Deployment of the HpO® Artificial Intelligence system to improve the efficiency of the Kampala water supply network



National Water & Sewerage Corporation (NWSC) is the public entity responsible for water services in over 250 municipalities in Uganda. The project covers the capital city of Kampala, which is operated by a sub-entity of NWSC, Kampala Water.
The drinking water network under the study consists of several thousand kilometers of pipes. The ageing of the networks, which were first laid in 1930, is affecting the performance of the drinking water supply.
The rate of unbilled water (NRW) is 40%. The share due to leakage is estimated at over 20%, or 75,000 m³ lost per day. Kampala Water repairs 4000 leaks per month, with no significant improvement. These are only visible or easily identifiable leaks.
The rational and targeted renewal of the network is thus a major challenge for Kampala Water. Altereo has therefore proposed an expert approach to network asset management based on its patented and award-winning HpO® artificial intelligence technology.
The project is funded by the French Ministry of Economy & Finance through the FASEP scheme.


Project overview

The project lasts 18 months and is divided in 2 main missions:

  • Mission A : Understanding and data improvement
    An essential step with the deployment of the HpO Collect® mobile application for field agents.
  • Mission B : Optimised network renewal programme
    Valuation of network failure data by HpO® artificial intelligence to predict the future risk of failure and to target and optimise leakage detection and network replacement.


Description of the mission

Mission A : Understanding & Data improvement (1 year)

Phase 1 : Understanding the context and the organisation (6 months)

The project starts by understanding the context of NWSC/Kampala Water in terms of existing practices, processes, skills, IT tools and data:

  • Understanding focused on the asset management and leakage control team
  • Familiarity with the leak detection and repair unit
  • Analysis of existing software, in particular the Geographic Information System (GIS)
  • Audit of existing data and measurement of discrepancies with the prerequisites for the implementation of HpO® AI

Phase 2 : Improvement of data describing networks and anomalies (6 months)

Existing knowledge about the network and its failures needs to be strengthened before any analysis can be made. It is necessary to have complete knowledge on two aspects:

  • Pipes (diameter, length, materials, installation or laying date)
  • Failures (date, location, type of anomaly, reason)

Data improvement work is done in collaboration with Kampala Water at 3 levels:

1. Field surveys and investigations to recover missing information

2. Deployment of the mobile application HpO® Collect

HpO® Collect has been specially designed to ease the collection of relevant data at the time of leak repair.
In less than a minute, HpO® Collect enables the field agent to collect data about the failures of mains, service connections and fixtures: GPS location, material, diameter, type of leak, cause, photos, etc.
The data capitalised by HpO® Collect is stored in a secure database which will be used for visualisation and simplified analysis.
The richer the history of failures, the better the asset is known and the more it will be possible to optimise the replacement strategy. Once a sufficient amount of failures has been collected, Altereo will be able to implement the rest of the HpO® approach, including the use of artificial intelligence to predict the risk of failures in the future.

3. Deployment of the HpO® Citizen mobile application to the population

Altereo also plans to deploy a public application that will allow the population to instantly report to Kampala Water any leaks visible on the public domain.

4. Support for updating the missing networks in the GIS

The mapping of the pipes in the geographic information system is not exhaustive. Only 3,000 km of pipes have been digitalised to date. Altereo will identify the sectors where the networks are not digitised in the GIS. Altereo will then oversee Kampala Water’s teams to survey the networks in the field and complete the GIS. This process will be continued over time by Kampala Water.

Mission B : Change management and asset replacement programme

Phase 1 : Implementation of a renewal programme on the Kampala network

During the second year of the project, after having upgraded the descriptive data of the drinking water network and built up a exploitable stock of failures, Altereo will be able to implement the rest of the HpO® approach, including the use of artificial intelligence to calculate the risk of failure in the future.

The benefits of this innovative approach are:

  • Optimising leakage search and detection
  • Targeting the pipes to be replaced as a priority
  • Drastically increasing the efficiency of the renewal budget

Kampala Water will therefore be able to significantly optimise its operations and investment programmes based on the results provided by HpO®. These results will be able to be articulated with pipeline replacement projects for which financing has already been secured from the AFD (French Development Bank).

Phase 2 : Capacity building and sustainability of skills

In parallel with the renewal programme, Altereo, in partnership with Watura, will provide Kampala Water and NSWC with an e-learning platform that will enable them to follow the training sessions. The training & capacity building programme in English will include the principles of asset management and water system performance, the importance of the key phases of the project and the use of the HpO® module interfaces left at Kampala Water.