Thermo-Cognitive Advanced Process Control (TC-APC) models, simulates and optimizes your process in real time to improve yield, conversion and throughput, and reduce energy use, with a quick payback.
Applications
Ten processes, the plant around them, one argument, and eleven honest statements of where the work stands.
Every page below names the instrument the plant steers by and what that instrument cannot tell it.
And says, at the top, whether this has been deployed or not.
Dryers, milk and chemical powders
Outlet temperature tells you the air is hot. It does not tell you how wet the powder is.
Milk, food and chemical spray dryers are run on outlet temperature and a lab result an hour old, so every lot is dried well inside its moisture limit, to be safe. That margin is paid twice, in fuel and in powder weight that could have been sold. On a 150 TPD milk powder dryer, each 0.1 point of moisture is worth about ₹1.2 crore a year.
Starts with a free audit of your own data.
Natural extracts and oleoresins
The extract passed on residual solvent. What it lost getting there appears in no reading.
Concretes, absolutes, resinoids and spice oleoresins are all made the same way, and all decided in the last hour of solvent removal, where the last parts per million of solvent are bought with heat and paid for in aroma and color. The end point moves with every lot, so a fixed recipe strips every batch for as long as the hardest one needs.
Engineering proposal. Starts with a free check.
Aroma and specialty distillation
A boiling point tells you something is boiling. It does not tell you what.
Batch and continuous vacuum fractionation. Under vacuum a small pressure drift moves the apparent boiling point further than the transition you are trying to detect, so the cut is made on a signal that cannot resolve it. Cutting early fails silently and cutting late fails loudly, which is why the bias runs one way in almost every plant.
Published analysis. Technical note available.
Soda ash carbonation
The conversion is known when the analysis comes back. By then the tower has moved on.
A hundred feet of steel doing three jobs at once, with twenty three ionic species spread across twelve vertical zones and no instrument reporting any of it. This is the one deployment we own: fourteen towers, a hybrid model solving over three hundred simultaneous equations against live plant data, validated over six months.
Deployed and validated. Reference call available.
Sugar crystallization, vacuum pans
Supersaturation decides the crystal. In seventy years nobody has measured it online.
Hold the strike inside the metastable zone and crystals grow. Leave it and you make grain nobody wanted, which you then pay steam and time to unmake. The claim that supersaturation cannot be derived online turned out to be a claim about a function nobody had constructed.
Model developed and validated against laboratory and plant data.
Reaction selectivity
Temperature buys conversion and hazard in the same movement. Nothing tells you which one you have just bought.
Nitration, Grignard and condensation. Here part of the margin is protecting something other than the yield, so the boundary is treated as absolute and never traded. The only thing that can move is how well you know where you are standing relative to it.
Published analysis, Chemical Weekly, October 2025.
IPA and DIPE separation
Near the azeotrope the temperature goes flat exactly where the cut point sits.
On a continuous column in a propylene direct-hydration plant, the instrument goes quiet over precisely the range that decides the product. To keep the co-product out of the bottoms, operators raise internal traffic, and prime alcohol is carried overhead with it, into the recovery loop.
Engineering proposal. Starts with a data audit.
Solvent recovery, DMSO and DMF
The solvent is back in the tank. How much went to residue, and how much decomposed on the way, appears in no reading.
The last of the batch is where the money is and also when the solvent is least able to take the heat. There is a point where the next kilogram recovered costs more in degradation than it is worth, and it moves with every charge.
Engineering proposal. Starts with a data audit.
Protein gelation, surimi to crab stick
A gel strength grade tells you the paste was strong eight months ago, in another country.
Two modules, independent of each other. The first reads the paste in the bowl while the cutter runs and needs no gel strength figure at all. The second handles formulation against a quality target and does need a current parameter set, which we say plainly rather than work around.
Engineering proposal. Module 1 starts with a data audit.
Boilers and combustion
An oxygen reading tells you the excess air is right for the coal you assumed you were burning.
Oxygen trim holds the oxygen at a number. It does not choose the number, it does not know what is on the grate today, and it is a steady-state device, so every ramp is uncontrolled. Boilers are safe and running in a wide comfort band rather than a tight window.
Engineering proposal. Starts with a data audit.
Plant Ledger, the whole plant
A log sheet tells you what was made. It does not tell you what was lost.
The same argument at the scale of the plant. Every batch balanced, every meter checked against the physics, early warnings from the instruments already installed, and the closed-loop engines on this page added where you choose. Machines and discrete lines are covered too.
Engineering proposal. Starts with a month of log sheets.
All eleven begin the same way.
You send data from one unit
Operating data with the laboratory results beside it. Nothing is connected to anything and nothing on your plant changes while we look at it.
We tell you what the margin costs
How wide it actually is on your unit and what it is worth, with the uncertainty stated openly rather than folded into a single confident figure.
Including the answer no
If the unit is already being run close to its limit, we say so and there is nothing to buy. That has to be a real possibility for the exercise to be worth anything to you.
What happens after that, in full, including the guardrails and what the system does on every failure mode.
Where the same argument appears, and we have not written it up.
These have come up in earlier work or in conversation. None has a page, because none has been worked through to the standard of the eleven above. They are here because if one of them is yours, the first conversation is the same conversation.
Chocolate tempering
The crystal form decides snap, gloss and shelf life, and it is inferred from temperature and time rather than measured while it forms.
Pharmaceutical sterilization
Lethality accumulates through the cycle and is verified afterward. The margin is held wide because the position during the cycle is not known.
Limestone kilns and material handling
Calcination is decided inside the kiln and judged from what comes out of it, some time later, and by then the feed has changed.
Other thermally or composition limited processes
Wherever the quantity that governs the money is inferred from a proxy, updated too slowly, or confounded by something else that moves. Tell us the process and the instrument and we will tell you whether the argument applies.
The mark
Find the optimum. Hold it while it moves.
The curve is a yield or selectivity response. The amber point is the optimum, and the dashed drop is where it sits on the operating axis.
The second half of the line is the part that matters. Finding an optimum is a study, and studies end up in drawers. Holding one while the feed, the fouling, the weather and the wear keep moving it is a control problem, and that is the work.
Start with the note
Cut-point determination in the vacuum fractionation of odour-critical products
Written for a distillation column, and the measurement argument underneath it is the one every page here rests on. Three pages. Free, no form, and we do not ask for your email address.
Or describe your own unit
Tell us the process and what it is steered by
If the variable that decides your yield is not measured online, the argument on these pages applies to you and we will say how. If it is measured, we will say that too, and there is nothing here for you.
projects@yielddynamics.in Talk to IIT Bombay engineers Call +91 94230 07004Monday to Saturday, 9:00 to 19:00 IST