Apply AI HVAC optimization to chillers, pumps, cooling towers, and selected setpoints as one coordinated commercial and industrial energy-saving system.
Low-disruption
Integration
Maintained
Operator authority
The control problem
Commercial HVAC is still a massive, under-optimized control problem. Most plants already have BMS visibility, trends, alarms, and operator screens; the persistent waste comes from static setpoints, poor sequencing, and subsystems that are not coordinated through supervised control against real load, weather, and equipment efficiency.
What we optimize
The solution is designed around the operating system you already have in the field, then extended with supervisory logic that can coordinate the assets that matter most.
A supervisory layer evaluates live conditions and recommends approved central-plant and major air-side setpoints without replacing existing controls.
ClimaMind focuses on the assets that drive plant energy: chillers, pumps, towers, AHUs, PAUs, heat exchangers, boilers, and selected system setpoints.
The BMS remains the operating interface and native control layer. Operators keep visibility, overrides, and local authority.
Adjacent systems stay outside the optimization boundary unless they are explicitly mapped into the deployment scope.
How savings happen
The economic value comes from removing unnecessary operating buffer while making control decisions more precise, bounded, and measurable.
Buildings often run with extra safety margin because schedules and rules are coarse. ClimaMind trims avoidable overcooling, overheating, ventilation, and simultaneous heating/cooling while staying inside comfort and equipment constraints.
Instead of broad settings applied everywhere, ClimaMind adjusts by equipment state, zone demand, time window, weather forecast, and occupancy pattern so each condition receives only the energy it needs.
When the building can coast safely, ClimaMind can reduce runtime or shift load away from costly periods without changing the existing BMS control stack.
Secure intervention
Before AI writes to approved control points, ClimaMind can start in advisory mode, define hard limits, apply bounded tuning steps, and preserve a clear path back to native site control.
ClimaMind can begin by recommending setpoint changes without writing to live control points.
Commands outside authorized equipment limits are rejected automatically.
Gradual, rate-limited adjustments avoid aggressive transients.
Operators can return control to the native site path while fail-hold preserves safe operation.
Audit Log
For approved deployments, ClimaMind connects the operating context, control decision, and operator authority in one reviewable record—so teams can understand what happened, why it happened, and what followed.
01
Timestamped BMS telemetry, relevant weather and schedule context, equipment availability, and the data-quality state used for the reviewable control cycle.
02
The recommended control move, applicable logic or model version, expected operating effect, and the approved limits checked before a command can proceed.
03
The proposed or issued command, acknowledgement or rejection status, plus the approvals, manual overrides, and return-to-native-control events that determine authority.
04
Alert conditions, acknowledgement and assignment history, operator notes, and resolution or escalation steps that explain how an exception was handled.
How value is measured
Measurement fits the site's operating constraints, metering, and commercial boundary so results can be reviewed by operators and sponsors.
The measurement package depends on site conditions, available data, and the agreed commercial boundary.
01
Choose the lightest defensible method for the commercial question: alternating days, a baseline model, or settlement-grade IPMVP when required.
01
Used where site operations allow a controlled comparison between baseline and optimized operation.
02
Used when direct A/B cycling is not practical and the site needs a baseline-based reporting path.
03
A settlement-grade option for contracts, incentives, financing, or third-party review; requires a pre-agreed M&V Plan and complete data access.
02
Once the method is set, the evidence package tracks outcomes, comparable operating conditions, and control traceability.
Plant meters, utility interval data, or BMS energy points define the energy result when they are available.
Tariffs, demand charges, time-of-use periods, and billing assumptions translate energy movement into cost impact.
Weather, load proxies, schedules, occupancy, and equipment availability explain whether periods are comparable.
Keeps the setpoint and operating record available so measured outcomes can be tied back to runtime decisions.
Commercial model
Performance-based pricing is the default. Total HVAC energy-cost share is available when customers need a fixed budget or procurement-friendly structure.
Default
Payment aligns to measured savings when the site boundary and reporting data support it.
Long-term / special requirements
Available when a total HVAC energy-cost basis fits a long-term customer relationship or procurement requirement better.
Where it fits
The control layer stays conceptually consistent even when the facilities and operating constraints change.
Central cooling systems serving multi-tenant or comfort-sensitive operations.
Multi-asset environments with varying loads across buildings or thermal loops.
Reliability-sensitive environments that need explicit oversight and fallback paths.
Cooling infrastructure where operator control boundaries and measurable outcomes both matter.
Deployment-readiness briefing
Review control scope, guardrails, and measurement approach with our team before any closed-loop rollout.