Looking back, I was fortunate to be in the right place at the right time to witness the emergence of the digital water industry. I helped shape two of the largest digital twin platforms for water utility networks, first as the lead architect of the Bentley GEMS platform and later as the product manager of Autodesk/Innovyze InfoWorks WS. In 2000, I created what may have been one of the first digital twins in the water industry: a WaterGEMS hydraulic model connected to SCADA and used to optimise pumps for Bethlehem, PA. At that time, this real-time model was experimental; in retrospect, it was a digital twin ahead of its time.
Today, the water industry is experiencing a major transition, one that could be more consequential than the shift to desktop computers in the 1990s. Several factors have come together to make this possible: a significant drop in data acquisition and storage costs; easier data processing, integration, and access; and the rise of artificial intelligence (AI). These forces are transforming not only our software tools but also how water resources are managed.
Figure 1. Decision-Making Steps (inside ring) aligned with the Digital Transformation Milestones (outside ring) (source: the author)
But the digital transformation of the water sector did not start in the 1990s. Digital transformation, the historic shift of decision-making steps (Problem Definition, Observation, Modelling, Analysis, Decision, Action, and Evaluation) from the human mind to external tools and, ultimately, to computers, began thousands of years ago with early water resource observation tools, such as Nilometers in Ancient Egypt. However, the transformation gained significant momentum recently with the advent of digital computers. Today, we have already entrusted computers with the Observation, Modelling, and Analysis decision-making steps. The next phase will involve transferring decision intelligence from humans to machines, first under human supervision and later fully automated.
The water industry is experiencing a major transition, one that could be more consequential than the shift to desktop computers in the 1990s
This is the moment where digital twins (DTs) become central. A DT is a decision-making platform that integrates a predictive model (deterministic/hydraulic, statistical, machine learning, or AI) with descriptive models, such as asset data (e.g., GIS) and real-time data (e.g., SCADA). DTs are the ultimate goal of digital transformation, marking the transition from human-centric management to increasingly autonomous operations.
DTs already can observe system behaviour, analyse alternatives, recommend actions, and, in some cases, execute decisions automatically. Looking ahead, they will be the backbone of the industry’s move from reactive operations to proactive, and eventually autonomous, management. In the near term, DT value lies in automating routine decisions, such as pump scheduling or anomaly detection, with humans firmly in control. Over time, as our trust in DTs grows and AI capabilities mature, DTs will evolve beyond operational optimisation to support long-term planning, system assessment, and adaptive operational evolution.
.jpg)
Figure 2. Qatium Water Distribution Network Digital Twin (Source: Qatium with permission)
As a member of the Qatium advisory board, I have had the opportunity to see this future. Qatium water distribution network DT has demonstrated that it is possible to provide an open, cloud-native DT platform that integrates hydraulic models, GIS, and SCADA, without the traditional barriers of cost, complexity, or specialised expertise. It also offers an open development platform integrated with AI, providing a gateway to explore AI use cases in a controlled setting. This democratisation of DTs is essential if technology is to scale across utilities of all sizes, not only the largest and most digitally mature.
As our industry faces increasing pressure from climate variability, ageing infrastructure, workforce transitions, and regulatory complexity, DTs offer a path to more resilient, efficient, and transparent water management. The last step in the digital revolution gave us the ability to model our systems; the next allows these models to learn, reason, and act.
