We built an AI energy platform that connects a resort group's building sensors to its PMS, so heating, cooling and lighting adjust around guest arrivals and departures. Energy use fell 19%.
Client: Resolve ArchiveIndustry: HospitalityService: AI & Automation
Overview
Client
Resolve Archive
Global portfolio of luxury resorts
Engagement snapshot
The challenge
The resorts' older building management systems (BMS) only reacted. They used manual set-points or basic motion sensors that could not tell a quick staff visit from a guest in the room. To make sure suites felt right the moment a guest walked in, air conditioning ran at full power in empty rooms, wasting millions in seasonal operating costs.
The front-desk software and the building equipment never shared data, so the HVAC had no idea when a guest was actually due to arrive.
Big spikes in energy use at check-in times triggered high-demand surcharges from local utilities.
Facilities teams had to adjust the older systems by hand all the time, which led to mistakes and waste.
Without detailed energy data, the resorts could not achieve the verified LEED or ISO certifications that professional investors expected.
The solution
We installed gateways that convert data from different sensor protocols (BACnet, Modbus, Z-Wave) into one event stream.
An AI assistant reads check-in and check-out events from the PMS and pre-cools a room before arrival, or lets its temperature drift when it is empty.
Every set-point change and energy saving is logged, giving an auditable trail for ESG and carbon reporting.
Key features
AI reads check-ins from the PMS to pre-cool rooms before arrival and let empty rooms drift safely.
Business impact
19% less energy used
Predictive maintenance flags HVAC motor faults weeks ahead, and every energy saving is logged for ESG reporting.
Business impact
Uptime from 95% to 99.2%
For each empty room, the AI works out how far the temperature can drift to save energy, based on how long it takes to get back to comfort.
Business impact
Zero guest complaints
Carbon savings are calculated automatically for each guest night, which meets regulatory ESG reporting rules.
Business impact
A path to LEED Gold
When utility prices jump into a high-cost tier, the network switches off non-essential loads automatically.
Business impact
19% lower bills
Before and after
Before
Fixed schedules
HVAC ran on fixed timers or basic motion sensors and ignored whether a guest was booked in.
After
Planned around arrivals
Energy use shifts with arrival times, weather forecasts and past occupancy patterns.
Before
Separate systems
The front-desk software and building hardware never talked to each other.
After
One connected platform
The PMS, IoT sensors and utility pricing stay in sync through an automatic bridge.
Before
High peak usage
Property-wide surges caused expensive spikes in utility demand charges.
After
Lower peaks
Automatic load balancing across each property cuts peak demand charges by 15%.
Results
19% Operating Cost Reduction
Predictive load management cut kilowatt-hour use across the portfolio, and energy costs fell 19%.
Before: ReactiveAfter: 19% Lower
14x Faster Response
Edge processing changes room set-points within moments of a guest checking in.
Before: 2.5sAfter: 180ms
99.2% Uptime
In 85% of cases, predictive maintenance alerts caught HVAC motor faults 3 weeks before the system failed.
Before: 95%After: 99.2%
Testimonial
“Energy use across our resorts is down 19% and peak demand charges are 15% lower. Rooms pre-cool before guests arrive, HVAC faults show up weeks ahead, uptime rose from 95% to 99.2%, and payback took 11 months.”
Patricia MorganVP of Engineering and Facilities, Resolve ArchiveKey results
Trust and control
PMS data is scrubbed of personal information. Only occupancy status and arrival times reach the IoT network.
PrivacyProcessing happens on site first, so room control keeps running even if the internet connection drops.
Edge firstThe cloud monitoring hub supports 100,000+ IoT devices across regions at the same time.
Scales with demandThe client owns 100% of the IP in the custom load-control models and gateway drivers.
100% ownedFaster delivery
About 10 weeks of setup was saved by reusing these components, and the team spent it on the room control logic.
Provided tested drivers for BACnet, Modbus and Z-Wave, which we adapted to each resort's equipment.
Passed check-in and check-out events to the AI controller, with personal details removed.
Watched sensor and HVAC health across the portfolio and raised the predictive maintenance alerts.
Tracked live utility prices and demand charges, so the system knew when to shed load.
Their older building management systems used fixed set-points and basic motion sensors that could not tell a quick staff visit from a guest in the room. To make sure suites felt right on arrival, air conditioning ran at full power in empty rooms, wasting millions in seasonal operating costs.
It reads check-in and check-out events from the PMS, along with weather forecasts and past occupancy. Empty rooms drift to an energy-saving temperature, and the AI works out when to start cooling so the room is comfortable before the guest arrives. Set-points change within 180ms at the edge.
Energy use fell 19% across the portfolio, and the system paid for itself in 11 months. Peak demand charges dropped 15%, the guest comfort score was 100%, and equipment uptime rose from 95% to 99.2%. Maintenance alerts caught 85% of HVAC motor faults 3 weeks before failure.
If your rooms are cooled whether anyone is booked in or not, we can help you link your PMS and building systems so energy follows real occupancy.