Abstract
We examine the performance of four discrete differential evolution (DE) algorithms for the solution of capacitated vehicle routing problems (CVRPs). Twenty seven test instances are employed in the experimental analysis, with comparisons of final solution quality and time to convergence. The results indicate that two approaches presented significantly better results, but that all algorithms are still lacking in their ability to converge to the vicinity of the global optimum.
Original language | English |
---|---|
Title of host publication | Proceeding of the Genetic and Evolutionary Computation Conference - GECCO '13 |
Publisher | ACM |
Pages | 77-78 |
Number of pages | 2 |
ISBN (Print) | 9781450319645 |
DOIs | |
Publication status | Published - 2013 |