There has been a lot of talk about the growing infrastructure deficit. Across the US, Canada and Europe, experts and policy makers share a growing concern around the many billions required to catch up with this deficit, or in layman’s words: Public infrastructure is aging and decaying. While there may be disagreement about the actual extent of the deficit, there is no arguing that it is there to stay – and that some types of infrastructure age with less grace than others.
Water infrastructure is by no means the only public good at jeopardy. Roads, bridges, power lines and public schools are all affected by the growing infrastructure deficit, totaling – according to civil engineering experts – hundreds of billions or even several trillions of dollars. Coupled with water shortages that extend beyond the need to replace infrastructure and into the need to locate more sources of water supply, the problem with water infrastructure may seem like a ‘mission impossible’. But it is not.
For water networks, unlike many other decaying assets, it is quite likely that parts of the investment gap can be addressed using little to no hardware. An emerging category named ‘Water Infrastructure Monitoring’ carries the promise of optimizing – through the use of advanced software – the process of pinpointing the assets that need to be serviced, and taking corrective action, thereby prolonging asset life and diverting scarce budgets to the right places.
The premise is simple: take the data you have about the network and use it to make the network better. If you apply your insights on the network with diligence and discipline, you will make it live longer. And while water distribution infrastructure is not well-instrumented, the approach can work with surprisingly few data collection points. And where there is apparent shortage in data, you can augment your data with additional readings. Until recently that meant only one thing: adding more meters and sensors.
In some cases, applying a layer of smart sensors onto the existing distribution network or sewage system may prove cost-effective; several companies, such as Echologics, Syrinix or Pure Technologies, are offering an overlay of sensors (mostly acoustic) to complement the existing flow meters. No need to replace aging pipes just because their term is up, or as T. Bert Lance put it – ‘if it ain’t broke, don’t fix it’.
But more often than not, adding sensors in all ‘blind-spots’ is not feasible or not economical. Could there be a way to gain better visibility into the water network without digging and/or plugging new gear into the pipes? Hydraulic modeling is often cited as such an approach, and companies such as Bentley Systems or PPIC have actually incorporated smart hydraulic equation solvers into their solutions. Unfortunately, hydraulic models require constant calibration, making these applications less practical than they may seem at the outset.
The use of advanced statistical and mathematical models to generate a holistic view while also picking up specific network events is not as trivial in water as it is in other space. Shedding light on the behavior of the network and identifying anomalies using only software, turns each and every sensor reading into a relevant, valuable data source, and can illuminate a previously obscure network layout with golden nuggets of knowledge and insight. At TaKaDu, for example, we apply complex and innovative algorithms to data already existing in the network in order to shed light on the network’s behavior. It is not as straight-forward as a layman reading this post may assume, but not impossible as experienced water engineers may tend to speculate.
What can be detected using algorithm-heavy SaaS (software-as-a-service) solution, which can’t be picked up otherwise? Quite a lot, in fact. In just few months of operation in a European capital, TaKaDu has identified dozens of leaks, bursts and inefficiencies, alongside hundreds of other significant network events, thereby improving the utility’s response-time and saving water and energy, but even more importantly – helped generate a coherent view of the network status, directing the pipe-replacement crews and Active Leak Detection efforts to the right places in lieu of an ‘oldest-first’ FIFO approach.
True, budgets for pipe replacement would still be as hard to come by as before. Nevertheless, pointing the limited resources to the right place could mean the difference between an uncontained, escalating water infrastructure gap, and an optimized use of each and every dime put to work for the benefit of the public.