Genetic-Algorithms-Based Approach for Bilevel Programming Models

摘要: Many decision-making problems in transportation system planning and management can be formulated as bilevel programming models, which are intrinsically nonconvex hence difficult to solve for the global optimum. Therefore, successful implementations of models rely largely on development an efficient algorithm handling realistic complications. In spite various intriguing attempts that were made solving these algorithms unfortunately either incapable finding optimum or very computationally intensive impractical a size. this paper, genetic-algorithms-based (GAB) approach is proposed efficiently models. The performance illustrated compared with previous sensitivity-analysis-based using numerical examples. computation results show GAB much simpler than heuristic algorithms. Furthermore, it believed more likely achieve based globality parallelism genetic