multivariate production systems optimization - Stanford University
multivariate production systems optimization - Stanford University
multivariate production systems optimization - Stanford University
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
Chapter 6<br />
NONLINEAR OPTIMIZATION<br />
There are numerous problems for which people strive to find an optimal solution. These<br />
problems might be minimizing the fuel consumption of an engine, maximizing the<br />
efficiency of a process, or minimizing the volume occupied by a cluster of objects. It is<br />
trivial to optimize a problem that depends on a single variable. It is when several variables<br />
are optimized simultaneously that matters become complicated.<br />
Mathematics can provide a powerful tool to optimize a problem, regardless of the<br />
number of variables involved. Numerical <strong>optimization</strong> is the location of the extrema of a<br />
mathematical model. The mathematical model, referred to as the objective function, accepts<br />
multiple decision variables as input and returns a single value, the objective variable, as<br />
output. Optimization strives to locate the minimum or maximum value of the objective<br />
function.<br />
Nonlinear <strong>optimization</strong> methods based on Newton’s technique locate the extrema by<br />
approximating the objective function with a nonlinear quadratic model. Let the objective<br />
function, F, be a nonlinear function of the vector of decision variables, x.<br />
F = f x (6.1)<br />
Newton’s method approximates the nonlinear objective function, F, with a quadratic<br />
model, Q, which is also a function of the vector of decision variables, x.<br />
The quadratic approximation<br />
Q x ≈ F x (6.2)<br />
Q x = c T x + 1 2 xT G x (6.3)<br />
may be conceptualized, in two dimensions, as a bowl. If the matrix G is negative-definite,<br />
the expression is bounded above and unbounded below (the bowl is upside-down) and will<br />
therefore have a maximum (See Figure 6.1). Conversely, if the matrix G is positive<br />
definite, the expression is bounded below and unbounded above (the bowl is right-side-up)<br />
61