A memetic algorithm for uncertain Capacitated Arc Routing Problems

The Capacitated Arc Routing Problem (CARP) is a widely investigated classic combinatorial optimization problem. Being a deterministic model, it is far away from the real world. A more practical problem model of CARP is the Uncertain CARP (UCARP), with the objective of finding a robust solution which...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	2013 IEEE Workshop on Memetic Computing (MC) s. 72 - 79
Hlavní autoři:	Juan Wang, Ke Tang, Xin Yao
Médium:	Konferenční příspěvek
Jazyk:	angličtina
Vydáno:	IEEE 01.04.2013
Témata:	evolutionary algorithm Maintenance engineering Memetics Optimization robust optimization Robustness Sociology Statistics UCARP Vehicles
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	The Capacitated Arc Routing Problem (CARP) is a widely investigated classic combinatorial optimization problem. Being a deterministic model, it is far away from the real world. A more practical problem model of CARP is the Uncertain CARP (UCARP), with the objective of finding a robust solution which performs well in all possible environments. There exist few algorithms for UCARP in previous work. In this paper, a Memetic Algorithm (MA) and its modified version in time consumption for UCARP are proposed. Experimental results on two benchmark test sets show that with an integrated fitness function and a large step-size local search operator, the new MAs show excellent ability to find robust solutions for UCARP. We also present a less time-consuming version of our MA which shows significant advantages in time consumption.
DOI:	10.1109/MC.2013.6608210