A Local Convergence Analysis of Bilevel Decomposition Algorithms
Angel-Victor DeMiguel (avmiguellondon.edu)
Abstract: Decomposition algorithms exploit the structure of large-scale optimization problems by breaking them into a set of smaller subproblems and a coordinating master problem. Cutting-plane methods have been extensively used to decompose convex problems. In this paper, however, we focus on certain nonconvex problems arising in engineering. Engineers have been using bilevel decomposition algorithms to tackle these problems. These algorithms use nonlinear optimization techniques to solve both the master problem and the subproblems. In this paper, we propose two new bilevel decomposition algorithms and show that they converge locally at a superlinear rate. Our computational experiments illustrate the numerical performance of the algorithms.
Keywords: Decomposition algorithms, bilevel programming, nonlinear programming, multidisciplinary design optimization.
Category 1: Applications -- Science and Engineering (Multidisciplinary Design Optimization )
Category 2: Nonlinear Optimization
Citation: London Business School working paper 2001
Entry Submitted: 04/29/2004
Modify/Update this entry
|Visitors||Authors||More about us||Links|
Search, Browse the Repository
Give us feedback
|Optimization Journals, Sites, Societies|