Publicação

An exact solution approach for a multi-follower bilevel optimization problem

Ver documento

Detalhes bibliográficos
Resumo:Bilevel problems are hierarchical optimization problems in which the set of feasible solutions is determined by the set of optimal solutions of a parametric problem. These problems model hierarchical decision processes involving a leader that takes his decision by considering the reaction of a follower. In this paper, we address a bilevel optimization problem with multiple followers, and we explore an exact solution approach that is an extension of an algorithm proposed recently for the case where there is only one follower. The algorithm consists in two phases. The first phase is based on dynamic programming rules, while the second consists in solving the reformulated problem that results from the first phase. This reformulated problem is solved using a commercial solver. Computational results are reported at the end of the paper.
Autores principais:Mansi, Raid
Outros Autores:Alves, Cláudio; Carvalho, J. M. Valério de; Hanafi, Said
Assunto:Analytical methods Decision support systems Numerical methods Operations planning
Ano:2011
País:Portugal
Tipo de documento:comunicação em conferência
Tipo de acesso:acesso restrito
Instituição associada:Universidade do Minho
Idioma:inglês
Origem:RepositóriUM - Universidade do Minho
Descrição
Resumo:Bilevel problems are hierarchical optimization problems in which the set of feasible solutions is determined by the set of optimal solutions of a parametric problem. These problems model hierarchical decision processes involving a leader that takes his decision by considering the reaction of a follower. In this paper, we address a bilevel optimization problem with multiple followers, and we explore an exact solution approach that is an extension of an algorithm proposed recently for the case where there is only one follower. The algorithm consists in two phases. The first phase is based on dynamic programming rules, while the second consists in solving the reformulated problem that results from the first phase. This reformulated problem is solved using a commercial solver. Computational results are reported at the end of the paper.