Hi, I'm the Data to Policy Chatbot
BETA: This site is work in progress
INFO
CLOSE
BETA: You are looking at the early access version of the Data to Policy Navigator – Let us know your feedbackto help us improve.
Want to hear from us? Subscribe here
Country
BulgariaRegion
City of Burgas, BulgariaProJECT YEAR
TYPe OF DATA
AI Capability
AI-driven forecasts and analytics
SDG
The issue of leaking water pipes in public infrastructure has escalated in recent years. Aging pipes, which have been in service for decades, are beginning to show signs of wear and tear. Deteriorating pipes leads to major leaks and significant water loss. In some European countries, the situation is particularly dire: the amount of potable water lost due to poor pipe conditions exceeds 40 percent in some areas, and in certain regions, this figure reaches 70 percent. This issue poses a serious challenge to the sustainability of municipal water supply systems.
As part of an initiative by the European Union on AI for public policy, the Burgas municipality in Bulgaria together with water supply and sanitation company EKSO SRL, proposed an AI-driven approach to identifying leaks in water pipes. The process uses vibration sensors along the pipes to detect the unique vibration patterns produced by water leaks. This enabled the city’s water distribution network to be monitored in real time for potential leaks. However, manually analysing the sheer volume of data produced posed a challenge. To address this, AI-based machine learning (ML) models were employed to analyse the data. ML models increased the efficiency of the leak detection process and improved accuracy. An added benefit is that, unlike human operators, ML models can operate around the clock, ensuring that leaks are detected as soon as possible.
The project will have a significant impact locally, providing inhabitants of the greater Burgas area with a more reliable water supply. The municipality of Burgas can also use these AI-based systems when maintaining, servicing and repairing the water infrastructure. This will enhance operational efficiency, reduce leaks and optimize long-term operating costs. The municipality, in partnership with the state-owned Burgas water supply and sanitation operator, hopes to significantly reduce water losses. These losses can have a significant socioeconomic impact, leading to reduced water availability for households, and challenges for irrigation and sanitation. Thanks to this pilot project, the municipality has already benefited from improved water analysis and lifecycle optimization.
Burgas Municipality has been heavily investing in a safe and efficient water supply and improved sanitization infrastructure. Recent initiatives include a loan of 32.4 million euros to the Burgas state operator, which will co-finance investments in the water network. Additionally, over 63 million euros from the European Commission Cohesion Fund will enhance access to drinking water and improve sewage infrastructure for residents in the Burgas district. The aim of these projects is to increase efficiency of the water management infrastructure, cut operating costs and align with national and European Union legislation for drinking water and wastewater and sludge treatment.
The effects of the Burgas pilot are exemplified in two use cases, each with a corresponding pilot policy.
In the first use case, vibration sensors placed along Dr. Nider Street (Burgas) enabled leak detection in near real-time. The sensors also helped to optimize maintenance schedules by drawing on historical data. The project set a precedent for scaling AI-driven solutions across the city's water distribution system, which could potentially extend to other liquid-carrying pipes. The policy for this initiative employed AI and sensor technology to detect leaks in the water pipe infrastructure, with the AI model achieving up to 99.88 percent accuracy. Objectives included improving reaction times to network failures, reducing water losses and optimizing maintenance planning. The key performance indicator was the number of alerts triggered by the AI model indicating leaks. The pilot dashboard displayed sensor data and AI predictions. Future enhancements could target false alarms, improve leak localization and enhanced dashboard functionality so the model can be replicated across different pipe networks.
The Burgas pilots exemplify a forward-thinking approach to policymaking, using ML and user-friendly visualization for better, faster and more accurate data management. This helped the municipality address complex challenges and enhance the resilience and sustainability of the water supply and sanitation system. The two pilot use cases validated an evidence-based approach to tackling urban issues and highlighted how AI can be harnessed for long-term solutions.
This solution piloted within the AI4PublicPolicy project was competitive in terms of cost, continuous monitoring and data accessibility. The project generated several findings: that improvements can be made by adding more sensors to cover a longer distance; that the method could be tested on larger pipe diameters; and that data needs to be better integrated with a local area map. All stakeholders also agreed that a more concentrated distribution of sensors along the network could improve future reliability of the monitoring system by compensating for possible malfunctions of single sensors.
