AutomatedBuildings.com Article - How I Learned To Stop Worrying and Love Building Data

Building data has the potential to change everything as we know it in our industry, and change can be scary. Your view of the landscape for a data-driven buildings industry depends largely on where you sit; startups and sensor and controller manufacturers are delighted, while operators, technicians and IT administrators are classically apprehensive. A buildings industry with an abundance of sensors, actuators, devices, and smart things that tend to fail raises critical questions: Where will all that data go? How can we possibly deal with all of it? How will data impact my job and the services that I render? How will we manage buildings with all that measurable data, and smart equipment capabilities?

This isn't the first time we've faced such sweeping changes, in fact, today's market focus on data and the use of data in buildings is merely a resurgence of the fervor that started at the end of the last century but ran cold because of challenges associated with cost-effective building data management. It's taken a few years for technical challenges like open communication protocols, data storage, and web-accessibility to catch up with the concept of a data-driven buildings industry, but it seems like our fears surrounding building data are being laid to rest for good.

Several technologies and industry culture shifts have been instrumental to helping people start to love building data, instead of fear it, and we will present those to you within the context of Clockworks^TM, KGS Buildings’ Software as a service (SaaS) and ‘big data’ solution for buildings.

First off is the 'cloud'. Many people tend to think that a 'cloud' is simply a datacenter made up of a network of servers which can be used as necessary for a given website or application. While that is partially true, the deeper value of the cloud is more subtle and usually mysterious to the end user. However, it can be a critical element to building data management at a portfolio-wide and global scale. The real values of the cloud are scalability and efficiency in managing data and applications, driving costs down while expanding functionality of applications. A network of on-demand, traffic managed, load-balanced, virtualized servers, working together with geo redundant storage, and distributed caching in a ‘cloud’ can accomplish more than any single server could do on its own, especially for collecting, storing and analyzing large volumes of data.

A well-instrumented HVAC system, logging at five minute intervals, can produce gigabytes or even terabytes of data within a short period of time. This can quickly fill the limited storage capacity and greatly tax the processing capabilities of a single server; and that’s before you start to analyze and serve data for users. A well architected cloud solution can solve data storage, processing, and management problems, as well as many other inherent security and networking challenges at significantly less cost than building and maintaining your own server farm. Stand-alone solutions are short term, hardware will eventually not be sufficient, and untimely failures will occur. This results in additional maintenance costs, which can become unsustainable at a large scale. Software will always need upgrades, and deployed legacy systems will still need maintenance throughout their useful lives, resulting in more hidden costs, concerns, and unnecessary user entrapment. On the other hand, the cloud makes it possible to get software upgrades automatically, scale capacity while maintaining performance, manage and access all client buildings in one platform, eliminate substantial maintenance, and provide web accessibility from anywhere.

KGS Buildings’ monitoring and automated ongoing commissioning software-as-a-service, Clockworks, is a good example of what the cloud can enable. Clockworks was built for large scale data collection, storage, and most importantly, automated analytics. Our cloud based platform is broken up into components architected to achieve individual levels of on-demand resources for automated scalability. Efficient data management means that there is no handling difference between handling 100 and 100 million data transactions per hour, and that terabytes of data may be processed and stored indefinitely without performance implications.

This is the true value of the cloud, and having used it, we know that this scale of efficiency is possible. We've engaged whole campuses of buildings, with tens of thousands of points, on a single cloud based solution for building data storage, collection, and daily analytics. The cloud is king when it comes to transforming the industry with data - without it, you can't reach a meaningful unified scale, and the industry had to wait for that technology to mature before learning to love building data.

Next is artificial intelligence. Saying those words conjures images of HAL and more importantly the litany of failed prior attempts to make successful expert systems, but fortunately, there’s a different approach to succeeding here. In buildings, the challenge is taking data from a variety of sources and using automated diagnostics to turn it into actionable information. Previous attempts to apply artificial intelligence have faced numerous barriers such as poor integration capabilities, lack of common communication protocols, lack of scalable storage, lack of metadata, lack of scalable and shareable computing resources, and lack of diagnostics and analytical tools that can be quickly and cost effectively deployed. On the other hand, there is a plethora of methods to analyze building data to identify opportunities for energy and cost savings. Statistical analysis methods, grey box and black box model-based methods, rule-based fault detection and diagnosis, and expert systems have been actively researched for a long time.

While analytical methods were developed in industry and academia to analyze building data, the technology for applying these methods commercially to provide value to customers had to mature. Today, it is possible to leverage modern hardware and open communication protocols to access data, use cost-effective cloud-based data storage and computational resources, and leverage software architectures implementing a variety of analytical methods and presenting information and data in a variety of interfaces and data and informational interfaces to finally apply advanced ‘artificial intelligence’ methods. It is now possible to turn the reams of data from buildings into information that people can understand and use to make decisions about operating, managing and servicing buildings. In Clockworks, all you have to do is specify what sort of equipment you're monitoring, what type of data exists for that equipment, and then point and click to apply analytics - Clockworks figures out all sorts of things that can be analyzed about the equipment based on what you know and are measuring about the equipment.

Although emerging technologies, including the cloud, have made it possible to automate data-driven diagnostics, real change requires communicating actionable information. Traditional engineering consultants provide in-depth analysis and action plans, but they can be costly. Simple alarms are used in most control systems, but many are ignored as “nuisance alarms”. Automated diagnostic systems have the advantage of being less expensive than an engineer, always present, and more in-depth than an alarm.

In Clockworks, we’ve tried to bring data closer to action by adding plain English text to the numerical and graphical results of our diagnostics. For example, if one diagnostic finds simultaneous heating and cooling on an air handler, it will output a statement similar to the following: “The preheating coil and cooling coil are heating and cooling simultaneously”, and list possible causes such as “Valve is in manual override” or “Valve is leaking by”, and estimate the cost of wasted energy such as $5,000 over the month. With such a report in hand, an operator can go check valves and sensors of the affected air handler, confirm the issue, and make repairs or call in a service contractor.

It is ultimately the building owner, operator or service personnel who determines the actual issue and takes action. Unfortunately, the current culture in building operations is far more reactive than proactive, and this may be the greatest hurdle for the success of data-based diagnostics. We’ve spoken with building operators who have a general distrust of automated fault detection. Other operators see potential, but spend so much time “putting out fires” that they don’t feel they have time to devote to proactive operations. Perhaps the largest negative motivator is that many operators are judged, not by the energy performance of the buildings in their care, but by the number of tenant complaints. Because of this, operators are often reluctant to make even beneficial changes to their operations.

Conclusion

Utilizing building data to its full potential may dramatically change the building industry, but there are still many fears and barriers slowing it down. However, as customers increasingly demand solutions that utilize building data to provide better services and save energy and money, solution providers will have to adapt to the changing needs of the market. Thankfully, new technologies like the cloud and software platforms like Clockworks provide scalable and cost-effective solutions enabling people to use data to improve building operational efficiency. Like us, owners and service providers may find they can stop worrying about whether and how to implement a solution for using building data, and start deriving more value from it.

KGS Buildings, LLC provides their customers with tools and services to answer their unmet need for combining information and building technology to yield greater building operational efficiency. Through a Software-as-a-Service offer, Clockworks, and building engineering support services, KGS helps contractors, owners, engineers, and operators use building data to better manage building assets.

Led by a team of partners, Stephen Samouhos, Nick Gayeski, Siân Kleindienst, and John Anastasio, KGS Buildings works in many different types of buildings, ranging from large commercial real estate portfolios and universities, to multi-family affordable housing complexes and retail centers; across the market, KGS has delivered real results for reliable energy savings, pro-active maintenance programs, and more efficient operations through data driven building information tools. An 'MIT-startup' built purely from sweat equity, KGS was founded 5 years ago by Nick, Siân and Stephen during their PhD years, and has since focused on the very unique market demand for combining sound building science with scalable information technology. With global partnerships to deliver Clockworks, and a growing customer base into the hundreds of buildings, KGS is excited about a future where people can easily use data to more effectively manage building assets, reduce energy costs, and streamline operations.

Events	Want Ads
Our Sponsors	Resources