Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Cited by Google
Similars in SciELO
Similars in Google
Share
Revista Facultad de Ingeniería Universidad de Antioquia
Print version ISSN 0120-6230On-line version ISSN 2357-53280
Abstract
CAMERO, Andrés; TOUTOUH, Jamal; FERRER, Javier and ALBA, Enrique. Waste generation prediction under uncertainty in smart cities through deep neuroevolution. Rev.fac.ing.univ. Antioquia [online]. 2019, n.93, pp.128-138. ISSN 0120-6230. https://doi.org/10.17533/udea.redin.20190736.
The unsustainable development of countries has created a problem due to the unstoppable waste generation. Moreover, waste collection is carried out following a pre-defined route that does not take into account the actual level of the containers collected. Therefore, optimizing the way the waste is collected presents an interesting opportunity. In this study, we tackle the problem of predicting the waste generation ratio in real-world conditions, i.e., under uncertainty. Particularly, we use a deep neuroevolutionary technique to automatically design a recurrent network that captures the filling level of all waste containers in a city at once, and we study the suitability of our proposal when faced to noisy and faulty data. We validate our proposal using a real-world case study, consisting of more than two hundred waste containers located in a city in Spain, and we compare our results to the state-of-the-art. The results show that our approach exceeds all its competitors and that its accuracy in a real-world scenario, i.e., under uncertain data, is good enough for optimizing the waste collection planning.
Keywords : Deep neuroevolution; deep learning; evolutionary algorithms; smart cities; waste collection.