The building data analytics revolution in three acts

More than any other technology, the fast-advancing field of semantic, or tag-based, building data analytics is changing how we design, construct, and operate high-performance buildings. Increased transparency of operational building data is impacting accountability. This in turn is fueling growth in smart building technologies and encouraging more building-industry professionals to hone expertise in wrangling insights from this data. It is a virtuous cycle encompassing three major activities:

1. Closing the design loop with operational building data;
2. Creating actionable intelligence from building performance data; and
3. Using the ROI from deeper energy savings as a platform to creating smarter buildings.

Act I: Closing the Design-Operations Loop

First, delivering a building’s operational data to the right people at the right time has potential to bridge a major disconnect in building design, construction, and operation. Design teams rarely get enough feedback about how their solution is performing after it has been delivered, and operations teams rarely have opportunity to offer input when a building is being designed. In this scenario we are faced with the challenge of designing and operating ultra-low energy buildings, while balancing ultra-high expectations for enhanced occupant comfort and end-user experience.

The New Bund Enterprise Center in Shanghai by DLR Group. Rendering by DLR Group.

The design industry is evolving from a reliance on predicted EUI, or energy use intensity, to gauge successful projects for a greater emphasis on outcomes and actual post-occupancy building performance data. Optimizing building performance early in design, with an operations feedback loop for ongoing performance, is now a requirement, not an option.

As my DLR Group colleague Amarpreet Sethi explains in this video, our goal is always to analyze building performance data early in design to bring quantitative insight to the decision-making process. Design team members are typically first to forecast how occupants will use the building. These behavioral predictions factor into simulations that look at key high-performance building indicators relative to energy consumption, indoor air quality (IAQ), occupant thermal comfort, and overall user experience.

Data analytics is also used extensively in the commissioning of building systems prior to final handover to the operations team. Critical to ensuring that design strategies are executed as intended, connected-commissioning (CCx) approaches help us optimize design during construction as needed to deliver the building at peak performance on the first day of occupancy. A building performance data analytics platform makes it practical to evaluate 100 percent of the results streaming from controllable points during the functional test phase, such as every terminal unit, rather than a 10-to-20 percent sampling. Once a building is instrumented for analytics, the detailed operations of systems can be analyzed at will. Over the long-term, this feedback loop will help high-performance designers ensure that efficient design strategies are continually optimized based on the ongoing dynamic needs of the end user.  

Baoshan Long Beach Winder Tower in Shanghai by DLR Group. Rendering by DLR Group.

When design teams have this inherent framework in place for validation of post-occupancy building performance, they can apply the same rigor of analytics to the building’s operational data set. Extending the process that has traditionally occurred with the design data model can close the design- operations loop.

Act II: Turning Operational Data into Actionable Intelligence

Our buildings are overflowing with data points. We have utilities data, metered data, building automation systems trend data, IAQ data – and that is just the start.

With the vast quantities of a building’s operational data available, the use of a semantic, or tag-based, data analytics framework has become an essential tool to leverage insights that can help us optimize the ongoing operations of our buildings.

Our industry tends to favor open protocols like BACnet and the use of semantic data modeling such as Project Haystack because these make the job of normalizing data for analysis easier to communicate.Early this year when ASHRAE announced semantic data modeling is galvanizing building performance, the announcement signaled that the various development communities and standards bodies working on semantic tagging are collaborating for a unified standard that will ultimately be adopted into BACnet. It is good news for the early adopters of open-protocol semantic tagging.

Algorithms, or tuned fault detection and diagnostics (FDD) rules, can now be created and customized to match a facility’s equipment and identify equipment performance issues. These rules-based data analytics tools reveal the operational patterns of building systems, automatically crunching through vast amounts of data to find and report anomalies or issues that previously could not be uncovered cost-effectively through conventional analysis.  This allows an experienced building performance analyst to use a monitoring-based commissioning (MBCx) approach to monitor and verify building performance for the ability to visualize this data graphically, analyze it, and prioritize issues based on calculated impact to the building’s energy use, IAQ, occupant comfort and operational efficiency. This process provides a curated, prioritized list of issues with quantified impacts on building performance to be presented to the building facilities team, on a weekly, monthly or quarterly basis, and facilitates a proactive, rather than reactive, approach to operations and maintenance.

Baoshan Long Beach Winder Tower in Shanghai by DLR Group. Rendering by DLR Group.

Once a data gathering framework and analysis protocol is established, we can leverage actionable intelligence to achieve deeper energy savings, enhanced IAQ, increased occupant comfort, and a proactive approach to building maintenance – all of which impact a facility’s operational financials and enhance the end-user experience.

From here the possibilities for improvement are endless. More sophisticated manipulation of a building’s data set makes room for further innovations such as automated work-orders, four-dimensional plots of building sensor data, DVR’s of whole-building energy-use down to the individual terminal unit level, and much more, for the continuous monitoring, visualization, and optimization of building performance.

Act III: Utilizing Energy Savings for Smarter Buildings

The third phase of the virtuous cycle is all about what can be built once you have laid the foundation for deeper energy savings by collecting, visualizing and analyzing operational building performance data. Adoption of new technology is often still justified by the ROI of energy savings, despite the well-documented benefits of enhanced occupant health, wellbeing, and productivity, in addition to operational efficiencies gained with smart building technologies. The good news is that operational data analytics are helping us find new layers of energy savings, with a very attractive ROI. 

From the data gathering framework, it is a short path to deploying IAQ monitoring, thermal comfort optimization, occupancy awareness, personalized lighting controls, and real-time conference room occupancy tracking and scheduling. These types of smart building apps offer investment returns that can be challenging to quantify, but are likely to deliver much greater value to property owners and tenants over the life of use.

The Mob Museum in Las Vegas, Nevada by WRL|DLR Group. Photo by Kevin G Reeves.

More smart building features also mean more Internet connectivity and software in buildings. Standard metadata tagging helps simplify the resulting data and maintain compliance with cyber-security best practices. With a tagging-based framework, operational data models express more than a basic identifier and time-stamped measurement. They also carry contextual metadata describing a piece of equipment or device, as well as its relationships to other devices and systems in the building. Tagging can convey hardware make, model, and version numbers, as well as firmware and software updates. Open source tagging can also be consistently incorporated into BIM models, streamline the deployment of analytics, and make the feedback loop more accurate and efficient for design professionals. 

The best news is that the technology to do all of this is already in place. Operational data analytics for deeper energy savings has become the gateway to more powerful smart building strategies. Of course, the virtuous cycle brought about by this revolution in operational data analytics is not something to watch from the sidelines. It is not a play in three acts. It is a call to action for our industry to be ready to harness the power of building data analytics.