CCWI2017: F58 'Water Advisory Demand Evaluation and Resource Toolkit'
journal contributionposted on 01.09.2017 by Daniel Paluszczyszyn, Sunday Iliya, Eric Goodyer, Tomasz Kubrycht
Any type of content formally published in an academic journal, usually following a peer-review process.
The purpose of this feasibility study is to determine if the application of computational intelligence can be used to analyse the apparently unrelated data sources (social media, grid usage, traffic/transportation and weather) to produce credible predictions for water demand. For this purpose the artificial neural networks were employed to demonstrate on datasets localised to Leicester city in United Kingdom that viable predictions can be obtained with use of data derived from the expanding Internet-of-Things ecosystem. The outcomes from the initial study are promising as the water demand can be predicted with accuracy of 0.346 m 3 in terms of root mean square error.