Flash flood evacuation with discrete particle swarm optimization / Marina Yusoff
During flash flood evacuation processes, the most challenging task is to move people to safer locations. Uneven distributions of transport, untimely assistance and poor coordination at the operation level have always been the major problem in evacuation process. Currently, no proper procedure is ava...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2011
|
| Online Access: | https://ir.uitm.edu.my/id/eprint/41268/1/41268.PDF https://ir.uitm.edu.my/id/eprint/41268/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | During flash flood evacuation processes, the most challenging task is to move people to safer locations. Uneven distributions of transport, untimely assistance and poor coordination at the operation level have always been the major problem in evacuation process. Currently, no proper procedure is available in managing the evacuation vehicle assignment problem (EVAP) and evacuation vehicle routing problem (EVRP). A discrete particle swarm optimization (DPSO) algorithm is proposed to solve the EVRP and the EVRP. Discrete particle position is proposed.to support the implementation of the DPSO known as myVAP-A for the EVAP. Particle positions are initially calculated based on the average passenger capacity of evacuation vehicle. Computational experiments were done with different numbers of PFA using two types of sequences for vehicle capacity: random and sort ascending order. Both of these sequences were tested with inertia weight and constriction factor (CF). Performance of each vehicle allocation was analyzed in four variations namely myVAP-A using inertia weight with random vehicle capacity, myVAP-A using inertia weight with sort ascending order of vehicle capacity, myVAP-A using CF with random vehicle capacity, and myVAP-A using CF with sort ascending of vehicle capacity. Flash flood evacuation datasets from Malaysia were used in the experiment. myVAP-A using inertia weight with random capacity was found to give the best results compared to the other variations of experiment and outperformed a genetic algorithm (GA) with random vehicle capacity and a GA with sort ascending of vehicle capacity in solving the EVAP. |
|---|