You can’t have failed to escape the recent press surrounding the water industry and leakage (especially here in the UK. It is estimated that 3 billion litres escape through leakage every day in England and Wales (the equivalent of 1,180 Olympic-sized swimming pools). Tracking down three billion litres of lost water – BBC News
In the face of unprecedented temperatures and prolonged drought, a problem that has plagued companies across the industry for years is coming to a head. The halcyon days of plentiful supply masking fundamental flaws in water networks look to be over. Regulators will likely take a dim view of anything but plans to reduce leakage over the next AMP significantly.
There will be no shortage of solutions to this complex challenge underpinned by networks that are under-maintained, the underlying infrastructure that is aging, and countless pressure variants that cascade through unmanaged parts of the network. On the engineering side, innovative solutions are available in the market now, and early-stage research is proving promising in other Active Leakage Control Efficiency in the Sustainable Economic Level of Leakage Calculation (ukwir.org).
On the technology side, solutions abound, and countless case studies enthusiastically tout the effectiveness of this logger, model, or digital twin at creating sustainable reductions in leakage. Unfortunately, the fact remains that leakage is still a problem and no one company seems to be on the fast track to fixing it. We do not portend to have solutions to the underlying issues (unless we are ready to press pause on our lives for a few years while we dig everything up and fix the core problems). However, our experience has led us to a few observations that may be of use to companies looking to make a dent in their leakage over the next AMP:
- There are no silver bullets. Digital twins first came into the lexicon in the early 2000s but has only really become technically possible since the mid 2010s. These give the promise of bridging the IT/OT divide, modell all kinds of operational scenarios to maximise efficiency. Unfortunately, many dreams of a digital twin utopia have been quashed due to beleaguered data quality challenges, data gaps, and operational teams’ poor uptake. Despite advances in processing power and the models that underpin digital twins, “small data” challenges and culture still create headwinds for their success. Rather than buying into the latest sales pitch for a silver bullet technology solution, think about the long-term capability you are building to respond to the problem.
- Build a capability to drive response. If you don’t have it already, build a leakage team. This is not just a team to focus on reactive response; it should be a partnership across multiple disciplines in the organisation: operations, capital planning, regulatory response, OT, IT, innovation, and procurement. Initially, this team may take a “skunk works” view of the world with many solutions to the problem. Still, it should evolve into a capability designed to support reactive, proactive, and predictive correction of leaks. This requires that your organisation be driven by both process and data (with an underlying operational excellence framework) and can functionally form and disband teams reasonably agilely.
- Data must be brought together. Any leakage operation in your organisation will require that many data sources be integrated. This would include base nightline calculations across DMAs, elevation data, logger information, work history, asset data, etc. If this data does not exist in a central place, where it is joined up and conformed for consumption by individuals or systems, any dream of a digital twin or leakage prediction is likely lost. Don’t underestimate the challenge of bringing this data together and improving its quality, both of which are major efforts in and of themselves.
- PoCs are hard to scale. Innovation funding is great for identifying early-stage technologies that will likely have a practical application. Unfortunately, many innovation projects are assumed to be able to scale without much effort, which is far from the truth. Innovation projects tend to take the form of a proof of concept. In a proof of concept, there should be a bounded scope of something trying to be “proven” and a defined period of time to prove things out. After that timeframe, the concept is deemed to have been proven or not and can then be reorganised to scale and pilot in the operation. Doing it the right way takes time, which many companies want to short circuit, so they take a PoC and ask it to work on a more extensive data set or for a larger scope. Inevitably, this results in disaster, as the PoC was never built to scale and cannot support a full operation.
- Take a hybrid approach. As with our initial point, a lack of silver bullets means you will need to try multiple approaches to solving the leakage problem. This may involve a blend of technology, capability, and engineering solutions that position your organisation, initially, to better respond to leaks reactively. Then you may add more capabilities that improve your ability to proactively target leakage hot spots and put crews on the ground before a slow burner becomes a major incident. Finally, once you have built up strong capabilities in both reactive and proactive response, you should then jump into predictive leakage monitoring. Too often, companies try to become masters of the crystal ball rather than build up a set of solutions that will address a wide range of leakage types and incidents.
These are by no means an exhaustive list of things to do, but a view from individuals who has been engaged deeply in the industry but has also had the perspective of other heavy industries trying to tackle similar problems. What do you think? Have we gone completely out of touch with the reality on the ground? Or have we tapped into a nerve that can help push for some sustainable change in this space? Let’s chat!
Jeff Gilley is a Senior Principal at The Oakland Group