What the season costs you, and when you find out
Most of these numbers exist somewhere in the mill today. The problem is that they arrive weeks late, from four different departments, after the cane they refer to has already been crushed.
Stoppages live in a register
Hours are written down, summarised late and disputed. Which section actually cost the day — mills, boiler, cogen, electrical or cane shortage — never gets settled with evidence.
Recovery losses show up in the lab report
By the time pol in bagasse tells you something was wrong, the mill settings, imbibition rate and preparation index from that shift are long gone.
Steam percent cane is managed by feel
It decides how much bagasse you save and how much power you export, and it is one of the few levers that pays in cash during the season itself.
Cane arrival and crush plan are not connected
Harvest-to-crush delay costs sucrose before the cane reaches the first mill, and the yard, the weighbridge and the tandem are managed as three separate problems.
Auxiliary consumption quietly eats export revenue
Every extra kilowatt consumed in-house is a kilowatt not sold. Auxiliary load creeps across a season and almost nobody watches it daily.
Boiling house data is manual and late
Brix, pol and purity are recorded by hand, entered later, and never sit alongside the process conditions that produced them.
What we connect to in your plant
A sugar mill is really four plants sharing a season — cane handling, milling, boiling house and cogeneration. We connect them as one system, because that is how the losses move between them.
Business and lab systems
The signals we read, and what each one tells you
During the season, the only signals worth having are the ones that change a decision before the next shift starts.
| Signal | What it tells you |
|---|---|
| Cane crushed per hour against rated capacity | Crush rate loss as it happens, rather than as a weekly average that hides the bad nights. |
| Stoppage start and stop with section tags | Downtime split cleanly across mills, boiler, cogen, electrical and cane shortage, without anyone maintaining a register. |
| Mill hydraulic pressure, roller torque and imbibition flow | The settings behind extraction, captured alongside the shift that chose them. |
| Preparation index and fibrizer motor current | Cane preparation quality, which sets the ceiling on everything the tandem can achieve downstream. |
| Bagasse moisture and pol in bagasse | Milling loss attributed to a shift and a set of settings instead of to the season as a whole. |
| Brix, pol and purity across juice, syrup, massecuite and molasses | Where sucrose is actually being lost between the mill and the bag. |
| Evaporator vapour pressure, pan cycle time, centrifugal cycle and load | Boiling house throughput and the bottleneck that is really limiting it. |
| Boiler steam flow, pressure, temperature and bagasse feed | Steam percent cane, live, which is the number that converts into exportable power. |
| Turbo-generator export and auxiliary consumption | Net exportable power by shift, and where in-house load is eating it. |
| Weighbridge tickets and harvest timestamps | Cane staleness before crushing, and the sucrose lost before the first mill ever sees it. |
Season performance, updated by the shift
The numbers your management committee reviews weekly, available while there is still season left to act in.
Crushing rate vs capacity
TCD and %
Capacity utilisation
% of available hours
Downtime hours by cause
hours, ranked by section
Sucrose recovery
%
Reduced boiling house recovery
%
Pol % bagasse
%, by shift
Total losses
%, split by stream
Steam % cane
%
Power exported
MWh per day
Auxiliary power
% of generation
Imbibition %
% on cane
Sugar quality
ICUMSA and grain size
Cane staleness
hours harvest to crush
Season progress vs plan
tonnes crushed
AI trained on your process. Not a generic factory.
Models that understand a crushing season — where capacity is fixed, the clock does not stop, and every loss is measured against a window that closes.
Stoppage classifier and choke prediction
Learns the current and pressure patterns that precede cane carrier chokes, mill jams and centrifugal trips, and tags every stoppage to a section and a cause without anyone maintaining a register.
Recovery loss attribution
Ties pol in bagasse and in final molasses back to mill settings, imbibition rate, preparation index and shift, so the loss has an owner and a specific correction rather than a season-end discussion.
Steam percent cane optimiser
Finds the boiler and process setpoint combinations that produced the lowest steam per tonne of cane, turning saved bagasse into exportable power during the season it was saved.
Crush plan against cane arrival
Forecasts arrivals against yard stock and current crush rate to cut staleness and avoid both mill starvation and yard overflow, which are the two ways a good crushing day gets wasted.
Your first thirty days
Connection is read-only and done one section at a time. Nothing about it requires you to stop crushing.
Connect the tandem and weighbridge
Milling tandem instrumentation and weighbridge data, read-only. Crush rate against capacity becomes visible by the hour rather than by the week.
2
Automate the stoppage log
Stoppages get captured and classified from the machines, and last season's downtime is rebuilt so this season has something honest to be measured against.
3
Add the boiling house and lab
Brix, pol and purity join the process timeline, and the loss balance across bagasse, filter cake and molasses becomes a live picture.
4
Bring in cogeneration
Boilers and turbo-generators come on, giving steam percent cane and net export by shift — usually the first number that pays for the project.
What plants aim for in the first quarter
1–3%
increase in capacity utilisation from recovered crushing hours
0.1–0.3 pts
improvement in sucrose recovery
2–6%
reduction in steam percent cane, converted to export
same shift
visibility of losses, instead of post-season analysis
Target ranges for scoping a pilot. Replace with your own validated customer results before this page goes live.
