Enabling intelligent water asset monitoring with AI for WaterNSW
Overview
Water resource management is inherently complex – requiring a delicate balance between safeguarding natural ecosystems and maintaining operational efficiency. Our client, WaterNSW, a leading Australian bulk water supplier, understood this challenge all too well. With a responsibility to manage vast canal networks while protecting native aquatic species, they identified the growing need for a smarter, more proactive approach to monitoring.
From detecting canal trash blockages that disrupt water flow to tracking the health of their unique eel populations, traditional methods were proving insufficient. Recognizing the potential of cutting-edge technology, our client turned to us to engineer a completely novel solution - an AI-enabled water asset monitoring platform.
This case study explores how HCLTech and WaterNSW collaborated to overcome these challenges. Leveraging our deep expertise in AI-driven engineering, we delivered a robust, scalable solution that is transforming how water assets and ecosystems are managed. As a trusted technology partner, we helped our client enhance operational visibility, protect biodiversity by tracking the health and movement of elusive baby eels and maintain efficient water distribution. Discover how AI innovation is shaping the future of sustainable water management – one intelligent insight at a time.
The challenges
Our client is responsible for managing an extensive network of canals that facilitate water distribution and support vital aquatic ecosystems. Among the most sensitive indicators of environmental health within these systems is the native eel population, whose well-being reflects the broader ecological balance.
However, monitoring these ecosystems and ensuring the smooth operation of the canal infrastructure presented significant challenges.
Their traditional monitoring methods – manual video reviews and routine site visits to camera installations and trash racks – were not only time-consuming, but also resource-intensive. These approaches delay response times and increase operational costs, making it difficult to proactively manage issues such as trash blockages that could obstruct water flow to cities or threaten aquatic life.
Our client sought an innovative solution to address these challenges. They needed an advanced video analytics platform that could automate monitoring efforts, reduce the need for manual intervention and trigger site visits only when truly necessary. However, realizing this vision came with its own set of hurdles, particularly the need to implement a reliable AI-enabled system that could function seamlessly across vast and often remote water networks.
Our approach and solution
The client and the experts from HCLTech AI Engineering discussed and identified three key use cases crucial for their operations - trash rack monitoring, water level monitoring and detecting elvers (baby eels).
We worked to develop an advanced video analytics platform, HCLTech AI Video Management Solution (AIVMS), to process and analyze data from a mesh created by 80 different cameras that were mounted over canals and water catchment areas. This solution enabled the client to significantly lower overhead costs, minimize time spent onsite and reduce the need to send personnel into remote areas or dispatch vehicles to field locations. Here are a few highlights of what we did to create the video analytics platform.
Key highlights
- Created a new C-based custom model with modified IOU loss for the detection of very small objects such as elvers (baby eels) and earthworms. This helped monitor eel populations with AI-based identification and monitoring.
- Used AI-powered geometric and segmentation-based algorithms for the detection of water levels and the height of the pole in water.
- Leveraged the Azure cloud platform to perform data analytics on the live camera feeds and stored videos.
- Developed a highly customizable AI model to detect and analyze the movement of small aquatic wildlife, enabling accurate eel tracking and blockage identification to meet the client’s specific requirements.
- Implemented a scalable microservices architecture integrated with ServiceNow to generate alerts.
Benefits delivered
Improved water asset monitoring
- 100% accuracy when detecting water flow obstruction across all canals.
- Enhanced efficiency and productivity in monitoring operations through HCLTech AIVMS by reducing the reliance on physical site visits.
- The solution provides short video clips, enabling teams to remotely assess the severity of issues before taking action.
Environmental protection and sustainability
- 85 percent accuracy in distinguishing elvers(baby eels) from earthworms through a state-of-the-art AI model for wildlife detection, ensuring reliable, real-time insights to safeguard native species.
- Enabled continuous monitoring of the eel population has significantly strengthened ecosystem protection efforts.
- In addition, the solution contributes to enhanced environmental conservation by reducing the carbon footprint through fewer vehicle trips and minimizing the need for routine physical site visits.
Enhanced cost optimization and decision-making
- 75% cost savings delivered by converting routine site visits into triggered, need-based interventions and optimizing resource allocation.
- Saves substantial effort and time for ichthyologists, allowing them to focus on critical conservation tasks rather than manual monitoring.
- Enhanced data-driven decision-making through continuous trend analysis, empowering the client with actionable insights to proactively manage water assets and ecosystems.
As a trusted AI engineering partner, HCLTech has empowered WaterNSW’s water monitoring and environmental conservation efforts with advanced AI-powered solutions and deep domain expertise – driving innovation and delivering measurable impact.
