الفهرس 
ContentsOnly 14 pages are availabe for public view
 |  | from | 89 |  |  |
| | | from | 89 | | |
Abstract
A numerical method for.solving an optimization problem is an iterative method that, starting from an initial guess, generates a sequence of points. Kach point is a better estimate of the solution of the problem than the previous oneThe efficiency of the numerical methods is a very important issue that makes one method preferable than the other. One of the most important criteria in measuring the efficiency of any numerical method is to check its rate of conver¬gence.
Newton’s method is one of the most widely used methods for solving opti-mization problems because of its fast local rate of convergence. In particular. Newton’s method has a q-quadratic rate of convergence.
Interior-point methods are numerical methods that require the starting point
to be feasible with respect to some constraints and enforce the feasibility restric¬
tions through out the generated sequence of points. If the process is terminated
before reaching a solution, which is almost always the case for nonlinear prob-
lems, the terminating point is feasible and probably nearly optimal solution to
the original problem.
lu the Newton-interior point methods we try to combine the Newton’s method with the interior-point idea in such away that gets the benefits of both. In particular we look for a method that has the fast local convergence of Newton’s method and the feasibility restriction of the interior-point method.