4 min read

Nexus #11 (2/25/2020)

Analytics in mainstream news; the EcoDomus digital twin; Augury's machine monitoring using machine learning

👋 Welcome to Nexus, a newsletter for people applying analytics and other smart building technology—written by James Dice.

If you’ve been forwarded this email, you can sign up for your subscription here:

Sign up now

This is an experiment, and I’d love your feedback. If you have thoughts, questions, ideas or tips, join the discussion on LinkedIn or hit reply.


My latest ideas

+ When analytics insights fall flat—One of my essays was featured on the Gridium blog. Here’s a taste:

If you’re finding cool stuff with your analytics and it’s falling on deaf ears, it might be time to take a step back and consider this: EMIS is an exploratory analytics tool. We use it to find the needle in the haystack and to understand what the data is hiding in our buildings. But to get stuff done, we need to transition to explaining what the data means, including who, what, where, why, and how.

+ The latest on LinkedIn:

It’s commonly accepted in our industry that O&M staff are “in fire-fighting mode” practicing “reactive maintenance”.

Personally, I think the “fire-fighting" narrative is true (seen it for years). It’s also getting stale. Every seller of solutions claims to enable building operators to see the future and spend their time preventing problems instead of fixing them.

And yet, I haven’t seen maintenance practices or this stale narrative change in my 10 years in the industry.

One potential reason: I think we create confusion with what we’re proposing as the solution to reactive maintenance. We have:
- Preventative maintenance
- Prescriptive maintenance
- Proactive maintenance
- Condition-based maintenance
- Reliability-based maintenance
- Predictive maintenance

How do you define these? What’s the difference?


Ideas from elsewhere

+ NYC buildings prepare to drastically reduce emissions to avoid penalties (NY Post)—it’s not often building analytics software gets into the newspaper. That’s what happened a few weeks back, as the New York Post covered Local Law 97 and featured Prescriptive Data’s Nantum software.

While the coverage is exciting, the article only skimmed the surface of the issue.  If the spread of benchmarking ordinances and mayoral commitments to the Paris climate goals are any indication, this law is coming to a city near you.

So let’s go deeper with 5 details you need to know…

  1. It stems from New York State's Climate Mobilization Act designed to cut greenhouse gas emissions 40 percent by 2030 and 80 percent by 2050
  2. It builds on Local Law 87, which requires retro-commissioning and an energy audit every 10 years for buildings over 50,000 square feet, and Local Laws 33 and 84, which require annual rating and benchmarking
  3. It adds teeth in the form of substantial legal and annual, recurring financial incentives for compliance or inaccurate reporting—no longer can building owners ordinances and simply pay a small fine for non-compliance or poor-performing buildings
  4. It provides a straightforward limit on the amount of greenhouse gas emissions a building can have. In year 5, covered buildings will need to submit a report showing their carbon emissions. In year 9, the permitted emissions caps are significantly lowered, in an effort to get to 40% reduction. There are alternative compliance pathways, such as carbon offsets.
  5. The grid matters. Local Law 97’s emissions reduction targets, including its increased stringency from 2030 onward, are premised on New York City’s electric grid becoming much cleaner in the future. It’s unclear how building owners will meet their goals if the grid doesn’t get clean enough.

+ The EcoDomus Digital Twin (YouTube)—As a leading “Lifecycle BIM company” EcoDomus is well positioned for the digital twin market because they’re already in the practice of transitioning as-built BIM models to O&M teams. This 1 hour video walks us through their digital twin platform.

For those that don’t have an hour, use this link to check out a couple minute demo of the Sydney Opera House’s twin.

I learned some new use-cases for Digital Twins:

  • Maintaining a before/after view of every project in the building. Allows you to go back in time. Considering how much time I’ve spent digging through rooms of drawings trying to piece together different renovations, this sounds helpful.
  • Comparing design exhaust flows to real time exhaust flows and showing the ductwork as a heatmap according to the data
  • Ability to redline the digital twin to add an image to an issue (similar to Bluebeam)

Finally, since we defined the modern digital twin a few weeks back, I had to compare my definition to theirs. Their definition of digital twin (BIM + IoT) is missing two key parts of the definition:

  • Computation/Optimization
  • A platform for applications

Regardless, this video is well worth the watch.


+ Augury: A Sign of What Might Happen in the Future (Forbes)—I’ve had a lot of questions recently about AI (where is it applied in buildings?) and proactive maintenance (can you give me some examples?). One startup that checks both boxes and more people should know about is Augury.

They install sensors that monitor vibration, temperature, and magnetic field data of heavy machinery like pumps and chillers. The data is uploaded to cloud applications that run machine learning diagnostics to detect malfunctions (before they occur!) and issue actionable alerts as necessary.

The machine learning prediction algorithms compare each piece of equipment’s unique vibration signals to that of a large database of equipment that has malfunctioned. In other words, they know what a pump sounds like when the shaft is about to break.

Augury’s software can be integrated with other software, allowing automatic parts ordering, work order scheduling, and budgeting. I'd imagine it could be used to send a command to stop the system that’s about to fail.


OK, that’s all for this week—thank for reading Nexus!

If you have thoughts on this week’s edition, head on over to LinkedIn:

Add your take on LinkedIn