Introduction - Benefits - Applications
Project Haystack is an open source initiative to streamline working with data from the Internet of Things. We standardize semantic data models and web services with the goal of making it easier to unlock value from the vast quantity of data being generated by the smart devices that now permeate our homes, buildings, factories, and cities. Applications include automation, control, energy, HVAC, lighting, and other building related systems.
"The adoption of Project Haystack's open standard for building related data is a game-changer for all those involved in managing building portfolios - instead of every building using different data labeling, as has been the norm until now, a common, standardized approach hugely simplifies comparative analytics, diagnostic processes, and performance optimization." Chris Irwin - VP Sales EMEA & VP Marketing
Macro trends in technology are making it increasingly cost effective to instrument and collect data about the operations and energy usage of buildings. We are now awash in data and the new problem is how to make sense of it. Today most operational data has poor semantic modeling and requires a manual, labor intensive process to "map" the data before value creation can begin. Pragmatic use of naming conventions and taxonomies can make it more cost effective to analyze, visualize, and derive value from our operational data.
We are very used to the term “data” to describe numeric information; in the context of buildings temperature and energy usage are the most important such parameters, needed to assess comfort and efficiency respectively. We all understand that temperature is measured in degF or degC and energy is measured in kW or BTU, but we are less familiar with the concept of “metadata” which is the term used to describe collectively all the labels we add to data to define its context and meaning. To say the temperature is “21” is meaningless until one adds the labels (or “tags”), stating the units o measurement (deg C or F), the type of sensor the temperature data is coming from (e.g. room, duct or pipe), the location of the sensor (where is it in the building?) and what equipment is the data being fed to so as to provide control. This metadata is also known as semantic tagging and is fundamental to enabling computer software applications to be able to properly understand the data they are required to process. Project haystack has established a dictionary of definitions for such metadata which establishes a common vocabulary. Without semantic tagging automated analytics processes are impossible. Unfortunately, until recent years adding such semantic tagging has not been supported by the BAS or BMS software typically used to manage buildings, and add the metadata afterwards to enable analytics has been a relatively costly exercise. What is needed is a new generation of software that natively adds the tagging metadata as part of the normal set-up processes when configuring a project.
Once data is tagged it can be analyzed more easily but the analytics software still needs to understand how the data is structured or classified to remove potential ambiguities. To illustrate the importance of context, let’s consider the word “hot” . If I am on a beach in the sun then one could easily understand what I mean if I say “this is very hot”, but in a different context, such as in an Indian restaurant while eating a curry “this is hot” has a very different meaning. In a building-related context, the word “On” will have different significance depending on whether I am speaking about a heating system, a lighting circuit, or a fire alarm. So to properly define the data related to a building a methodology for creating both relationship structure and classification is required. These are technically referred to as an ontology and taxonomy.
A popular implementation of open-source Project Haystack is nHaystack. This module enables Tridium Niagara stations (JACE and WebSupervisor) to serve Haystack data via a RESTful protocol. Using nHaystack, applications receive data complete with essential metadata descriptors. With a Haystack-tagged system, you can define tags once and realize value over and over again. nHaystack makes it easier for system integrators to add meaning to the Java-based Niagara component model.
There are numerous other implementations being worked on by the Project-Haystack community too. For example, there are groups working in C++ and DART. All share the common goal of making it easier to unlock value from the vast quantity of data being generated by the smart devices that permeate our homes, buildings, factories, and cities. They realize that the way forward starts with defining tags for the most common types of components and uses. The resulting self-describing models will pay off at the device, equipment and building-levels, as well as when they are shared by various applications for visualization, control, fault detection, analytics, maintenance ticketing, room scheduling, etc.
The real-world value of using tagging and data modeling can be seen through these examples – or “Payoffs.”
Building data needs to to be structured and fully defined for automation of optimization and analytics. Project Haystack is the most widely deployed metadata standard for buildings and offers great potential for the future.
Stay informed about our latest news and industry trends