Eventos Anais de eventos
COBEM 2019
25th International Congress of Mechanical Engineering
Capacitated Implementation of Growing Neural Gas Applied to the Electric Distribution Companies Measuring Logistics in Brazil
Submission Author:
Anselmo Paiva , MA
Co-Authors:
Artur Albuquerque Silva, Italo Silva, robert santos, pedro henrique , Anselmo Paiva, Geraldo Braz junior, Eliana Márcia Garros Monteiro
Presenter: Geraldo Braz junior
doi://10.26678/ABCM.COBEM2019.COB2019-2397
Abstract
The main tasks related to the billing process of an electrical energy distribution company are the measuring of client energy meters and invoice delivery. The energy distribution companies have the job of organising the measuring process by creating groups of consumers. Thus, each employee must have routes to perform the measuring and delivering invoices in each group. To solve this clustering problem, the development of computational methods is important, so that the process can be improved. Therefore, this work proposes Capacitated Growing Neural Gas (C-GNG), a modified Growing Neural Gas using capacitated clustering concepts. Applied to the electric distribution companies context in Brazil. Aiming to obtain a better clustering than the current one, that was manually generated by company experts. The obtained results, demonstrates improvements in terms of cohesion, separation and silhouette using the proposed C-GNG when compared with the manually crafted solution.
Keywords
Capacitated Clustering, Growing Neural Gas, Districting Problem
