Smart Water Demand Forecasting: Learning from the Data

Maria Xenochristou; Zoran Kapelan; Chris Hutton; Johannes Hofman

doi:10.29007/wkp4

Smart Water Demand Forecasting: Learning from the Data

Maria Xenochristou, Zoran Kapelan, Chris Hutton, Johannes Hofman

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

196 Downloads (Pure)

Abstract

Accurate forecasts of demand are essential for water utilities in order to manage, plan, and optimize the operation of their network. This work aims to develop a new method for short- term water demand forecasting by utilizing a new data-driven approach based on Random Forests, as well as consumption recordings, household, and socio-economic characteristics, and weather data. Initial results, obtained on real-life consumption data from the UK, demonstrate the potential of this method and show the importance of disaggregating consumption when attempting to determine the influence of weather on water demand. In this study, adding weather input to the model achieved improved forecasting accuracy, especially for the aggregation of properties with medium occupancy and affluent residents during summer months.

Original language	English
Title of host publication	HIC 2018. 13th International Conference on Hydroinformatics
Publisher	EasyChair Publications
Pages	2351-2358
Number of pages	8
Volume	3
DOIs	https://doi.org/10.29007/wkp4
Publication status	Published - 20 Sept 2018
Event	International Conference on Hydroinformatics - Palermo, Italy Duration: 1 Jul 2018 → 6 Jul 2018 Conference number: 13 https://www.hic2018.org/

Publication series

Name
Publisher	EasyChair Publications
ISSN (Electronic)	2516-2330

Conference

Conference	International Conference on Hydroinformatics
Abbreviated title	HIC 2018
Country/Territory	Italy
City	Palermo
Period	1/07/18 → 6/07/18
Internet address	https://www.hic2018.org/

Keywords

demand forecasting
machine learning
random forests
water management

Access to Document

10.29007/wkp4Licence: Unspecified

Smart_Water_Demand_Forecasting_Learning_from_the_DataLicence: Unspecified

Cite this

@inproceedings{843617e188144aa99355c8040686c63d,

title = "Smart Water Demand Forecasting: Learning from the Data",

abstract = "Accurate forecasts of demand are essential for water utilities in order to manage, plan, and optimize the operation of their network. This work aims to develop a new method for short- term water demand forecasting by utilizing a new data-driven approach based on Random Forests, as well as consumption recordings, household, and socio-economic characteristics, and weather data. Initial results, obtained on real-life consumption data from the UK, demonstrate the potential of this method and show the importance of disaggregating consumption when attempting to determine the influence of weather on water demand. In this study, adding weather input to the model achieved improved forecasting accuracy, especially for the aggregation of properties with medium occupancy and affluent residents during summer months.",

keywords = "demand forecasting, machine learning, random forests, water management",

author = "Maria Xenochristou and Zoran Kapelan and Chris Hutton and Johannes Hofman",

year = "2018",

month = sep,

day = "20",

doi = "10.29007/wkp4",

language = "English",

volume = "3",

publisher = "EasyChair Publications",

pages = "2351--2358",

booktitle = "HIC 2018. 13th International Conference on Hydroinformatics",

note = "International Conference on Hydroinformatics, HIC 2018 ; Conference date: 01-07-2018 Through 06-07-2018",

url = "https://www.hic2018.org/",

}

TY - GEN

T1 - Smart Water Demand Forecasting: Learning from the Data

AU - Xenochristou, Maria

AU - Kapelan, Zoran

AU - Hutton, Chris

AU - Hofman, Johannes

N1 - Conference code: 13

PY - 2018/9/20

Y1 - 2018/9/20

N2 - Accurate forecasts of demand are essential for water utilities in order to manage, plan, and optimize the operation of their network. This work aims to develop a new method for short- term water demand forecasting by utilizing a new data-driven approach based on Random Forests, as well as consumption recordings, household, and socio-economic characteristics, and weather data. Initial results, obtained on real-life consumption data from the UK, demonstrate the potential of this method and show the importance of disaggregating consumption when attempting to determine the influence of weather on water demand. In this study, adding weather input to the model achieved improved forecasting accuracy, especially for the aggregation of properties with medium occupancy and affluent residents during summer months.

AB - Accurate forecasts of demand are essential for water utilities in order to manage, plan, and optimize the operation of their network. This work aims to develop a new method for short- term water demand forecasting by utilizing a new data-driven approach based on Random Forests, as well as consumption recordings, household, and socio-economic characteristics, and weather data. Initial results, obtained on real-life consumption data from the UK, demonstrate the potential of this method and show the importance of disaggregating consumption when attempting to determine the influence of weather on water demand. In this study, adding weather input to the model achieved improved forecasting accuracy, especially for the aggregation of properties with medium occupancy and affluent residents during summer months.

KW - demand forecasting

KW - machine learning

KW - random forests

KW - water management

U2 - 10.29007/wkp4

DO - 10.29007/wkp4

M3 - Chapter in a published conference proceeding

VL - 3

SP - 2351

EP - 2358

BT - HIC 2018. 13th International Conference on Hydroinformatics

PB - EasyChair Publications

T2 - International Conference on Hydroinformatics

Y2 - 1 July 2018 through 6 July 2018

ER -

Smart Water Demand Forecasting: Learning from the Data

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this