Water Demand Prediction for Housing Apartments Using Time Series Analysis

Water Demand Prediction for Housing Apartments Using Time Series Analysis

Arpit Tripathi, Simran Kaur, Suresh Sankaranarayanan, Lakshmi Kanthan Narayanan, Rijo Jackson Tom
Copyright: © 2019 |Pages: 19
DOI: 10.4018/IJIIT.2019100104
(Individual Articles)
No Current Special Offers


Water management has always been a topic of serious discussion since infrastructure, rural, and industrial development flourished. Due to the depleting water resources, this is now even a bigger challenge. So, here is developed an IoT-based water management system where ultrasonic sensors are employed for predicting the depth of water in the tank and accordingly pumping the water to the sub tank of the apartment. In addition, the time series analysis Auto Regressive Integrative Moving Average (ARIMA) and Least Square Linear Regression (LSLR) algorithms were employed and compared for predicting the water demand for next six months based on the historical water consumption record of the main reservoir/tank. The information on the amount of water consumed from the main reservoir is pushed to the cloud and to the mobile application developed for utilities. The purpose is to access the water consumption pattern and predict water demand for the next six months from the cloud.
Article Preview


Water is one of the most important and essential commodities for everyone which can include humans, animals, plants and so on. Water scarcity is one of the major issues in many societies. In many places, storage tanks are used as intermediary buffers towards storing the water and meeting the peak demand.

Most of apartments, housing units, offices have storage tanks towards mitigating the variability due to lack of 24/7 water supply. The traditional water management system in any society does not guarantee the optimal way to conserve and use minimal water. The most basic and practical technique towards conserving the water is pumping the water into tank. A lot of water is wasted in overloading of main/underground tanks and then filling up the sub tanks. Energy required to pump water is very much demanding towards consumption irrespective of the outcome. So now with the advent of Internet of things, quite amount of work been done which is been discussed below.

Significant amount of research (Amatulla et al., 2017; Chanda et al., 2017) been carried out towards developing an IoT based water monitoring system prototype towards water usage monitoring by employing flow sensor and accordingly controlling the flow usage for curbing the water. This information stored in cloud for further analysis. In addition to above the mentioned research for an apartment/block for water flow monitoring, an IoT based prototype (Sandeep and Hemalatha, 2015) been developed for rural area water tank in specific for predicting the level of water and quality by employing appropriate sensor. The information sent to mobile for decision processing

The challenge in all the above mentioned IoT prototype system is that these systems operate based on sensor value and accordingly controlling the flow of water and so forth. There is no intelligence employed towards analyzing the data by employing machine learning or any other appropriate algorithm for taking intelligent decision towards controlling the flow of water to consumers and so.

So, towards this, quite amount of research carried out by employing Machine learning algorithms which are discussed below.

Significant amount of Research (Candelieria and Archettia, 2014) been carried out towards water demand prediction by employing SVM regression which is a statistical method applied for forecasting the hourly water demand for the entire day. This was validated for water distribution network for Milan.

In addition to the statistical method in machine learning, a neural network algorithm called Kohnen Self organized maps (Chrysi et al., 2015) constructed for predicting the water demand for consumers in Greek island of Skíathos based on water consumption pattern. This analysis is purely based on quarterly water bill only.

Also, some have employed the traditional Artificial Neural Network for predicting the hourly water demand at the demand nodes of water distribution Network (Gwaivangmin and Jiya, 2017) towards monitoring and control of water towards scarcity issue. This has been done for Laminga Water treatment plant and its water distribution network, Jos of Nigeria is taken as a case study. Following with ANN for water demand prediction, Backpropogation Neural Network employed for (Dongjun, and Sungil, 2016) water demand forecasting/prediction of residential building in Korea. The prediction was based on climatic, geometric variables and residential building type as the input for neural model. In addition, average water consumption data set for 2012 to 2014 used as a case study. This BPNN model resulted in proper decision making for residential water management in Korea through optimal water estimation of water consumption. Last but not the least water demand prediction employed multiple kernel regression (Herreraa et al., 2014) which is a statistical method based on daily value of climatic variables which are temperature, wind velocity, millimeters of rain and atmospheric pressure. The accuracy and computational efficiency were computed for making adequate management decisions in the smart cities environment.

The challenge in all these systems is that they have applied either regression or Neural Network for demand forecasting. Also, most of the system have taken quarterly water bill, yearly water consumption data set, or climatic variable integrated for predicting the water demand for consumers. In none of the system, seasonality not taken into consideration for forecasting by applying Time Series model which is very much needed for real time data set like water for consumers.

Complete Article List

Search this Journal:
Volume 20: 1 Issue (2024)
Volume 19: 1 Issue (2023)
Volume 18: 4 Issues (2022): 3 Released, 1 Forthcoming
Volume 17: 4 Issues (2021)
Volume 16: 4 Issues (2020)
Volume 15: 4 Issues (2019)
Volume 14: 4 Issues (2018)
Volume 13: 4 Issues (2017)
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing