I’m a paragraph. Double click here or click Edit Text to add some text of your own or to change the font. This is the place for you to tell your site visitors a little bit about you and your services.
I’m a paragraph. Double click here or click Edit Text to add some text of your own or to change the font. This is the place for you to tell your site visitors a little bit about you and your services.
I’m a paragraph. Double click here or click Edit Text to add some text of your own or to change the font. This is the place for you to tell your site visitors a little bit about you and your services.
Per Model Cost
Elipsa Platform - Frequently Asked Questions From Users
You asked, and we answered. Here are the answers to the most frequently asked questions from the Elipsa User Community. Learn how Elipsa seamlessly integrates AI with your BMS to take buildings from smart to intelligent.
Don't see the answer you are looking for? Let us answer you live. To book a meeting with us, click here
Elipsa is a secure cloud solution, but on-premises offerings are in development
Elipsa can use whatever protocol, connected device, data analytics software, or cloud hosting environment you have, but for reasons of simplicity, we prefer MQTT. Click here to learn why.
Elipsa uses best in class security standards for data in transit including fully encrypted MQTT connectivity with username and password authentication. For data at rest, Elipsa is hosted on AWS taking advantage of their security suite. For more information or details regarding our security standards please contact us at info@elipsa.ai
Elipsa simplifies AI for customers by automating the heavy lifting from the very start. We realized early on that very few buildings come with neatly tagged data. This is why we have developed an AI-based auto tagging tool to help any organization get started.
How it works:
Download your current data schema from your BMS into a CSV file
Drop the CSV in the Elipsa auto-tagging tool
Elipsa uses AI and machine learning to map your current data schema to our standardized Haystack ontology with an average of 90+% accuracy.
Users are able to review, amend, and approve the predicted tags as needed; cutting the data tagging process from weeks or months to as little as a few minutes or hours.
Once tagged the building data flows seamlessly into Elipsa’s platform and/or other third-party systems. It’s your data!
To learn more, read our post “How-To on Haystack Auto-Tagging”
Elipsa integrates seamlessly with your existing infrastructure and BMS. No additional equipment or software is required. Elipsa also offers a range of pre-built connectors to partner systems, such as Tridium's Niagara Framework. In addition, Elipsa integrates seamlessly with any third-party system capable of transmitting data to our MQTT broker. No heavy integration work required.
Smart Buildings utilize IoT to connect to a reporting system, typically a building management system. Smart buildings provide historical context and point-in-time reporting.
Intelligent Buildings take this a step further by applying machine learning and AI to this data to predict the current and future state of a buildings critical systems. These predictive insights enable organizations to proactively plan for suboptimal scenarios and address them before they occur while reducing system ware and wasted resources.
We provide a more in-depth explanation in the post “Why settle for "smart buildings"? "Intelligent" Building Monitoring is here today!”
Elipsa provides
Auto-tagging of building data
AI-based Fault Detection and Diagnostics,
Digital Twins/Scenario Planning,
Energy Monitoring and Optimization
Energy Measurement and Verification
FDD (or Predictive Maintenance, as it is known in some industries) allows organizations to address mechanical issues and inefficiencies proactively by alerting operators to potential issues before they occur.
This visibility into the current and future state of a building’s equipment reduces wasted resources and energy costs while ensuring safe and productive built environments.
To learn more about the benefits of AI-based FDD and intelligent monitoring read “From Smart to Brilliant: The Power Intelligent Building Monitoring”
In a study conducted by Berkley National Laboratory that looked at 550 buildings utilizing FDD, the average savings was 8%. The study results can be found here: “Building fault detection and diagnostics: Achieved savings, and methods to evaluate algorithm performance”
If you have historical data, you can start monitoring your equipment with Elipsa in just a few hours. Simply upload a CSV file containing your historical data and Elipsa will automatically select and train the models that fit your equipment. Once you connect your live data via MQTT, monitoring begins automatically, providing actionable insights. It's just that easy.
Absolutely, even without historical data to train the AI models, the platform automatically deploys rule-based monitoring and alerting derived from ASHRAE 36 standards. In parallel, while your system streams data to the platform via MQTT, we are building the historical dataset in the background to train your AI-based FDD models. The initial AI training period typically takes two weeks and then the models continue to retrain on a monthly basis. Once the training period concludes, Elipsa begins using a combination of AI and rules-based monitoring coupled with advanced analytical tools to monitor your operations going forward.
You can build an initial model with as little as 2 weeks worth of data knowing it will retrain and get better over time. If you do have historical datasets, we recommend 3 months worth of data for FDD model building. Other types of models such as predicting energy should utilize longer data sets of at least a year to account for seasonality.
The elipsa platform for FDD learns what normal operation of equipment is to monitor for abnormalities over time. Most of the time equipment that you look to monitor is already running sub-optimally. So, are you prevented from creating an effective AI model if you are using data that is not fully perfect and normal? No! Elipsa's proprietary AI algorithms are able to find normal even within abnormal data. As a result, you can upload data from equipment with histories of various levels of health and still be able to automatically determine what the effective baseline is to monitor from.
No, Elipsa requires no code to be written by the end user in order to implement. The user does not even need to select the data points or components that they look to monitor for. Elipsa has created proprietary templates for a wide variety of equipment types. Once you utilize our auto-tagging features to specify the type of equipment and points, Elipsa automatically builds AI models to monitor for the specific components that you have enough points for. The result is an automated implementation tailored to each and every piece of your equipment without any heavy lift. For more details, check out our article "The How-To on FDD Templates and Dashboards"
Yes, Elipsa bases portions of its rules-based monitoring on ASHRAE 36’s fault detection guidelines. This provides users with both AI-based predictions and standardized, industry-accepted rules ensuring safe and efficient building operations. To learn more about how we incorporate ASHRAE 36 read: “ASHRAE 36 Fault Detection Made Simple”
Elipsa was built to accommodate large-scale building portfolios but is just as valuable in a single building.
Users can easily configure multiple sites and then automatically configure the equipment and monitor for individual sites using Elipsa's auto-tagging.
Elipsa provides standard energy reports that pertain to many of the elements found in the various ESG reporting frameworks but are not based on any specific organizations standards; if there is a particular standard that a customer is interested in implementing we can produce customized reporting through our professional services group.
Yes, the Elipsa platform is flexible in the data it can ingest as well as the AI-based outputs. If weather is an available data point, the system will incorporate that data into the predictions. This type of analysis is typically done using Elipsa’s digital twin and scenario planning tools.
If the equipment's data flows to the BMS, or is capable of being transmitted via MQTT, Elipsa can track and report on it.
Elipsa was designed with the end users in mind: operations, engineering, finance, ESG teams can quickly access the insights they need with minimal training. From easy-to-understand visualizations and alerts to downloadable reports and AI-generated suggestions, Elipsa provides historical content and predictive insights that can be leveraged across the organization.
Yes, we are happy to set up a demo using your real-world data. Just contact us at demo@elipsa.ai to get started.
Elipsa has a software-as-a-service business model. The pricing model consists of an initial set-up fee and annual service fees based on the number of active data points in the system and the number of sites monitored. Please contact Elipsa here for more information.