Other industries · Building operations
Smart Building, Old Habits: Fix the Operating Model
By the humblx team7 min read
In short
Smart building operations depend less on sensors than on the operating model behind them: who owns each signal, how fast they must act and how closure is proved. Triage alarms and requests, name zone and system owners, route everything through one desk and require timestamped proof before a ticket closes.
- Most smart buildings invest in signals, improvise workflows and skip evidence.
- Triage every alarm and request into critical, comfort, trend, informational or nuisance.
- Give every floor a zone owner and every major system a system owner.
- A ticket is not closed until there is a timestamp and a photo or reading.
- A light weekly rhythm keeps the model alive after the launch enthusiasm fades.
Plenty of Indian office towers marketed as 'smart' share an awkward secret. The building management system can see every chiller valve and the temperature on every floor, yet a complaint about a warm meeting room still reaches the facility team as a phone call that a supervisor jots into a register. The sensors got cleverer. The way the building is run did not. If you have paid for BMS upgrades, IoT meters and occupancy analytics and your team still spends the day firefighting, the gap is rarely technology. It is who owns each signal, what happens after it fires, and how anyone knows the loop was closed.
What is a smart building operating model?
A smart building operating model is the set of roles, routines and rules that turns building data into action. Hardware tells you something is wrong. The operating model decides who hears about it, how quickly they must respond, what counts as done and where the evidence is kept. Without one, a smart building is simply a well-instrumented building run on habit.
It helps to think in three layers:
- Signals: BMS alarms, energy meter readings, helpdesk tickets, QR complaints from occupants and inspection findings.
- Workflows: the route each signal takes to a named person, with a response time and a resolution time attached.
- Evidence: the record that proves the work happened, ideally a timestamp plus a photo or reading rather than a tick in a register.
Most buildings spend generously on the first layer, improvise the second and quietly skip the third.
Why do smart buildings still run on old habits?
Visit the facility office of a Grade A tower in Gurugram, Whitefield or Hinjewadi at half past nine on a Monday and the picture is often the same. A BMS screen shows a dozen unacknowledged alarms. A WhatsApp group buzzes with requests from the client's admin team. A supervisor holds a printed checklist from last Friday. Each is a channel, and none of them talks to the others.
The habits that survive a smart building upgrade usually look like this:
- Alarm fatigue. The BMS raises so many minor alarms that the team learns to ignore all of them, including the ones that matter.
- WhatsApp as the ticketing system. Requests arrive as messages, get a thumbs-up and disappear. Nobody can say next month how many were raised or how long they took.
- Paper preventive maintenance. AHU and DG schedules live on a wall chart, and completion is filled in after the fact, sometimes a week at a time.
- Vendor silos. The lift vendor, HVAC contractor and housekeeping agency each report in their own format, if they report at all.
- Reports built for the review meeting. The monthly MIS is stitched together by hand the night before, which says more about the supervisor's stamina than the building's health.
None of this is a failure of people. It is what happens when a building gets new instruments but keeps its old organisation chart.
Which signals should trigger work, and which should not?
The first fix is honest triage. Not every alarm deserves a work order, and not every complaint deserves the same urgency. A simple matrix makes the rules explicit and gives the team permission to ignore noise without feeling negligent.
| Signal type | Example | What happens | Illustrative target |
|---|---|---|---|
| Critical alarm | Chiller trip, DG fails to start, lift entrapment | Breakdown work order raised automatically; on-duty engineer called | Acknowledge within minutes |
| Comfort request | Meeting room too warm, washroom out of soap | Helpdesk ticket routed to the zone owner | Resolve within the hour |
| Trend deviation | A floor using noticeably more power than its usual weekday pattern | Reviewed in the weekly engineering huddle | Within the week |
| Informational | Filter pressure drop creeping towards its limit | Folded into the next planned maintenance visit | As per PPM calendar |
| Nuisance | A faulty sensor flapping on and off | Suppressed, with a separate task to repair the sensor | Within the month |
Treat these targets as placeholders. Set your own from the contract and from what occupants actually expect, then publish them. A target that lives only in the facility manager's head is not a target. Our guide to facility management SLAs and KPIs covers how to choose measures that hold up.
Who owns what in a connected building?
In a building run on habit, the facility manager owns everything and therefore nothing in particular. A working model assigns ownership by zone and by system, so every signal has exactly one name beside it.
Zone owners
A zone owner, often a soft-services supervisor, answers for everything an occupant experiences on a floor or wing: cleanliness, washroom consumables, comfort requests and small repairs. They will not repair the chiller, but they own the ticket until the person who can is engaged and the occupant has been told.
System owners
System owners are the engineers and vendors responsible for HVAC, electrical distribution, lifts, fire systems, plumbing and the DG set. Their scorecard is uptime, planned maintenance completion and repeat-breakdown rate, not the number of messages answered.
One desk for every request
A single desk, staffed or virtual, receives every signal, classifies it and routes it. This is usually where a building gains the most for the least money, because it ends the practice of occupants ringing whichever technician they happen to know. A central helpdesk is the hinge the rest of the model turns on.
How do you close the loop with proof?
A smart building produces evidence that something went wrong without anyone lifting a finger. It almost never produces evidence that something was put right. That second half has to be designed on purpose.
Closing the loop means each ticket or work order ends with three facts: who did the work, when, and what the location looked like afterwards. For cleaning and soft services that is usually a QR scan at the spot plus a photo. For engineering it means readings recorded against the asset, parts consumed and a short note on the root cause. The building maintenance checklist is a sensible starting list of what to capture.
What should the weekly operating rhythm look like?
Operating models survive on routine, not on launch-day enthusiasm. A light weekly rhythm keeps signals, people and evidence in step without adding meetings for their own sake.
- Monday huddle, twenty minutes: open critical tickets, planned maintenance due this week and anything carried over. Zone and system owners only.
- Daily walk-round: supervisors scan QR points along their route and log exceptions on the spot rather than at the end of the shift.
- Midweek comfort and consumption check: compare floor-level power use with comfort complaints. A spike in both usually points to one cause, such as a stuck damper or a schedule left on manual override.
- Friday close-out: every ticket past its target either closes or gets a written reason and a new date.
- Monthly client review: a dashboard drawn from live records, not compiled by hand, covering SLA compliance, repeat faults and maintenance completion.
Keep that monthly review honest by measuring against what the contract actually says. The SLA compliance calculator is a quick way to check the percentages vendors report.
Where does software fit, and where does it not?
Software cannot design your operating model, but it can make the model hard to ignore. The useful jobs are narrow: capture every request in one place, route it to a named owner with a clock running, and refuse to close it without proof. humblx was built around exactly that loop, first for hospital housekeeping and now for offices, airports and malls, combining facility maintenance work orders, a unified helpdesk and QR-plus-photo verification. Energy and utility management is a coming-next module, so for now most teams keep their existing BMS and meters and use the ticketing layer to turn deviations into tasks.
What software should never do is add one more screen that nobody watches. If a tool does not reduce the number of channels a supervisor checks each morning, it is part of the problem.
Action list: reset the operating model in 30 days
- List every channel through which a request or alarm arrives today, personal phones included.
- Sort signals into critical, comfort, trend, informational and nuisance, and write a response target for each.
- Name one zone owner per floor and one system owner per major system.
- Route all occupant requests through a single desk and retire the informal channels.
- Require proof of closure: QR scan, timestamp, and a photo or reading.
- Run the weekly rhythm for four full weeks before judging the result.
Frequently asked questions
What is a smart building operating model?
It is the combination of roles, routines and rules that converts building data into completed work. It defines which alarms and requests create tasks, who owns each one, the response and resolution times, and what evidence proves the job was done. Sensors and a BMS supply the data; the operating model decides what happens next and makes it visible.
Why do smart buildings still have so many complaints?
Usually because the technology upgrade stopped at detection. Alarms and requests still flow through phone calls, WhatsApp groups and paper registers, so nobody owns them end to end. Alarm fatigue sets in, vendors work in silos and closure is rarely verified. Fixing ownership, routing and proof of closure tends to reduce complaints faster than adding more sensors.
How should BMS alarms be prioritised?
Group them by impact. Critical alarms such as a chiller trip or lift entrapment should raise a work order automatically and call an engineer. Comfort issues go to the zone owner through the helpdesk. Trend deviations are reviewed weekly, informational alarms are folded into planned maintenance, and nuisance alarms are suppressed while the faulty sensor is repaired.
Do we need to replace our BMS to fix operations?
Rarely. Most operational gaps sit between the BMS and the people who act on it. A single helpdesk, clear ownership, response targets and proof-based closure can be added around an existing BMS. Replace hardware only when it genuinely cannot provide the signals you need, and only after the operating routines are working.