What do we mean by water asset health and is that definition changing?
Per Edoff
Head of Global Energy & Utilities Presales for Data/Analytics,
Asset management and Value Based Maintenance
Building relationship with technology partners is key to addressing this and to the delivery of long-term asset transformation.
What we currently mean by asset health
Asset health is regularly valued in terms of its cost and benefits with respect to service. It’s measured by failure rates or rates of deterioration to forecast future probability of failure.
Currently water companies have key customer experience targets to achieve, measured through the outcome delivery incentive, ODI. These annual performance commitments and outcomes are re-assessed at the end of each AMP. This leads to the situation where short term service is maintained at the same time as underlying asset health is deteriorating. Such a situation is unsustainable in the longer term, ultimately asset failures will overwhelm the ability of the utilities to respond in a timely manner and maintain service.
So what needs to change, we need:
- A greater “enabled focus” on long term investment planning and wider asset renewal evaluated through TOTEX and outcomes management based on whole life cost analysis.
- Good, long duration data on asset health is an important asset in itself; a lead indicator is understanding the state of the asset before failure, this helps avoid significant costs and cliff edge failures.
The maturity of asset data systems and analytics varies across organizations, this includes disconnected data systems and a lack of skills and experience of doing complex data analytics. Building relationship with technology partners is key to addressing this and to the delivery of long-term asset transformation.
Machine leaning-driven process optimisation for wastewater treatment plant.
Recently, a French wastewater treatment plant began a project to optimise its processes from a cost, time, and quality perspective, while avoiding the risk of regulatory penalties that may result from water quality issues. Leveraging plant data and Atos’ big data solution, the plant team developed multiple smart solutions to optimise plant operations, leading up to 20% cost reduction and lower environmental impact driven by water quality improvement.
The output of this project provided algorithms for machine learning and advanced analytics created and packaged centrally in the Atos Codex Data Lake Engine downloaded and executed in real-time at the edge. Combined with clean historical data, third-party external data and batch and real-time IOT data collected at the plant, the algorithms are constantly trained, and models created. Thanks to analytics and Atos Codex Smart Edge, a comprehensive tool for real-time decision making has been achieved. This led to predictive process optimisation on a day-to-day basis and reduced costs for better profits management.
