Research On BBO Algorithm And Its Applications In Emergency Transportation Planning

This thesis mainly researchs on Biogeography-Based Optimization(BBO) and it’s applications in Emergency Transportation Planning(ETP). After analyzing ETP problem deeply, typical models in the roadway, railway and air ETP are established respectively, and efficient algorithms based on BBO are developed and applied in several actual ETP problems. Ways to solve ETP problems are being explored. The main contributions of this paper are as follows:1、According to the characteristics and demand of roadway ETP, method, a problem model based on destination sequence coded solution with multiple departure points and multiple destinations is established, an efficient sequence-based migration operator is designed, furthermore, a Seq-BBO algorithm to solving this problem is designed and implemented, and this makes convenient for BBO algorithm to be used to this kind of combinatorial optimization problem. The ETP problem in Yunnan earthquake disaster relief is taken example for, and the advantages of the Seq-BBO algorithm beyond the other popular algorithms is compared and analyzed.2、Facing the characteristics and demand of railway ETP, objective function related to transportation time and material weight is designed, and a nolinear integer planning problem model with complicated constraints is established. Punishment function is used to handle constraints, and transportation capacity utlization rate is used to improve solutions. An efficient problem simplifying and dimension reducing method is proposed. This provides an idea for the model research of complicated FTP. A hyper heuristic algorithm combining with BBO named BBO/HH to solving this problem is developed. Three different kinds of intelligent algorithms are used as the bottom searching methods, whether every component of the solution should be evolved or not is decided according to the BBO migration model, and the hyper heuristic strategy is applied to high-level algorithm space for searching appropriate operator, thus the efficiency of this problem is enhanced remarkably. The advantage of BBO/HH algorithm is verified through the computational experiments on a set of instances about railway ETP and the experimental comparison with several kinds of algorithms.3、According to the characteristics and typical demand of air ETP, a constrained optimization problem model with multiple sources and multiple flight batches is established, and an efficient problem simplifying method is designed. To this problem, a combined algorithm named BHS fusing BBO algorithm and Harmony search algorithm(HS) is developed. The advantage of BHS algorithm is verified through the computational experiments on a set of instances about air ETP and the experimental comparison between several algorithms.