Five Ways AI can Address the Water Crisis

One of the United Nation’s (UN) Sustainable Development Goals (SDGs) is Goal Six: ensuring availability and sustainable management of water and sanitation for all. The inclusion of the water crisis in the SDGs effectively underlines the importance and consequential nature of water in our everyday lives. As such, it is not surprising that artificial intelligence is being applied to the achievement of this goal.

One of the most predominant concerns around water management across the world is water leakage. According to the United States Environmental Protection Agency (EPA), an average household leaks more than 10,000 gallons of water wasted every year, equivalent to the water needed to wash 270 loads of laundry. And that’s only in the US, imagine countries in the developing world, wherein the quality of the infrastructure in place is inconsistent.

By Daan Mooij

Now, given all these leaks, artificial intelligence can be applied to detect and monitor these leaks through a system. For example, in a city, it is difficult to detect from which houses, which taps, which toilets, which pipes, or water towers the leaks may be coming from. However, the ability to detect these leaks and manage them can be incredibly powerful. WINT, a venture based out of New York, uses AI to detect and stop leaks. Using sensors and a powerful AI model, the system alerts users when water is leaking and automatically shuts off the leak. Systems like this are often applied into houses and buildings.

Machine Learning (ML) coupled with operating real-time sensors in wide networks can go so far as to anticipate when a leak may occur. A system like this involves the use of components such as sensors, network monitoring, cloud storage, and supporting applications. Moreover, systems like this are most aptly applied to water distribution systems, like cities.

By Jos Speetjens

Using systems like this allows cities to have a strong overview of water usage across the city. It allows city officials to expect a more accurate prediction of the scale and complexity of water leakages. This is beneficial for all cities, but especially for older cities where the water distribution system is difficult to upgrade and more prone to water leaks. The accurate prediction of scale and complexity can lead to correct and timely identification of water leakages. And, in turn, this will lower water losses and increase revenue for the city.

Roughly 71% of the world is covered by water, with the ocean covering about 96.5 percent of all Earth’s water. Of all that water, only 3% is fresh, drinkable water. In many countries in the world, households pay for water. A notable exception is Ireland, wherein domestic households are exempted from paying for water. The Irish have historically been incredibly proud of this fact. However, with the worsening of climate change, the government has created a charge for excessive use of water, to be implemented in the near future.

According to the UN, today, 1 in 4 people, or 2 billion people in the world, lack access to safe drinking water. Desalination of water occurs naturally during the water cycle, but given the extreme climates the world has been seeing in recent decades, innovation is required to catch up with the water requirement by 2040. Most desalination plants consume far too much energy and are inefficient.

By Ivan Bandura

In 2021, a study led by Yuyang Wang, Zhonglin Cao, and Amir Barati Farimani looked at a more efficient water desalination process using AI. In essence, desalination is the removal of salt or ions that are not favourable to the human body. In engineering this process, the materials used are pivotal. For example, two-dimensional nanomaterials, like graphene may lower the energy required for desalination if chosen well. Improving this process is where AI steps into the picture. In the study, they proposed a data-driven AI framework for discovering the most efficient graphene nanopore for water desalination.

Asset Management as it is termed and used by utility companies, refers to the maintenance of the infrastructure in place in order to provide consumers with electricity, water, gas, etc. AI coupled with computer vision can help in this arena.

The inspection of urban sewer systems’ pipes is important, given that undetected structural and other faults may result in severe pollution and flooding incidents. Normally, inspection of these drainage systems uses CCTVs, wherein videos are recorded and analysed by a team afterwards to assess the health of the pipes.

By Sigmund

According to the International Water Association, an AI-based solution coupled with computer vision would automate the analysis of the CCTV footage and the detection of faulty pipes. Image processing is followed by a machine learning algorithm, which has been trained with pre-labelled images.

The communication chain between consumers and utility companies will be improved with the integration of AI systems. In many countries in the world, particularly in the developing world, corruption in utility companies is not uncommon. The consequences of such corruption affect everything in that particular society, from citizens’ quality of life to the environment.

By Pavan Trikutam

The Anti-Corruption Resource Centre recently published an article on corruption and water governance in the Mekong River Basin. The river is one of the world’s greatest rivers, accounting for up to 25% of global freshwater. It covers a distance of nearly 5,000 km from its source on the Tibetan Plateau in China to the Mekong Delta. Moreover, it flows through six countries: China, Myanmar, Thailand, Lao PDR, Cambodia, and Vietnam. These governments share the basin and display varying levels of corruption and authoritarianism.

Now, in situations, systems, and governance involved in places like the Mekong River Basin, accountability can be an extremely important and consequential factor. As such, the incorporation of AI that is explainable can make this accountability more achievable.

The ways in which AI can be applied to address the global crisis are abundant. However, if one looks at the start-ups and companies currently at play in this area, it remains the case that a lot of the use cases are similar. On this note, it means that competition for the best execution is at play, which is beneficial for everyone. On the other hand, it is exciting to see further research into novel ways in which AI can be applied. Such is the case with desalination of water – how much further can AI take us?