How can big data be used to reduce uncertainty in stormwater modelling?

Abstract : Urban growth is an ongoing trend and one of its direct consequences is the development of buried utility networks. With growing needs among consumers, new networks are being in- stalled and more underground space is being occupied. Locating these networks is becoming a challenging task. Mispositioning of utility networks is an important problem for both indus- trialised and developing countries and will worsen as cities expand and their networks increase in size and complexity (Jamil et al. (2012), Metje et al. (2007)). Over the past century, it was common practice for public service providers to install, operate and repair their networks sepa- rately Rogers et al. (2012). Now local authorities are confronted with the task of combing data produced by different parties, having distinct formats, variable precision and granularity (Chen and Cohn (2011)). Although in certain countries contractors are now obliged by law to position all buried networks within set precision ranges, finding data related to older network branches is a cumbersome task. Once located these data are often unavailable at the desired precision or are prone to errors or omissions. This study is a part of a global project which aims to recreate a storm water and a sewage network in settings where no accurate information regarding the position or characteristics of buried utility networks is available. The methodology consists in detecting objects, such as manhole covers or inlet grates, from areal photographs and very high resolution satellite imagery and use alternative sources of big data in order to extract interesting descriptor about them. The big data is original information scrapped from the internet such as calls for tenders, newspaper articles, consumer complaints etc. Information extracted with text mining techniques such as used in Kergosien et al. (2015) are particularly interesting to confirm or infirm the position of the previously detected manhole covers and inlet grate. This infor- mation is then used to build an attribute table of the underlying pipes. Once located, standard industry recommendations for pipe selection (diameter, slope, depth, junctions, etc.) are used to define a statistically probable network, including uncertainty associated to each characteristic. The final objective of this work will be to carry out hydraulic simulations using a classical mod- elling software and assess the output hydrographs sensitivity to location and descriptor errors.
Type de document :
Communication dans un congrès
Spatial Accuracy, Jul 2016, Montpellier, France. pp.322-329, Proceedings of Spatial Accuracy 2016. 〈http://www.spatial-accuracy.org/〉
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01417491
Contributeur : Carole Delenne <>
Soumis le : jeudi 15 décembre 2016 - 16:29:14
Dernière modification le : lundi 22 octobre 2018 - 09:54:03
Document(s) archivé(s) le : jeudi 16 mars 2017 - 18:49:04

Fichier

Chahinian_etal2016_SA.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-01417491, version 1

Citation

Nanée Chahinian, Anne-Laure Piat-Marchand, Sandra Bringay, Maguelonne Teisseire, Elodie Boulogne, et al.. How can big data be used to reduce uncertainty in stormwater modelling?. Spatial Accuracy, Jul 2016, Montpellier, France. pp.322-329, Proceedings of Spatial Accuracy 2016. 〈http://www.spatial-accuracy.org/〉. 〈hal-01417491〉

Partager

Métriques

Consultations de la notice

811

Téléchargements de fichiers

176