Skip to main content

Open Water Management - from droplets of participation to streams of knowledge

Article Category

Article available in the folowing languages:

Big Data helps households save water

Despite water being a precious resource, its usage is typically communicated every few months. An EU-funded project has developed low-cost sensors and applications, empowering consumers with detailed information and stimuli that help them adopt sustainable water use.

Food and Natural Resources

Saving water is easy through simple actions, like spending less time in the shower or selecting an eco-efficient programme for your dishwasher. The real challenge is how to remind ourselves to save water in the first place and to embrace sustainable habits in our daily lives. This can only be done through exposure to stimuli and information that immediately tells us about our water use and encourages adoption of more environmentally friendly behaviour. The EU-funded DAIAD project addressed these challenges by developing the first integrated demand management system for water for use in both residential homes and utilities. Project coordinator, Mr Spiros Athanasiou explains: ‘DAIAD employs Big Data and machine learning (ML) technologies to exploit data from smart water meters and help consumers change their behaviour to use water more sustainably’. He adds, ‘It also provides novel large-scale analytics to improve short-, medium-, and long-term demand management for water utilities, by extracting the untapped value of smart water meter investments.’ Making the most of data Researchers collected and analysed data from trials conducted for over a year in Alicante, Spain, and St Albans in the United Kingdom, involving 457 consumers in 149 households. The information came from a diverse range of sources, including detailed water consumption data, socio-demographics such as age and income, and weather data like air temperature and precipitation. Results revealed the participants’ water consumption habits and shower behaviour over time, including temperature and energy used. Furthermore, 3 external trials exploiting DAIAD technologies were carried out in Germany, Spain and Netherlands across an additional population of around 4 750 consumers. ‘All real-world experimental data produced during our field trials are provided as open data, in an effort to assist other researchers in replicating our work,’ states Mr Athanasiou. By using Big Data and ML to analyse and predict water consumption, DAIAD can extract actionable insights for consumers, available through their mobile phone, tablet or over the web. The data is retrieved through a smart water meter obtained from the consumer’s water utility and includes personal tips and actions to help people understand their water use. It can even compare consumers’ water usage with similar households or the entire city in which they live. Take an environmentally friendly shower If a consumer does not have a smart water meter installed, they can use the intelligent water meter developed by DAIAD for their shower. The ‘amphiro b1’ is an intelligent monitoring system that does not require a battery, harvesting the energy it needs from the water flow, and informing users of their water and energy usage while they are still in the shower. The device, which is already commercially available, is installed into the showerhead to help users immediately begin saving water, energy and money, and protecting the environment. ‘Amphiro b1 is suitable for all ages, especially for children, as it uses an interactive polar bear image as a symbol for efficiency. The less efficient children are, the less ice the bear has to stand on,’ says Mr Athanasiou. The DAIAD system can therefore directly benefit water utilities and individual consumers both in Europe and throughout the world by providing detailed information about their water usage that is easily interpretable. It will also help to make Europe a global leader in technology for sustainable water use and help protect the environment.


DAIAD, water, demand management system, Big Data, machine learning, amphiro b1

Discover other articles in the same domain of application