 0\), where \(F\) is twice continuously differentiable, \(A\) and \(B\) are (generally rectangular) matrices. The method is based on an active set strategy and solving a sequence of auxiliary quadratic programming problems. Calculation of some complex chemical equilibria is presented as an application of the method.