Download or read book Grasp-based Heuristics for Continuous Global Optimization Problems written by Michael J. Hirsch and published by . This book was released on 2006 with total page pages. Available in PDF, EPUB and Kindle. Book excerpt: ABSTRACT: In almost all areas of the applied sciences, optimization problems abound. An optimization problem can be defined as optimizing a function of several variables subject to some constraints that limit the feasible region. These problems can be defined over discrete or continuous spaces (or some combination thereof). In global optimization, it is reasonable to assume that multiple local optima exist, different from the global optimum. Solution techniques for global optimization problems attempt to overcome locally optimal solutions in the search for a globally optimal solution. The general global optimization problem is known to be NP-hard. Thus, there has been significant research directed towards finding heuristics to solve global optimization problems. When very little is known about the problem structure, i.e., little or no a priori information, the problem can be called a black-box optimization problem. This research introduces a new heuristic for continuous black-box global optimization problems. This heuristic is named C-GRASP, for Continuous Greedy Random Adaptive Search Procedures. In addition to fully detailing this new heuristic, we apply C-GRASP to standard global optimization test problems, as well as several challenging real-world problems.