Key Data Tagging Best Practices for Smart Buildings

Smart buildings require data to operate at their most efficient. The more data gathered, the more efficient buildings will become. Greater efficiency within built environments leads to higher rents (from 3-7% greater) and higher sales prices (16% more on average) when these assets are sold. In smart buildings, Internet of Things (IoT) technology like smart sensors and controllers help gather, integrate, and evaluate data with the help of advanced analytics software. However, this collected data, which falls under the “big data” umbrella, requires standardization to turn it into productive data. 

Regardless of its origin or purpose, proper management of data requires it to be collected and translated to make it useful. At its core, building data is really no different than other big data, which benefits from specific organizational strategies. Data tagging, also called data modeling, helps organize building data based on identifying tags, helping to translate the raw data into a usable form. Using data tagging best practices, system integrators and other stakeholders will be able to organize data to make it productive while also ensuring interoperability between a building’s systems and equipment. 

The Importance of Data for Smart Buildings

In a sense, data tagging is very much like creating a virtual library out of a pile of books. In that respect, data tagging is really no different from the “ancient” card cataloging systems found in 20th-century libraries, which provided a sample of data about each book. In fact, these cards accomplished a very similar task to metadata, essentially providing data about data. 

As buildings become increasingly automated and digitized throughout their lifecycle, the amount of building data available to stakeholders has risen dramatically. This collected data also needs to be utilized by a building’s various systems. However, interoperability is often hampered by the inability of different smart systems and devices to communicate with each other. This limits scalability when developing strategies to increase efficiency and productivity. 

Why Data Tagging Is So Important

Each dollar invested in energy efficiency saves over two dollars on the cost of supplying energy. A simple yet effective way to improve energy efficiency involves organizing building data. Tagging best practices contribute greatly to this, allowing the building automation system (BAS) tasked with collecting and storing data to retrieve and use it more easily. 

The importance of data tagging best practices has only increased with the integration of microcontrollers, sensors, networking, connectivity, and other technology within built environments. Database management relies on metadata to more easily interpret data within a dataset and apply it to real-world situations. 

Data Tagging Best Practices

Basic data tagging best practices involve:

• Don’t recreate the wheel. Use existing data modeling standards and document where any deviations or extensions exist.
• Use data modeling tools and processes that ensure tags are added correctly and adhere to the adopted standard as well as automatically update tagging in the case of required changes to the data model. 
• Refrain from using too many tags, and improve efficiency by pruning extraneous tags. 
• When tagging systems that are outside the scope of an existing standard, limit the need for new tags by choosing broad terms to cover different applications of the same concept. 
• Take care in applying  relational tags to avoid conflicts and inconsistencies. 
• Don’t go it alone. There is an extensive network of data modeling experts with specific experience in your domain.

Disparate data sources, changing personnel, insufficient structure within the dataset, and communication with legacy systems are all challenges when applying a data model. Following a clear set of best practices will help mitigate these obstacles. 

Organize Data

Data tagging best practices involve using fewer, better-quality tags that allow relationships between data to become easier to recognize. The depth of the data tagging should take into account context and how stakeholders will use the data. 

To improve the organization of data, best practices include: 

• Ensure tags are independently queryable, e.g. one can easily obtain a list of all sensors without knowing the exact sensors that are in the building.
• Include metadata on key assets that assist in troubleshooting. This includes metadata such as equipment make and model, floor area of rooms, or building primary use type.
• Apply relational tagging in depth to ensure a complete understanding of the interconnected nature of the data, e.g. sensors connected to controllers, controllers connected to equipment, equipment located in a room but serving another room, equipment receiving air from one equipment and water from another, or rooms on a specific floor and occupied by a specific tenant.
• Keep datasets clean by fixing or removing corrupted, duplicate, incorrect, partial, or wrongly formatted data. 

Once tagging is complete, data should be organized and easy to access for a variety of use cases. This ensures the benefits of the data model will be realized now and in the future as use-cases are sure to change and evolve over time. 

Ensure Interoperability

A primary objective for intelligent built environments includes the interoperability between devices, systems, and the software that supports them. This requires that they be able to communicate through the same communication protocol, or that interactions between them are translated via communication gateways. This often gets messy, especially in retrofitted environments where systems and equipment were not designed to interconnect. 

Though a significant problem in existing structures, it is also an issue that developers should consider for new buildings. Simplifying interaction between disparate systems and software has increasingly involved using standardized communication protocols. For large commercial buildings, stakeholders have moved towards using standard protocols like BACnet and Modbus for their BAS. 

Standardizing communications facilitates integration, though the difficulties inherent in mapping metadata in a legacy BAS still take considerable time and money. Because of the cost and complexity inherent in integrating technology into built environments, utilizing integration software to streamline this process makes sense. It may even be worthwhile to consider bringing in a master systems integrator to assist with optimizing and integrating building systems in order to increase interoperability. 

Make Analytics Productive

Most organizations underutilize analytics, including among building developers, designers, owners, and managers. Harvard Business Review Analytic Services published results in 2018 from a survey of global executives, finding that only 18% of respondents felt they were getting a sufficient return from their analytics investment. 

It was discovered that:

• 54% of analytics were not integrated into workflows
• 45% had inadequate staff training for interpreting and using analytics 
• 41% used siloed analytics that produced competing results
• 38% deployed and distributed outputs from analytics ineffectively 
• 31% saw time lags that negatively affected deciding and acting upon data 
• 26% found outputs too often conflicted with the way in which operations were customarily conducted 
• 23% that noted analytics outputs were poorly presented, making them difficult to interpret

Using data tagging best practices helps improve the performance of building analytics software. 

To make analytics more productive: 

• Leverage connections and relationships to provide context wherever possible.
• Automatically validate data integrity to confirm that results are born from trustworthy sources.
• Foster feedback by encouraging collaboration and contributions from all stakeholders.
• Minimize the need for manual entry. 
• Support stakeholders by considering different concerns, needs, and perspectives, and aligning goals accordingly. 
• Understand that data need not be perfect to provide useful insights and actions. 
• Utilize standard processes when modeling and reporting data to generate actionable insights. 
• When large volumes of data are unavailable, use newer forecasting methods for simulations that require less data. 

To fully harness the power of analytics software, it is important to utilize data tagging best practices. Organizing datasets assists with collaborative engagements inherent throughout a building’s life cycle, helping to streamline processes and increasing trust between stakeholders. 

To discover what products and services Buildings IOT provides to assist with organizing datasets within built environments, contact our team of experts today.