Skip to main content

How Big Data and analytics change the game for HVAC systems

<p>Big data is sending the winds of change through heating, ventilation and air conditioning systems for commercial building owners and operators.</p>

Disclaimer: The views presented in this article are personal views of the authors and in no way represent those of the company they work for or any industry body they are associated with.

Big Data refers to high-volume, high-velocity and/or high-variety information assets that require new forms of processing to enable enhanced decision making, insight discovery and process optimization. How big is it? Big Data is measured in terabytes (1,000 gigabytes), petabytes (1 million gigabytes) and zettabytes (1 trillion gigabytes).

The world is full of examples of Big Data. Walmart in the United States handles about a million credit card transactions every hour. Across the U.S., there are more than 40 billion credit card transactions every year. Using more than 60 years of historic data and models involving 82 billion calculations, one startup company is now able to accurately predict weather conditions 40 days in advance. IBM interprets 350 billion annual meter readings to better predict power consumption.

Globally in 2012, 2.7 zettabytes of digital content was created. The advent of smart equipment and greater digitization significantly has increased the volume of data points available from the field. Besides the volume, complexity of data and establishing interrelationships is a huge challenge. Globally, $28 billion of IT spending in 2012 was on Big Data technologies, and is expected to increase to $34 billion in 2013.

Over the last few years, it is not just the volume of data that has increased exponentially; it is also the variety – we capture more variables today on top of capturing more samples of each variable.

Taking the example of buildings — the traditional approach to energy-baseline definition would have been to correlate whole building consumption with outside air temperature, maybe also occupancy and level of operations. However, with advanced metering and building automation systems, we now can analyze data by individual service type, specific building location and even specific time of the day. We therefore have more variables to take into account (and of course the large volume of data) for the same analysis — which is great as it allows for more targeted analysis, but which also means that we need to identify the most important variables for analysis, as otherwise we could easily be lured into misleading analytics.

Green HVAC photo by nostal6ie via Shutterstock

Big Data always has existed. Today, technology allows us to capture and use Big Data to connect various elements of the ecosystem — facility data, performance and operational data of the systems installed in these connected facilities, maintenance histories and manufacturers’ data. Advancements in technology are making it easier to connect, capture and use Big Data. Technology finally has evolved to allow us to digitize building subsystems and make them transportable virtually as Big Data.

Therefore, the challenge of Big Data analytics really starts with big data management — and then the analytics. Big Data analytics must in turn lead to “big insights” and “big actions” — the challenge today is in completing the “Big Data and Analytics Value Chain.”

Big Data in HVAC: The case of chillers

In a building, the heating ventilation and air conditioning (HVAC) system typically consumes about 60 percent of the building’s total energy requirement. Of this, chiller plants consume 35 percent — translating to 20 percent, or one-fifth of the total building energy use.

So, chiller plants must be one of the first logical starting points for any building energy-efficiency initiative. On the average, about 30 operating parameters of a chiller can be monitored. If these parameters are captured and recorded every 15 minutes, it translates into more than 1 million records a year for a single chiller. For a medium-sized building, normally about 600 HVAC equipment and system data points are captured using building automation systems (BAS). If the information is captured every 15 minutes and stored for each building, there will be over 21 million records every year. For 5,000 similar buildings, over 105 billion records will be captured annually, translating into over 4.2 terabytes of data storage.

Building owners and operators can leverage Big Data and Analytics in many ways to create tangible value:

  • Advanced analytics. It can help with better understanding of building and equipment performance. It allows historical trending, pattern recognition and correlation between cause and effect of issues and events occurring in the various building and HVAC subsystems.
  • Intelligent insights. It enables benchmarking of a building’s HVAC system performance against industry standards or benchmarks. Owners and operators can cross check what their energy usage is in HVAC systems and how they stack up against peers.
  • Preventive maintenance. Through proper analytics on past performance data and issue trends, future potential maintenance issues can be identified through simulation and predictive technologies. Such actions will help extend equipment life, reduce operating costs and minimize disruption.
  • Informed decisions. Leveraging Big Data and Analytics, building managers also can model their future energy requirements and simulate their future operating budgets.
  • Value asset. Monetize raw data for parties interested in sustainability such as educational institutes, research bodies and policy groups.
  • Connected communities. At a fundamental level, virtualization of building subsystems allows harnessing dispersed experts by creation of a connected community of advisors to enhance performance of buildings. "Bringing the building to subject matter experts" is now possible through organized Big Data and the Internet.

Big Data and Analytics brings much additional value for HVAC systems manufacturers and service providers:

  • Improve future design by understanding how their equipment and systems are used by customers, facilitating the alignment between product development and customer needs
  • Anticipate future repair and replacement needs, thus improving service quality and planning
  • Increase service productivity with more accurate targeting of current and future issues identified by analytics on Big Data
  • Differentiate their relationship and service offerings from organizations that are not leveraging Big Data and Analytics, thereby creating new economic models

Big Data and Analytics helps us understand what is going on with buildings and the HVAC systems in them, the implications of that and what kind of actions are recommended to improve building performance. This is achieved as we progress on the dimensions of understanding the data and putting it in the context in which the data is created. As we get better in moving from data to analytics, we move from more descriptive analytics to predictive analytics.

Here are some examples of how we can use Big Data to create output in a progressive manner and how we can use that output to answer more analytical questions that brings increasing value. This is based on a model developed by Accenture to put in perspective Big Data and Analytics.


The urgency around Big Data and Analytics

Today, it has become imperative for building owners/operators and OEM/service providers to start making use of Big Data and Analytics. There are several key drivers for this:

  • Technology in "digitization of buildings" is now a reality. Customers and building operators should be able to demand to be supported and served by the subject matter experts and best advisors from OEMs and service providers.
  • Reliance on a building and HVAC system performance is becoming critical to a company’s success as it helps with occupant health, comfort, productivity and compliance.
  • In several industries, including pharmaceutical, food processing and scientific research, temperature and environment control of the facilities is critical for proper conduct of business. This makes the HVAC systems even more vital for such applications beyond health and comfort as the revenue of such industries is fully dependent on the facility environment.
  • Many countries have statutory requirements around energy usage. To comply with most of these statutes, building owners must submit fairly detailed proof of efficiency and performance of systems, and often provide aggregated results from those.
  • Most corporations are becoming more involved in their sustainability initiatives with the commitment coming straight from senior management. Big Data and Analytics creates technology and information platforms for the facility managers and sustainability leaders in organizations to drive sustainability initiatives.
  • Big Data and Analytics provides some alternative options to capture and institutionalize expertise on HVAC systems in a world competing for top talent.
  • Organized big data from buildings now need to be managed also by IT teams along with other business groups such as finance, HR, legal and knowledge management. There should be similar considerations on security and access. This is the convergence of building systems data with information technology.

How to handle Big Data and Analytics

Addressing opportunities and challenges around Big Data and Analytics in the space of building and HVAC systems requires a systematic approach. Some key steps involved in this process are:

  • Capture – connect and collect information from different equipment and building sub-systems
  • Curate – select and organize collected information
  • Manage – store and correlate information to derive knowledge and wisdom
  • Process – analyze and present information in an actionable format with economic impact indicators

The HVAC and buildings industry is evolving in interesting ways in response to the opportunities around Big Data and Analytics. Almost all the large equipment/system manufactures and service companies such as Johnson Controls, Honeywell, Carrier, Trane, Schneider, Siemens, Daikin and Lennox are investing aggressively in this space to become the market leader. This space is also seeing interest from mechanical contractors and facilities management companies. Large government and private organizations have undertaken their own initiatives around Big Data and Analytics to further their sustainability agenda or optimize their energy and operational budgets. Most large IT companies such as IBM, Wipro and SAS are also moving fast to build capabilities and solutions in this space. They are incorporating learning from different industry domains which also create and use Big Data to apply to the HVAC and buildings industry. Accessibility of the enabling technology is leveling the playing field.

Challenges in developing strategies around Big Data and Analytics

Big Data and Analytics is quite matured in certain industry domains such as banking, financial services, retail, defense and security. However, in the HVAC and buildings industry, it is still in its early days but evolving rapidly. Companies invested in Big Data and Analytics are addressing several challenges posed by the uniqueness of this discipline:

  • Organization of building data to enable analytics and advisory services is one of the biggest challenges to overcome in this industry.
  • Buildings and HVAC systems within them are dispersed over geographic expanses. Remotely connecting such sites to collect data in an inexpensive manner with an industrialized approach is a problem which many companies are trying to address.
  • Companies are also exploring effective solutions to store and manage the huge volume of data once collected.
  • Variation of scope, design and configuration leads to differences in HVAC systems that increase the challenges of data normalization.
  • After one is able to capture and curate the data, the next challenge of developing flexible platforms is to create visualization reports and dashboards on the fly to meet individual needs.
  • Deploy robust analytical platforms for predictive modeling, fault detection and diagnostics, and create economic value from Big Data and Analytics.

The core technology for HVAC has been stable over many years, but Big Data and Analytics looks set to bring the HVAC industry through a paradigm shift, especially with respect to how buildings are viewed, operated, managed and serviced. Big Data and Analytics is here and now. People already have started investigating the possibilities and investing in future. This is an area which needs to be understood well, but acted upon quickly.

More on this topic