- PUTS IN
- A geotagged report of burning, dust, a dug-up road or an idling genset — in under a minute, with a photo.
- GETS BACK
- A tracking code, the air in their own locality, and the GRAP advisory that applies to them today.
- ACCOUNTABLE FOR
- Honest, located reports. Nothing more is asked of them.
Gurugram cannot fix the air it cannot see — and cannot act on what nobody owns.
VAYU puts one record under the whole city: every reading, every citizen report, every ward, every facility and every statutory measure in the same place — then names the body accountable for each one and holds the clock on it. Pollution stops being a season the city endures and becomes a workload the city schedules.
The pollution is measured. The response is scattered.
Gurugram does not lack data — it lacks one place where the data becomes somebody's job before the episode peaks. Five systems that never meet produce five partial truths and no owner.
Five truths, no owner
A resident complains on one channel. A monitor records a spike on another. The ward knows the site. The department holds the notice. Nobody holds all four, so the episode is reconstructed after it has passed.
One record, one clock
The same five inputs land on one city record, tied to a ward, a named owner and a statutory lead time. The question stops being "what happened?" and becomes "who is on it, and by when?"
One loop, running continuously
Six stages. Each one has a surface in this app, a body that owns it and a record it leaves behind. Click a stage to see what it does.
Sense — the city reports itself
Regulatory monitors, weather, satellite-era forecasts and residents all write into the same intake. A resident's photo of a burning pile and a CPCB station reading are both evidence; they differ in confidence, not in standing.
What happens to one report, step by step
The handoff is the thing that usually breaks. Here it is the thing the system records: every step carries an owner and a timestamp, and nothing leaves a step without one.
Six parties, six jobs, no overlap
Clean air is not one body's mandate. VAYU gives each party the narrow slice it can actually act on — and shows the others that it moved.
- PUTS IN
- Standing local knowledge: the recurring burn spot, the site that works past hours, the stretch that is never swept.
- GETS BACK
- A locality view of what was raised, what the ward did, and what is still open — evidence to take to a meeting.
- ACCOUNTABLE FOR
- Mobilising the colony on pledges and advisories; confirming whether a fix actually held.
- PUTS IN
- Ground truth: which complaint is real, which site is the actual source, what the ward has to deploy.
- GETS BACK
- A ward dossier — 1685 assets, named officials, colonies, facilities and the live queue for their ward alone.
- ACCOUNTABLE FOR
- Closing the ward's queue inside the lead time GRAP sets for each measure.
- PUTS IN
- Compliance state for the 462 mapped sites: construction dust control, DG set hours, waste handling.
- GETS BACK
- Advance notice of the stage about to come into force, so shutdown is scheduled rather than raided.
- ACCOUNTABLE FOR
- Meeting the measure that names them before the inspection, not after the notice.
- PUTS IN
- Enforcement and works: sweeping, sprinkling, notices, site stoppages, PUC drives, power continuity.
- GETS BACK
- A queue filtered to the measures their own body owns, with the ward, the site and the deadline already attached.
- ACCOUNTABLE FOR
- Executing their named GRAP measures and filing the evidence that closes them.
- PUTS IN
- The decision: which stage to invoke, where to concentrate, what to escalate to CAQM.
- GETS BACK
- One map of the city, the open workload by ward and by body, and whether yesterday's order was actually carried out.
- ACCOUNTABLE FOR
- Calling the escalation early, and answering for the city's record with evidence rather than assertion.
Six layers over one geography
Each layer answers a different question, and they only become useful stacked: a reading means little until you know whose ward it fell in, what sits under it and who was already tasked there.
Illustrative geometry — the live map in City intelligence map draws the real MCG boundaries. Hover a layer to isolate it.
Move the index. See the city's obligation change.
This is not a forecast and not a model. It is the published CAQM Graded Response Action Plan read against this city's own inventory: drag the index and the page shows exactly what comes into force, which body owns each measure and how long they have.
What this does not do: it does not forecast AQI, attribute pollution to a source, or claim a health outcome. It answers one question — if the index reaches this level, what is mandated, who owns it, and which wards carry the load. The authoritative text is the current CAQM order.
The whole point is to act before the peak
Reacting at the peak is already too late — the measures that matter most need a day's notice to take effect. With forecasts, the ward inventory and the GRAP ladder in one place, the city can pre-position instead of scrambling.
Met and air forecasts flag a build-up window before it is visible outdoors.
VAYU · FORECAST INTAKEThe scenario ladder is run for the projected index; the tasklist is drawn before it is needed.
CITY COMMANDSweepers, water tankers and inspection teams move to the wards the model loads heaviest.
MCG · GMDAFacilities, schools and RWAs are told what comes into force and when — no surprise shutdowns.
HSPCB · WARD TEAMSThe queue is already assigned. Enforcement starts on hour one instead of on day two.
ALL BODIESClosure evidence is checked against the next reading; what worked becomes a lesson on the record.
VAYU · WARD TEAMSFewer bad-air days the city did not see coming.
VAYU does not clean the air. It makes sure that when the air turns, Gurugram knows within the hour, knows which ward and which site, knows which body owns the response, and can prove afterwards whether it happened.
About the numbers on this page: every count is read from this instance's own tables when the page loads — 36 wards, 1555 localities, 462 facilities, 1685 ward assets. Where an import has not been run, the figure shows as zero rather than as an estimate. This is a prototype environment and not an official government monitoring system.