Water intelligence for industrial parks · East Africa
See every litre in your park — where it comes from, where it goes, and what happens next.
Hydro-SaaS turns meters, boreholes, SCADA, lab results and weather into one live model of an industrial park's water: balance, pollutant loads and their sources, demand forecasts, and each factory's compliance.
Built by hydrologists and modellers at KAI Water & Energy, Kampala.
- live modules
- 5
- effluent limits built in
- NEMA 2020
- data sharing
- Tenant-controlled
- while in beta
- Free tier
Modules
One model of the park, six ways to use it
Park water balance
Bulk supply, boreholes and rain in; every tenant's use; wastewater, treatment and what reaches the wetland or river. Losses and unmetered use are shown, never hidden.
See it in the demo →
Pollutant loads and sources
Flow × concentration at every outlet. When the treatment plant inlet spikes, see which factory most likely caused it — and which ones it wasn't.
See it in the demo →
Demand forecast
Next ten days of park demand from the weather forecast, plus scenarios for new tenants against supply capacity.
See it in the demo →
Boreholes and drawdown
Pumping water level against the pump intake and safe level, days until it gets there, and falling specific capacity that signals clogging.
See it in the demo →
Factory effluent compliance
The tenant module: each factory's outlets checked against the NEMA 2020 effluent standards or its permit, with early warnings and anomaly alerts.
See it in the demo →
Next: treatment optimisation, receiving environment, drainage and flood risk
Plant control to stay within limits at least cost, wetland and river impact with satellite data, and storm drainage modelling.
Tested on real data
Which manhole did the pollution come from?
The KAI engine on the real Bellinge sewer network in Denmark (about 1,000 manholes, run in EPA SWMM): blind tracer tests, how often it names the right source, and how much a few well-placed sensors help. Open data, real results.
Open the real-data case →
Right manhole on the first guess, blind test releases (a random pick: 3 %)
Tested on real data · boreholes
Is it the pumping or the weather?
Real pumping tests and a 55-year water-level record from the Netherlands. Trained on 1960–1999, the engine predicted the next 15 years of levels to within 0.22 m; a weather-only model was off by 0.77 m. A MODFLOW 6 borehole model reproduces the field test to within 1 cm.
Open the borehole case →
Tested on real data · effluent
BOD on the day, not five days later
A soft sensor on a real treatment plant's records: effluent BOD5 from a turbidity probe, checked against 504 lab results. In a blind year it flagged every warning day of the plant's worst failure on the day; the last lab result the operator had flagged none.
Open the soft-sensor case →
Who it is for
Built for everyone with a stake in the park's water
Park operators and developers
One live picture of the park's water: supply security, losses, central treatment performance, whole-park compliance, planning for new tenants.
Tenant factories
Their own water use, borehole health and effluent compliance — and full control over what they share with the park.
Lenders and investors
Environmental monitoring of financed parks, with traceable data behind every number.
Privacy by design
Tenants stay in control
A park operator sees the whole park only as far as each factory allows. Every tenant chooses separately whether to share water quantities, effluent quality and compliance alerts, and can change or withdraw that choice at any time. What a tenant doesn't share still shows up honestly in the park balance as unmetered use, without exposing the tenant's data.
Example of a tenant's sharing settings
Running a park, or a factory inside one?
We start with a fixed-price park water diagnostic: water balance, load map, risks and a sensor plan, then switch on the live model.