✦ Ensmart BMS Academy Home About Us Products Solutions Case Studies eNews Blog Downloads Team Contact Design Engg Get a Demo →
Home › AI BMS Mentor › Sensors

Occupancy Sensor in BMS — Letting the Building See Who Is There

Sensors 33 views ✓ Reviewed by EnSmart engineer Updated 7 hr ago
Copied to clipboard ✓
Occupancy Sensor in BMS — Letting the Building See Who Is There — infographic

A Hyderabad IT Park, 11 PM, One Person on a Floor

Dhana is a software engineer at a Hyderabad IT park. She works late twice a week, often until 11 PM, finishing builds and pull requests before the next day's standup. The floor she sits on has 240 desks. By 9 PM, the floor is empty except for Dhana. The AHU runs at full design capacity. The lighting is at full brightness. The CO2 setpoint is calibrated for full occupancy, so the fresh-air damper is wide open. The chiller serving this zone runs as if 240 people are still working. Dhana is not unhappy with the temperature. She is, however, working under stadium lighting in a space that could comfortably be served by a single AHU at 30 percent capacity. The reverse failure also happens. Some employees come in on Saturdays to catch up. They arrive at 10 AM. The AHU schedule starts at 11 AM. They sit in a hot, stuffy floor for an hour because the building does not know they came in. Both failures share one cause: the building cannot see who is there. Every single one of these problems has one solution — occupancy sensing tied to HVAC and lighting.

Three Ways to Sense Occupancy

``` Method 1 — PIR (Passive Infrared) Detects motion through body-heat signature against background. Cheap, ubiquitous, easy to wire (typically a single DI per zone). Limitation: stationary occupants (someone sitting and reading) appear absent after the timeout period (usually 10-20 minutes). Best for: corridors, restrooms, occasional-use rooms. Method 2 — mmWave Radar Detects the chest-cavity micro-movements of breathing. Detects stationary humans even when motion-still. More accurate, slightly higher cost (10-15 percent more). Output: presence count or simple presence/absence. Best for: open-plan office floors, meeting rooms, individual cabins. Method 3 — Access Control Card Count The card-reader system already counts who entered and exited. Subtract exits from entries to get current floor count. Free if access-control is already deployed (most IT parks). Limitation: tailgating, accidental holds, unrecorded re-entries. Best for: building-level coarse occupancy. Method 4 — CO2 as Occupancy Proxy CO2 in return air rises with occupancy. Indirect but reliable for steady-state. Already required by ASHRAE 62.1 for fresh-air control. Best for: confirming occupancy in zones that lack PIR/mmWave. ``` Most modern BMS designs combine two or three methods to reduce false negatives: ``` Layer 1 — Access card count tells the floor it is occupied at all Layer 2 — Per-zone PIR or mmWave tells which zones are occupied Layer 3 — CO2 confirms steady-state for low-priority zones ```

What Changes When the Building Sees Occupancy

``` HVAC response by occupancy: Floor occupied (any zone): AHU runs normally per schedule Setpoints at occupied values (24 °C cooling typical) Floor partially occupied (single zone): AHU runs at reduced fan speed Cooling delivered to occupied zone only Other zones at unoccupied setpoints (28 °C cooling) VAV boxes in unoccupied zones close to minimum Floor unoccupied: AHU fan minimum or off (anti-mould cycle every hour) Fresh-air damper closed Setpoints at unoccupied (28-32 °C cooling) Lights off (security minimum stays on per policy) Floor unoccupied but security camera or guard motion detected: Lighting at 30-50 percent on egress path AHU minimum Logged for post-shift review ``` Lighting response is similar. Per-zone lighting on per-zone occupancy. After-hours dimming. Daylight harvesting integrated with PIR.

What Changes for Dhana

When Dhana stays late, the building sees only her zone is occupied. The AHU drops to 35 percent fan speed and serves only her zone. Lights in her aisle stay on; lights in unoccupied aisles dim to 20 percent. The chiller load drops; the energy meter records meaningfully less consumption for that night shift. Dhana is comfortable. The other 220 desks she does not occupy stop costing energy. ``` Before occupancy integration: Dhana's late shift consumed: AHU at 100 percent for 4 hours Full lighting for 4 hours Full chiller fraction for 4 hours Cost per night: heavy After occupancy integration: Dhana's late shift consumes: AHU at 35 percent for 4 hours Aisle lighting only Reduced chiller fraction Cost per night: 25-30 percent of before ``` Multiplied across a year of late-shift nights and weekend visits, the saving is substantial — and Dhana never noticed a comfort change.

What Changes for Saturday Visitors

When the first employee swipes the access card at 10 AM on Saturday, the BMS receives the swipe event. The schedule is overridden. AHUs start. Lights turn on. By the time the employee reaches their desk after a coffee, the floor is at temperature. ``` Schedule says: AHU starts at 11 AM Card swipe says: occupant entered at 10 AM BMS rule: "earliest-of card-swipe-or-schedule" wins Result: AHU starts at 10 AM, employee comfortable ``` This pattern — scheduled start with card-triggered early start — is one of the most appreciated features in any IT park.

Filtering False Occupancy

Sensors are imperfect. The BMS layer filters them: ``` Single-PIR trigger filtering: Require PIR active for at least 60 seconds before declaring occupied (filters quick-pass-through events). Stale occupancy: After 20 minutes of no PIR activity, declare unoccupied (filters frozen-PIR fault). Cross-validation: If access-card count is 0 but PIR shows occupied, log as "ghost occupancy" for investigation. Time-of-day weighting: At 03:00, weight CO2 occupancy higher than PIR (PIR may fault during low-activity hours). ``` These filters reduce false-positives and false-negatives to under 1 percent in well-tuned deployments.

Why This Matters Now

Indian commercial buildings are large, often partially occupied, and serve flexible work patterns. Hybrid work, late-shift teams, weekend support, festival weeks — every pattern of partial occupancy is now common. Schedule-only HVAC was designed for a world where everyone arrived at 9 AM and left at 6 PM, five days a week. That world is gone. Occupancy-driven HVAC is the design for the world that exists. The building should know who is there. With that knowledge, it serves the occupants well and stops serving the empty zones. Without that knowledge, it serves at full capacity all the time — comforting nobody and wasting heavily.

Related Topics

Related Topics

Related Topics

Related Topics

Was this answer helpful? ✓ Thanks — your feedback was recorded.

Related questions

Have a different question?

✦ Ask the AI BMS Mentor → More from Sensors →