With 2021 nearly here, we’d like to share what’s in store for the year to come as it relates to Asset Management, Digital Twins, and Artificial Intelligence.
Kicking off with asset management, let’s take a step back to see how it’s grown throughout the years. Utilities historically have struggled with reactive maintenance and how to respond to events, such as water main breaks and pipe failures. 2020 was a year in which we continued to help our clients understand that reactive at the last minute can be costly and comes with its own set of risks. To hear more about what lies ahead, watch our Water Talks webinar: A Look Towards 2021
Moving forward, we continue to see our clients pursue prescriptive asset management, allowing them to not only gain a better understanding of when things might fail but also enabling them to prescribe a reactionary method beforehand.
Now getting into what will be the next generation asset management
We've all been having or hearing a lot about sustainability, especially due to some of the repercussions of climate change, like limited potable water resources, flooding or drought issues. Prompting the question of how can a water utility operate more sustainably? How could using asset management methods help to better prepare us for these extreme environmental conditions and be more resilient? To find a resolution for this we had to gain a deeper understanding of where the root of the failure was coming from and how it could be avoided.
Bringing us to what this holds for the future of asset management and what we here at Innovyze are working to provide to our clients- cloud performance and SaaS flexibility.
Using asset management, you create a codified workflow by capturing institutional knowledge and putting it into a formula. By looking at various states of data, from historical to the present, and understanding data in real-time, develops a workflow that becomes reproducible and is used in predicting the future, this is prescriptive asset management. You get a prescribed action for a specific state.
But we still need to prepare and understand these multiple possible states. For instance, with optimization, you tend to choose a path through discovery, you find that path and focus less on non-optimal scenarios. Understanding resiliency, we also have to prepare for these very improbable states.
For instance, how would anyone have known to prepare for the impact of COVID-19? What is the order of these causes, of these states and how do we better prepare for them? We have to go through the exercise of what would happen if some of these things came to be. These are a lot of circumstances going on in a very complex environment but it’s cloud computing, Digital Twins and artificial intelligence that allows us to even try to work through the impossible.
What’s next for Digital Twins
A year ago, we were talking about digital twins as decision support tools for operational water, wastewater, stormwater network management and were predicting that we were going to be seeing more of them implemented during 2020. And that is precisely what happened for both water distribution networks and stormwater sewers.
For 2021, we will continue to look at different aspects of Digital Twins and examining the driving forces behind the digital transformation of water utilities. Going beyond one of the first most important qualities, visibility, we are looking towards sustainability and resilience. This ultimately helps utilities get ahead of any potential repairs, replacements or conflicts before they are even visible.
Sustainability refers to optimizing the use of resources, whether those are water resources or stormwater or carbon or money, it usually involves delivering the right service levels in an optimal way. Resilience is focused on exceptional circumstances. How do you continue to deliver a service when the network fails during extreme weather, huge rainfall, storm surge drought or during a pandemic?
Some other exceptional circumstances, like drought and flooding, can be controlled by using the Digital Twin which can adapt to these circumstances, predicting the impact on the consumers, and then allowing for re-planning on the fly. Using a Digital Twin you’re able to review the impact within the operations in real-time, allowing for faster communication and response. Looking ahead, water utilities can use Digital Twins to support them, both to improve visibility, sustainability and resilience.
Bringing AI into the fold
In regards to artificial intelligence, with the Innovyze product Emagin we've been primarily focused on predictive control applications. Predictive control is about using more variables from your environments and using that information to optimize how you navigate through the environment. That is the central premise of what predictive control is and where we started with the Emagin platform.
To take things even further, we took that concept and applied it to when operators are in the control room and operating wastewater treatment facilities, to see how much more information could be leveraged from the environment, in order to determine an optimal pathway for water-intensive operations.
If you look at any plant operator, at any given time, you know, they've got to decide whether to turn pumps on off, what concentrations to apply for chemical dosing, when to maintain pieces of equipment and what lab tests to run. They've got to do all of that, and there's only so much that a human operator can do at any time. We apply that predictive control concept to help them make the best decisions, at any given time with respect to their operations and control in the facility.
Now we're starting to think about how we can integrate them with the hydraulic models so that we can start generating more data. This is valuable for AI to learn from because we can start exploring scenarios that a facility or a network hasn't seen before, by generating data, and letting the AI learn the most optimal trajectory of control in these new unforeseen conditions. These conditions then form the guardrails to how the AI is recommending a predictive control workflow. But as you start looking ahead that's another area for integration because we can start using information from asset management systems, and maintenance constraints.
These are some of the things that we're exploring on the AI side, to continue to build upon the robustness and scalability of the existing technology with more integrations both on Digital Twin, as well as the asset management aspect. If we tie that AI intelligence to like our hydraulic models, we can start being more prescriptive in terms of planning decisions as it relates to distribution networks, and stormwater systems.
In the years to come, it's more about extending our prescriptive analytics capabilities, both with regards to operations and asset performance management.