Overcoming the Building Big Data Challenge

As the cost of sensors has dropped and the amount of computational power and data storage has increased, the amount of building data available has increased considerably. Moore’s law observes that overall computer processing power doubles every 2 years. There is value in having large amounts of data and processing power, but how to manage and get useful information out of the data is a challenge. With dozens of equipment vendors, sensor manufacturers, and building automation systems, data sets often come back with various formats, naming conventions, and syntaxes. While the potential for using large data sets to improve energy and operational efficiencies is significant, the difficulties of organizing and understanding the data is still being overcome.

Government Initiatives

The Standard Energy Efficiency Data (SEED) platform was built by the U.S. Department of Energy (DOE) to gather, sort and analyze complex building data. The software helps users combine, organize, and authenticate data from multiple sources. The data sets can then be shared among platform users with the intent of developing methods to calculate and demonstrate economic and environmental benefits of energy efficiency initiatives. The SEED platform is also useful for generating benchmarks and displaying key metrics for facilities. In late 2015, the DOE extended the SEED platform to the SEED Collaborative to encourage the partnership and participation of states and local governments. The collaborative includes several major cities, including New York and Atlanta, in addition to the California Energy Commission, and Washington, D.C. to name a few. The strategic partnerships between the SEED Collaborative and states and cities will help the platform reach additional software developers and service providers to improve interoperability and data transparency.

Open Source Efforts

Project Haystack is a corporation formed with the intent of developing semantic modeling solutions for smart device data. The open source effort includes a variety of automation software companies and associations working together to map existing building data models and taxonomies, with the aim to improve the cost-effectiveness of creating value from the data sets. Project Haystack intends to standardize semantic models for all building systems and other intelligent devices. If facility owners, systems integrators, and software providers use common naming conventions, they can expect easier integration for value-added services that facility data can provide.