For teams in building operations and facility management, the starting situation is often similar: buildings generate very large volumes of data every day, originating from multiple sources and intended to be usable by different roles. In practice, however, this data often remains confined to dashboards, BI views, or isolated solutions—and new, situational questions quickly lead to additional analysis or reporting effort.
This is precisely where akenza’s Genio comes in: Genio is an AI-based assistant within the akenza IoT platform that answers natural-language questions based on available IoT and building data. The approach has been implemented in the facility management context together with partners—including in the collaboration between ISS and akenza.
The challenge: many operational questions—but answers are not immediately available
In building operations, situational questions arise every day—about usage, anomalies, or the condition of sensors and infrastructure. Typical examples include:
- What was the highest recorded CO₂ level this week?
- How many workstations are currently occupied?
- Which sensors will soon require a battery replacement?
- Which machines had the highest energy consumption last month?
- What is the status of all motion sensors?
The bottleneck is often the ability to answer such questions quickly and reliably, as data resides in different systems, integration is lacking, or dashboards only provide fixed views. At the same time, raw data quality is not always sufficient, and custom analyses often require specialized expertise.
This becomes particularly evident when questions refer to a specific location within a building—for example, a floor, a room, or a zone. In such cases, it must be clearly defined which sensors belong to which area to ensure that queries remain consistent (e.g., “occupancy of floor 3”).
The key question was:
How can building data be consolidated and leveraged in a way that allows operational questions to be answered in natural language—without having to build a separate analysis for each new query?
The solution: Genio as an AI assistant within the akenza environment
The solution is based on the akenza IoT platform as the data layer and Genio as an AI-based assistant within this environment. Genio is designed to process natural-language questions and derive answers from the available IoT and building data—both with regard to current conditions and historical trends.
In practice, this means that instead of creating new reports or navigating through fixed dashboard filters, a question is simply formulated in natural language. Genio then provides an answer, which can be further refined as needed—depending on the level of detail required for the situation.
The approach is designed to be role-based. An example from the Genio context: a Building Security Manager asks, “What is the health of my IoT device fleet?” and receives a direct answer on the operational status of the devices—including indications if sensors will soon require attention due to low battery levels.
The result: questions become directly derivable from data—provided context and data foundation are in place
This approach shifts access to building data away from “dashboard searching” toward direct querying: operational questions can be asked and answered in natural language—across topics such as usage, comfort, security, maintenance, or energy.
At the same time, the implementation clearly highlights what the reliability of such answers depends on:
- Data quality: If raw data is incomplete or inconsistent, the meaningfulness of the answers decreases.
- Unambiguous building mapping: Whenever questions refer to floors, rooms, or zones, the assignment of sensor data to these areas must be modelled consistently and correctly.
- Performance: For large-scale queries, appropriate aggregations and database capabilities are required.
For operations, this means concretely: akenza provides the foundation with the akenza IoT platform to consolidate and make data sources usable—and Genio builds on top of this to enable natural-language queries in day-to-day operations, as long as the data foundation and building mapping are reliable.