The building industry terms we need to define – and you need to know
Digital twins, data lakes, semantic tagging. The list of technical jargon in our building industry goes on and on. I hear so many of these terms being used every day, but often their precise meanings differ depending on who you’re speaking to. For there to be any value in using our industry-specific lingo, we need to have some agreement on their definitions. I’ve compiled a glossary in an effort to bring some focus and unity to the language we use when discussing digitalization in buildings.
Change of Value
Collecting data when it changes a set state, for example a light switch turning from on to off, or a temperature sensor changing by more than 0.1 degrees. This is also known as COV data.
Gathering data from building systems and importantly, having the means to analyze the data that is being generated. When connecting systems, the parameters of what is required need to be defined at the outset (see fig. 1 below).
Once the type of data required has been identified, it needs to be collected (see fig 2. below). The most common methods of collecting data are BACnet, Rest (Representational State Transfer), API (Application Programming Interface), Modbus and MQTT.
By analyzing and processing the data that is gathered over time, a building’s self-learning can take place and new values, or issues, can be found.
A place to store a large collection of raw data, that can be structured or unstructured. Large volumes of diverse data from a variety of sources can be analyzed. A poorly organized data lake is known as a data swamp.
The outputs required from the building. The framework is constituted by the digital solutions needed and this differs from building to building. For example, do you need data analytics, people tracking, or meeting room booking?
The data required from every installed asset in the building, and how that information is structured. For example, the asset’s install date, make, model, size and documentation.
Creating a digital replica of the physical building. This does not mean just geometrically, but also including the information of the building, and making it available digitally. Many people think of a digital twin as just a three-dimensional model, but that’s only one part. Capturing the data is integral to creating a digital twin.
Taking physical information and making it digital. The building and its technology work seamlessly together to provide benefits for the inhabitants or employees of the building.
Fault Detection Diagnostics (FDD)
An engineering tool to identify faults or pieces of equipment not running to specific, pre-defined variables.
Collecting data from the building at set periods of time. The data is normally measured in regular intervals such as one five, ten or 15 minutes, depending on the system and requirements. Using an interval scale shows the exact difference in value and can document any events such as interruptions or swells.
Change of Value Data
Collecting data from the building when it changes state. This is important for pieces equipment that switch on and off, as it allows start and stop times to be tied to specific loads and optimization can occur.
Using data to predict when assets might fail and preventing the failure by prioritizing and directing labour to the most important issues and performing maintenance. Predictive maintenance can be used when traditional task-based maintenance is replaced using analytics, data monitoring and fault detection. It can also refer to predicting failure on equipment using a time-based analysis or regression model.
Schedule-based maintenance that is designed to increase the life cycle of equipment. By regularly performing maintenance on an asset, the likelihood of it failing lessens. This maintenance strategy falls between reactive and predictive maintenance in terms of complexity.
Repairing assets when they have broken down, in order to return them to their normal operating condition. Reactive maintenance has a place in many maintenance strategies but should not be the only strategy implemented.
Applying meta data makes it easier to analyze and utilize data once collected. By implementing a standard tagging protocol, it becomes easier and more cost effective to collect data from building systems and its associated assets and devices.
Language is crucial to this increasingly inter-connected world and standard phrasing is key to allow for clear communications, as well as future developments and innovations. Here I’ve listed the terminology I hear being used most frequently when discussing building technology and asset performance. What are the terms that you might define differently? And are there any words and phrases commonly used in the industry that you think should be included?