Forecasting urban household water demand with statistical and machine learning methods using large space-time data: A Comparative study

Duerr, Isaac; Merrill, Hunter R.; Wang, Chuan; Bai, Ray; Boyer, Mackenzie J.; Dukes, Michael D.; Bliznyuk, Nikolay

doi:10.1016/j.envsoft.2018.01.002

Cited by 54 publications

(30 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Zubaidi et al proposed using singular spectrum analysis (SSA) and a linear autoregressive model [71] for predicting water demand. Furthermore, Duerr et al [66] evaluated the following spatiotemporal statistical models and ML algorithms to forecast monthly water demand: linear and linear mixed models with month effects, a multiple linear regression model, and time-series models (AR(1) and ARIMA, spatiotemporal Gaussian process models, generalized additive models GAM), random forests (RF), Bayesian additive regression trees (BART), and gradient boosting machines (GBM)). The study found that time-series models outperformed other models, indicating the temporal dynamics of water consumption.…”

Section: Hybrid-based Methodsmentioning

confidence: 99%

Machine Learning and Data Analytic Techniques in Digital Water Metering: A Review

Rahim

Nguyen

Stewart

et al. 2020

Water

View full text Add to dashboard Cite

Digital or intelligent water meters are being rolled out globally as a crucial component in improving urban water management. This is because of their ability to frequently send water consumption information electronically and later utilise the information to generate insights or provide feedback to consumers. Recent advances in machine learning (ML) and data analytic (DA) technologies have provided the opportunity to more effectively utilise the vast amount of data generated by these meters. Several studies have been conducted to promote water conservation by analysing the data generated by digital meters and providing feedback to consumers and water utilities. The purpose of this review was to inform scholars and practitioners about the contributions and limitations of ML and DA techniques by critically analysing the relevant literature. We categorised studies into five main themes: (1) water demand forecasting; (2) socioeconomic analysis; (3) behaviour analysis; (4) water event categorisation; and (5) water-use feedback. The review identified significant research gaps in terms of the adoption of advanced ML and DA techniques, which could potentially lead to water savings and more efficient demand management. We concluded that further investigations are required into highly personalised feedback systems, such as recommender systems, to promote water-conscious behaviour. In addition, advanced data management solutions, effective user profiles, and the clustering of consumers based on their profiles require more attention to promote water-conscious behaviours.

show abstract

Section: Hybrid-based Methodsmentioning

confidence: 99%

Machine Learning and Data Analytic Techniques in Digital Water Metering: A Review

Rahim

Nguyen

Stewart

et al. 2020

Water

View full text Add to dashboard Cite

show abstract

“…Water demand modelling that reconstructs detailed household, temporal, and weather variables would enable planners to predict small area demands and test new tariffs (Clarke, 1997). In addition, these variables can enhance the understanding of water use behaviours and thus support improved demand management practices (Duerr, 2018). This is particularly important when the distribution of customer demand is highly skewed, particularly on peak demand days, when a small number of customers are responsible for a high percentage of the total water use.…”

Section: Overview and Aimmentioning

confidence: 99%

“…However, accuracy could be improved if more variables were included in the analysis. Duerr et al (2018) also developed a water demand forecasting model using property (e.g. land and building value, green space), temporal (e.g.…”

Section: Introductionmentioning

confidence: 99%

Short-Term Forecasting of Household Water Demand in the UK Using an Interpretable Machine Learning Approach

Xenochristou

Hutton²,

Hofman

et al. 2021

J. Water Resour. Plann. Manage.

View full text Add to dashboard Cite

This study utilises a rich UK dataset of smart demand metering data, household 10 characteristics, and weather data to develop a demand forecasting methodology that combines 11 the high accuracy of machine learning models with the transparency of regression methods. 12 For this reason, a Random Forest model is used to predict daily demands one day ahead for groups of properties (mean of 3.8 households/group) with homogenous characteristics. A variety of interpretable machine learning techniques (variable permutation, Accumulated Local Effects plots -ALE, Individual Conditional Expectation curves -ICE) are used to quantify the influence of these predictors (temporal, weather, and household characteristics) on water consumption. Results show that when past consumption data are available, they are the most important explanatory factor. However, when they are not, a combination of household and temporal characteristics can be used to produce a credible model with similar forecasting accuracy. Weather input has overall a mild to no effect on the model's output, although this effect can become significant under certain conditions.

show abstract

“…En la Figura 2, se observa que los proyectos trabajan con bases de datos iguales o superiores a un año, once artículos entre uno y diez años y tres estudios con 14, 21 y más de 50 años de información recolectada. El estudio ID [25] que no se muestra en el gráfico, presenta una distribución espacial y no temporal, es decir en un día se tomó una muestra en 48 puntos diferentes a lo largo del rio, lo que evidencia dos tipos de distribuciones para los análisis de calidad de agua en ríos [36].…”

Section: Evaluación Del Desempeñounclassified

Aprendizaje Automático Para La Predicción De Calidad De Agua Potable

Aguilar

Díaz

2020

ingeniare

View full text Add to dashboard Cite

La conservación y cuidado del agua es uno de los problemas medioambientales más importantes en la actualidad. La calidad de agua hace referencia a los valores apropiados de los parámetros fisicoquímicos y/o biológicos del agua para un uso específico y su monitoreo proporciona información útil para ser procesada por herramientas de aprendizaje automático con fines predictivos. Este documento tiene como objetivo presentar una revisión de las técnicas de aprendizaje automático utilizadas en la estimación de la calidad de agua. Los trabajos investigativos muestran que las redes neuronales (RN), sistemas de inferencia neuro difusa (ANFIS), y las máquinas de vectores de soporte (MVS) son las técnicas predictivas más utilizadas, los resultados obtenidos en las medidas de exactitud evidencian la viabilidad de estimar la calidad de agua en ríos, cuencas, lagos entre otros. Estudios comparativos plantean nuevas estrategias para mejorar la exactitud de las predicciones, la cual integra las propiedades de dos o más técnicas para formar modelos híbridos, su implementación con diferentes algoritmos y metodologías puede ser extensa y se perfila como un escenario futuro a desarrollar.

show abstract

Forecasting urban household water demand with statistical and machine learning methods using large space-time data: A Comparative study

Cited by 54 publications

References 24 publications

Machine Learning and Data Analytic Techniques in Digital Water Metering: A Review

Machine Learning and Data Analytic Techniques in Digital Water Metering: A Review

Short-Term Forecasting of Household Water Demand in the UK Using an Interpretable Machine Learning Approach

Aprendizaje Automático Para La Predicción De Calidad De Agua Potable

Contact Info

Product

Resources

About