Optimization of bicriterion quasi-concave function subject to linear constraints (Q752008)

The purpose of this paper is to give algorithms for the maximization and minimization of a bicriterion quasi-concave function \(g(c_ 1x,c_ 2x)\) subject to the linear constraints \(Ax=b\), \(x\geq 0\). First, one considers the maximization case. If the function g: \(R^ 2\to R\) is continuously differentiable on \(R^ 2\) and strictly increases in each argument, some preliminary results and an algorithm based on them can be derived. This algorithm is applied to a problem of assigning components optimally to a series-parallel reliability system so as to maximise the system reliability. Finally, one considers the minimization case. Again, some preliminary results and an algorithm based on them is obtained. This algorithm is also applied to the maximization of the reliability of a parallel-series reliability system by optimally allocating the components.