Tweet

September 2019
AutomatedBuildings.com

[an error occurred while processing this directive]
(Click Message to Learn More)


The Digital Twin

A First Step in Attracting & Retaining Commercial Tenants
Adam Benson


Adam Benson,
VP of Engineering
BuildingIQ

Digital Twin
Articles
Interviews
Releases
New Products
Reviews
[an error occurred while processing this directive]
Editorial
Events
Sponsors
Site Search
Newsletters
[an error occurred while processing this directive]
Archives
Past Issues
Home
Editors
eDucation
[an error occurred while processing this directive]
Training
Links
Software
Subscribe
[an error occurred while processing this directive]

In 1989, the movie Field of Dreams generated the famous line: If you build it, they will come. Unfortunately for commercial property owners, this is not always the case when it comes to tenants. It is well known that one of the largest challenges for commercial property owners is attracting and retaining tenants, followed closely by controlling operations costs and tightening resources.

And unlike in the late 80s when Field of Dreams debuted, today’s connected world has raised expectations for what defines an attractive commercial environment. This mindset challenges property owners, who often lack data-driven insights about their properties, to know how to make the right investments to match the demands of the modern commercial tenant.

Investing in the Future: IoT and the Digital Twin

A growing trend among building owners is to turn to the Internet of Things (IoT) investments, like wireless temperature and occupancy sensors, as a way to modernize their facilities while at the same time realize operational efficiencies. IHS Markit estimates that the number of connected devices will soar past 125 billion over the next ten years. For the building industry, this means petabytes of data generated from building devices can become the backbone of intelligent buildings.

However, this smart building future is not a given. IBM Research found 88 percent of IoT data goes unused. This means to recognize the most return on IoT investments, cloud analytics is foundational. And the more IoT data mushroom, the need for the cloud analytics-based “digital twin” becomes self-evident.

A digital twin is a digital representation of both the elements and the dynamics of how an IoT device operates and its interconnections in a physical space. By definition, a digital twin has to apply analytics at every step and can use these advanced analytics to dynamically recalibrate to its environment. It feeds on data and improves itself, the more data it processes, putting all IoT data to use in applications like buildings.

Optimized Operations with the Digital Twin

One of the most underutilized IoT investments in commercial buildings is the building management system (BMS) itself. The BMS was among the first IoT applications for buildings. When integrators began to add IP gateways, it effectively turned every device into an Internet host that could be monitored and controlled. Yet most buildings are run in maintenance mode or are poorly tuned—unable to modulate their behavior— let alone able to work around the design constraints of the building system. It is unlikely that the designer thought about things like hot-desking; integrated calendaring and room reservations; or laws mandating reduced carbon emissions. This creates inefficiencies in building operations and leaves the BMS poorly leveraged as an existing IoT network without the cloud and digital twinning.

The digital twin addresses these shortcomings by starting with data from the BMS, and other IoT devices, and layers on context, treating them as discrete assets to build a map of how the system (building) is behaving. This brings to light how discrete assets sit inside more complex systems and how systems are interconnected to other systems. Now when an issue is identified, the map can be run to determine if there is inertial impact and validate if the anomaly has undue influence on the system. Such an effort is practically impossible to retrofit into an existing BMS, but it is an inherent capability of the computational power, logic, and artificial intelligence that cloud-based digital twins can offer.

For example, a BMS system can modulate zone comfort based on the return air temperature or zone temperature. What it cannot do is understand that a comfort issue may stem from the number of people temporarily in the zone or the radiant heat from blinds that are raised, which normally are not. An occupancy sensor and a window blinds sensor combined with sun positioning data would provide the additional data to get to the root cause of the discomfort. No BMS alone can do this today. But a digital twin connected to the cloud could.

The above example illustrates how the digital twin can reveal how buildings actually operate in real-time versus their designed behavior. A digital twin accumulates both granular knowledge at the device level as well as system knowledge — even cross-system knowledge. Similar to a facilities team over decades of experience, the digital twin learns and evolves to improve decision-making, but it does so at an accelerated pace and with less bias than humans through artificial intelligence, machine learning and other techniques. It creates and preserves institutional knowledge within a building by creating a virtual catalog of how an asset operated and was maintained over 20, 40, 60+ years.

The Digital Twin and AI Engines

The concept of a digital twin highlights the value of trending building data, storing it for decades and applying algorithms, rules and processes over the top of it. An artificial intelligence (AI) inference engine combines a digital twin with tools that identify relationships across seemingly un-related data sets – it is the flow of these tools when combined together that identify which assets are contributing to the energy consumption or operational stance of a building. It will analyze the strength of those relationships to highlight which data points (assets) are contributing the most to the consumption of energy (typically used as a primary indicator of performance/impact) that period of the query.
 
In simple terms, AI engines leverage the digital twin and a series of mathematical and statistical functions and classifications that take building data sets and create highly normalized relationship maps. These maps, depending on the richness of data collected, can show strength and weakness of influence between mechanical components in a building. The power of an AI engine is in its ability to identify relationships and influences happening within a building’s mechanical systems that perhaps was not obvious without the use of big data and AI. As important when compared to standard industry practices of identifying faults, AI engines are not limited by reliance upon known, specific rules for flagging an issue or fault.
 
[an error occurred while processing this directive]Unlike other AI techniques, it is not a regression model of past performance to predict future performance, but more akin to “relationship builder” that measures the strengths and weaknesses of all data sets to find which elements of an asset influence one another. It does not need to know the programming logic or the mechanical flow – it will work that out on its own. The benefits of this are powerful in that historic knowledge, gleaned through countless hours talking with the chief engineer or combing through mountains of documentation sets and manuals, is not required. The asset and its inherent behavior are forever captured, annotated and recorded.
 
This is by no means a replacement of the people and technical services that manage and tune the mechanical and electronic elements of the building, but a set of tools and utilities that improve the performance of the team responsible for its maintenance and upkeep. AI engines on their own can not fix or resolve issues. It can only highlight and guide the building facility teams to anomalies and relationships in the building elements that are not performing at their most efficient. Artificial intelligence and the digitization of building data will not necessarily drive energy improvement or operational excellence on its own. But without requiring new sensors or expensive retrofits, digital twins and AI engines are about using the team you have, the current state of the building asset and optimizing it to its full potential to be more cost-effective.

How to Get Started with the Digital Twin

The first step on a building’s optimization journey is far simpler than one would imagine: unlock the data in the BMS. Doing so means building owners can start to make progress with the digital twin, which is a low-risk proposition. By definition, the digital twin evolves and improves over time, so going “all-in” is unnecessary. The BMS has massive amounts of data that despite decades of effort, is still largely locked away, in the worst case, or massively underutilized in the most common case. Many building owners have increased financial exposure due to contracts that lock down access to data, domain experts instead of databases, and an inherent bias among facilities teams to go it alone.

Beginning with the digital twin, an AI engine and the right set of services built upon it can create appealing, modern commercial spaces while at the same time generate efficiencies in operations and de-risking ownership. With this level of return on investment now and in the future, the new quote for building owners may become something like: If you optimize it, they will come.


footer

[an error occurred while processing this directive]
[Click Banner To Learn More]

[Home Page]  [The Automator]  [About]  [Subscribe ]  [Contact Us]

Events

Want Ads

Our Sponsors

Resources