Water and energy intelligence for a 300-acre FMCG manufacturing campus
This engagement brings a 300-acre FMCG manufacturing campus onto Aqueduct. The campus runs 8 business units, from grain milling and dairy to beverages, edible oils and personal care, each billed but none seen as one system. The program is designed to start with one high intensity plant, prove it in a quarter, then scale across the campus.
A 300-acre FMCG manufacturing campus runs eight business units side by side: flour and atta mills with 200+ TPD of grain milling, dairy and ghee, beverages and juices, herbal and ayurveda, edible oils, 30+ spice lines, personal and home care, and packaging and utilities. Each unit meters its own bill, but no one sees the campus as a single, comparable system, and the monthly bill gives one number with no view of which line or shift spent it.
Water is the sharpest pressure. The campus sits in a water stressed geography, and reuse from STP, RO reject, cooling blowdown and condensate is under instrumented. Disclosure under BRSR, pollution control board effluent norms and zero liquid discharge expectations calls for auditable, line level, timestamped water and discharge data.
The plan does not start with the whole campus. It starts with one contained, high intensity unit, the beverages and juices plant, where process water, CIP wash downs, rinsing, RO polishing, boilers and cooling all sit under one roof. aQ WaterGuard builds a live water balance in which supply equals floors plus fire plus RO feed plus unaccounted loss. Tanks, headers, RO and STP update every 60 seconds, and when the equation stops reconciling, the loss term rises before the bill does.
aQ EnergySmart turns the same lens on power: feeder level load and power factor, peak demand, solar offset and distribution losses. Portfolio views rank units by total use, intensity and use per unit of output, so a small, intense plant cannot hide behind a big one. Smart Controls can put pumps, VFDs and air handlers on schedules and automation rules where control is enabled, with owner approval.
The plan runs about 90 days from site survey to a live digital twin: a survey on day 0, an instrumentation plan by week 2, non intrusive metering in weeks 3 to 8 without interrupting production, and the pilot live by about day 90 with first losses named. Dairy and ghee is planned as pilot two, drawing on Milvian AI's dairy work, followed by edible oils and herbal units in quarters 2 and 3, and every unit on one campus view in quarter 4, with withdrawal, discharge and reuse records ready for BRSR, GRI 303 and CDP Water.
Start with one plant
Of eight units on a 300-acre campus, the plan starts with the beverages and juices plant as a contained pilot.
Non intrusive metering
Flow, pressure, level and power sensors are fitted in phases, and data starts flowing without stopping a line.
A balance that must close
Supply equals floors plus fire plus RO feed plus unaccounted loss, with tanks, headers, RO and STP updated every 60 seconds.
The loss term rises
When the equation stops reconciling, the loss term climbs and the leak shows up before the bill does.
Schedule the big loads
Smart Controls can put pumps, VFDs and air handlers on schedules and rules where control is enabled, with owner approval.
Before and after quantified
By about day 90 the digital twin is planned to be live, with first losses named and the before and after measured.
How it works at A 300-acre FMCG manufacturing campus.
Many thirsty units, one campus nobody could see whole
The campus runs flour and atta mills, dairy and ghee, beverages and juices, herbal and ayurveda, edible oils, spices, personal and home care, and packaging and utilities. Each unit meters its own bill, but no one sees the campus as a single, comparable system.
Water sits in a water stressed geography, and reuse from STP and RO reject, cooling blowdown and condensate is under instrumented. Disclosure under BRSR, pollution control board effluent norms and zero liquid discharge expectations calls for auditable, line level water and discharge data.
- One number on the monthly bill, no view of which line or shift spent it
- Reuse loops recoverable only if measured line by line
- Units of very different size that could not be compared fairly
- Disclosure needs that called for line level, timestamped records
One live campus on aQ Core, starting with one plant
aQ WaterGuard builds a live water balance for the pilot plant, where supply equals use plus fire, RO feed and unaccounted loss. Tanks, headers, RO and STP update every 60 seconds, and when the equation stops reconciling, the loss term rises before the bill does.
aQ EnergySmart applies the same lens to power: feeder level load and power factor, peak demand, solar offset and distribution losses. Portfolio views rank units by total use, intensity and use per unit of output, and Smart Controls can schedule pumps, VFDs and air handlers where control is enabled, with owner approval.
- 01
Site survey
The team walks the pilot unit, maps meters, panels and taps, checks wireless coverage and pulls 12 months of water and power bills.
- 02
Instrumentation plan
By week 2, a sensor list, gateway layout and water balance model go to the campus team for sign off.
- 03
Phased metering
In weeks 3 to 8, non intrusive flow, pressure, level and power sensors are fitted and data starts flowing, without interrupting production.
- 04
Live pilot
By about day 90, the digital twin is live for the beverages and juices plant, with first losses named and the before and after quantified.
- 05
Scale in waves
Dairy and ghee, edible oils and herbal units follow in quarters 2 and 3, then every unit joins one campus view in quarter 4.
| Layer | What the program plans |
|---|---|
| Pilot | The beverages and juices plant, with process water, CIP wash downs, rinsing, RO polishing, boilers and cooling |
| Water | A live water balance across tanks, headers, RO and STP, updated every 60 seconds |
| Reuse | STP, RO reject and condensate measured so recovery is verified, not assumed |
| Energy | Feeder level load, power factor, peak demand, solar offset and distribution losses |
| Portfolio | Units ranked by total use, intensity and use per unit of output |
| Controls | Schedules and automation rules for pumps, VFDs and air handlers, where control is enabled, with owner approval |
| Reporting | Withdrawal, consumption, discharge and reuse records for BRSR, GRI 303 and CDP Water |
Expected outcomes
By the end of one quarter, the program targets a pilot plant that is a live, self explaining system and a template for the rest of the campus. Later phases are designed to bring every unit onto one portfolio, ranked, benchmarked and ready for disclosure.
- Leaks, idle draw and power factor penalties found
- Bursts, overflows and demand spikes caught in 60 seconds, not next month
- Plants compared on intensity per tonne or per liter
- Audit grade water and energy records for BRSR, PCB and ZLD disclosures
After the pilot, dairy and ghee is planned as pilot two, drawing on Milvian AI's dairy work, followed by edible oils and herbal units. The final phase puts every unit on one campus view, with Smart Controls automating the biggest loads where control is enabled, with owner approval.
From the field to a verified result.
What changed on site.
- One number on the monthly bill per unit
- Reuse loops under instrumented and unverified
- Units of very different size compared unfairly
- Disclosure data assembled after the fact
- A live water balance updated every 60 seconds
- STP, RO reject and condensate recovery measured
- Units ranked by intensity and use per unit of output
- Line level, timestamped records for BRSR, PCB and ZLD
- Starting with one contained plant proves the method in about a quarter before it is asked to cover a whole campus.
- In a water stressed geography, reuse only counts when it is measured line by line, not assumed.
- One template, cloned unit by unit, turns eight separate bills into one comparable campus.
Logistics and corporate real estateWater stewardship across Amazon's sites, measured in real time
AviationA top 25 US airport live on Aqueduct in two weeks, with a major leak found at once
HealthcareUtility and critical room intelligence for one of India's top hospital chains
Common questions
Aqueduct brings every unit's water and energy onto aQ Core as one comparable system. Portfolio views rank units by total use, intensity and use per unit of output, so a small, intense plant cannot hide behind a large one.
A self contained, high intensity plant has clear boundaries, one supply, one discharge and one energy feed to reconcile. On this campus the beverages and juices plant was chosen as the pilot and template for the rest.
The plan for this campus is about 90 days: site survey on day 0, an instrumentation plan by week 2, phased metering in weeks 3 to 8, and a live digital twin by about day 90.
Yes. aQ WaterGuard measures STP output, RO reject and condensate line by line, so reuse is quantified rather than assumed. The records support BRSR, GRI 303 and CDP Water reporting that the company files.
No. The survey is non invasive, and sensors are non intrusive flow, pressure, level and power devices fitted in phases. The plan maps what exists and flags gaps without stopping lines.
See Aqueduct on your own facility.
Bring a site, a utility bill or a question. We will show you where the loss is.