Considered installation
Discreet monitoring at roof cavities, timber framing, sills and other relevant locations. Confirm attachment methods and suitability for each structure.
The intelligence at the heart of KyoSensa.
The AI-powered KyoSensa Insight Engine uses sophisticated machine learning across an interconnected intelligence architecture to understand how buildings behave, respond and recover over time. It brings together sensor evidence, Building Fingerprint, environmental context, spatial relationships and human expertise to identify meaningful change and support earlier, better-informed decisions.
Talk to KyoSensa01 / MEASURED EVIDENCE
Technology that sees what the human eye cannot.
Wireless sensors measure timber moisture, relative humidity and temperature, and those measurements are the authority. KyoSensa calculates context around them: rainfall and weather from external data, local seismic events and derived indicators such as mould risk. Context frames a question; it does not establish a cause on its own.
Weather & seismic context
Rainfall and local events help frame the questions that a change in readings raises.
ExploreHuman judgement
Review context and evidence before consequential decisions. Monitoring supports inspection and professional judgement.
02 / FROM SENSOR EVIDENCE TO BUILDING INTELLIGENCEFrom sensor evidence
From sensor evidence
to building intelligence.
Every KyoSensa Insight follows the same path from measurement to understanding, with AI-assisted risk detection and decision support turning readings into connected evidence that people can review.
- 01
Sense
Continuous evidence from discreet monitoring throughout the building.
- 02
Learn
AI and sophisticated machine learning build an understanding of how the individual building behaves.
- 03
Compare
Changing conditions are interpreted against the building’s history, environment and related monitoring points.
- 04
Detect
The Insight Engine identifies meaningful deviations, unusual relationships and emerging patterns.
- 05
Explain
KyoSensa connects the evidence into understandable context so people can review it and make better-informed decisions.
KYOSENSA INSIGHT ENGINE
Intelligence that works together.
KyoSensa looks at each signal in the context of the whole building. Its AI-powered interconnected intelligence architecture brings together multiple forms of building intelligence, allowing one source of evidence to be understood alongside the building’s history, environmental conditions, related areas and recovery behaviour.
KyoSensa combines building-specific data with environmental and geospatial context to better understand how each building responds to its surroundings.
A change at one monitoring point can mean much more when KyoSensa can compare it with what is happening elsewhere in the building and how the building has behaved before.
AI-POWERED INTERCONNECTED INTELLIGENCE ARCHITECTURE
Data tells you what happened. Intelligence helps you understand what it means.
Multiple specialised intelligence capabilities work together through the KyoSensa Insight Engine, with the Building Fingerprint at the centre.
What the Insight Engine does, continuously
- SENSESignals arrive
- LEARNPatterns take shape
- PREDICTHow the building should respond
- COMPAREExpected vs observed
- DETECTMeaningful change
- VALIDATEWhat later evidence showed
- EXPLAINEvidence you can act on
- CONTINUE LEARNING
What KyoSensa considers
- Sensor Data
- Rainfall Data
- Temperature & Humidity
- Seismic Events
- Building History
- Maintenance & Inspections
- Proximity Awareness
- Past Events & Recovery
Specialised intelligence capabilities, connected around Building Fingerprint
01
Expected Behaviour
Learns how the individual building normally behaves over time.
02
Environmental Response
Understands how the building responds to rainfall, humidity, temperature and changing environmental conditions.
03
Anomaly & Deviation Detection
Identifies behaviour that differs meaningfully from what is expected.
04
Recovery Modelling
Understands how the building responds and recovers following moisture and environmental events and helps identify when recovery is slower or different from normal.
05
Spatial & Zone Intelligence
Looks beyond individual monitoring points to understand relationships between areas of the building.
06
Mould Risk Intelligence
Evaluates building and environmental conditions associated with elevated mould risk.
07
Post-Seismic Intelligence
Adds context following significant seismic activity and helps identify behavioural changes that deserve closer attention.
08
Sensor Health & Data Quality
Helps distinguish genuine changes in the building from instrumentation or data-quality problems.
Multiple interconnected intelligence capabilities bring the evidence together around your building. Building Fingerprint provides the context; human review turns findings into considered next steps.
03 / BUILDING FINGERPRINT
How a building’s fingerprint is built.
Building Fingerprint is learned from the building’s own readings after installation, for each structure on the site. It describes the building’s usual moisture levels, its response to rain, humidity and temperature, its recovery after events and the relationships between monitored areas. New readings are compared with that established behaviour as well as with the monitored thresholds, so a change is understood in the context of the building itself, not by a threshold alone.
KyoSensa begins with broader building intelligence and becomes increasingly specific to your building as it learns. It is designed so that every monitored building can contribute to better KyoSensa models, while each customer’s raw data stays isolated and each building keeps its own Building Fingerprint.
A fingerprint is not trusted simply because time has passed. It is checked against readings it has not learned from before it is used for comparison, its confidence is stated, and it is learned again when the building changes, for example after major work. Persistent moisture risks stay visible even when they have become part of the normal pattern.
AI proposes. Raw evidence decides. Forecasts are recorded before their outcomes are known, and KyoSensa validates forecasts against what actually happened. A change to how KyoSensa forecasts is adopted only after a person at KyoSensa has reviewed the evidence.
Judge change in context
A departure is weighed by how long it lasts, where it appears and what the weather was doing.
Keep persistent risk visible
An elevated reading does not become safe simply because it has been present for a long time. Material risk and building history must be considered together.
Local or shared
Monitoring points are grouped by the structures and designated areas they belong to. Comparing them helps distinguish a local change from a building-wide response.
RECOVERY MODELLING
How recovery is compared.
After each significant rain event, KyoSensa measures how the building responded and how long it took to return toward its usual level. Each recovery is compared with earlier, comparable events at the same location; events affected by other factors, such as a post-earthquake watch, are kept separate rather than mixed in.
Until enough comparable events have been observed, KyoSensa says so rather than guessing. A slower or different recovery is a reason to look more closely, not a conclusion about its cause.
The building returns toward normal.
EventMoisture responseReturn toward normal
The recovery is prolonged or unusual.
EventMoisture responseProlonged or unusual recovery
POST-SEISMIC WATCHHow a building is watched
How a building is watched
after an earthquake.
When officially observed shaking near a building is strong enough, KyoSensa opens a Post-Seismic Watch. The watch continues through the next significant rain, because that is often when a disturbed roof or seal first lets water in.
The building’s response to that rain is compared with how it responded before the earthquake. Missing data is never treated as stability, and a single spike is not treated as change. A difference is passed for review as a reason to inspect, not as a diagnosis of damage.
Explore post-seismic monitoringTHE KYOTO PILOTDetected early.
Detected early.
Corrected before damage.
At the pilot timber property, the concealed subfloor showed wood moisture around 23–24% while the kitchen directly above stayed around 13–14%. Subfloor relative humidity remained near 85–95% over an extended period.
The contrast led to a targeted investigation that found a moisture issue, which was corrected before damage occurred. A visual inspection alone would not have revealed a concern, and without continuous monitoring the issue could have stayed hidden for a considerable time.
Reported wood moisture% WMC
Concealed subfloor23–24%
Alert at detection
These readings prompted investigation. The moisture issue was subsequently found and corrected before damage occurred.
Kitchen directly above13–14%
Within normal parameters
The kitchen stayed within normal parameters during this monitoring period, despite the issue beneath it.
0%10%20%30%
Historical pilot readings · 29 June–9 August 2026. These show the conditions that prompted investigation, not post-repair readings. These values are not universal safety thresholds.
04 / PRIVACY BY DESIGN
The building, not the occupants.
No cameras, microphones or audio-visual occupant surveillance are used for environmental monitoring. The value comes from physical measurements interpreted over time and in context.
YOUR NEXT STEPStart with your building.
Start with your building.
Or your entire portfolio.
Start a conversation
Direct owners, management teams and partners welcome.