Steel service centres can use AI to build custom processing software for coil-wise inventory with grade thickness width and heat number, slitting and cut-to-length planning, mult combinations and edge trim, yield and end scrap, customer-wise stock and reservations, mill test certificate handling, toll processing of customer material, dispatch by bundle and weight, and margin measured against replacement cost. Pentoggle is an AI platform that generates production-ready software from a plain English description, which means a service centre can build an application around its own lines and customers instead of adapting to a system built for a fabricator.
Most steel service centres in India already run Tally for accounting, some run Busy, and larger businesses may run SAP Business One. Those systems handle purchase, sales, GST and accounting well. This is not a proposal to replace them. Pentoggle builds the processing application around them, covering the workflows they were never designed for.
A service centre earns differently from a manufacturer. You buy a commodity whose price moves daily, hold it, process it modestly, and sell it into a market that knows what the commodity costs. The processing margin is thin and fairly well understood by everyone. What actually decides the year is the price at which the stock on your floor was bought, relative to the price at which it can be replaced.
Key takeaways
- Inventory in a service centre is a market position whether or not you intended one, and margin measured against purchase cost can look healthy while the business is losing ground.
- A coil is an identity rather than a quantity, carrying grade, dimensions, heat number and its own cost.
- The slitting plan decides yield, and it is decided against the order mix rather than one order at a time.
- Mill test certificates have to travel with the material, and reconstructing which coil a bundle came from afterwards is slow and unconvincing.
- The number that runs a service centre is realisation against replacement cost, not against what the coil cost you.
Why service centres are badly served by existing software
Tally records the coil purchase and the sales invoice, and produces an accurate margin against purchase cost. What it does not tell you is that the margin was earned by a coil bought four months ago at a price you cannot buy at today, which means the business made a trading gain and a thin processing margin, and only one of those repeats.
Packaged ERP handles items and quantities. A service centre's inventory is neither. Each coil is a distinct object with a weight that changes as it is processed, a heat number that must follow its output, a cost of its own, and a physical location. Systems that treat stock as a tonnage figure per grade lose the identity, and the identity is what the customer's certificate and your costing both depend on.
The rest lives in a coil register, a slitting plan drawn up by the production head, a weighment slip book and a stack of mill certificates in a file.
Most centres are running some combination of the first two columns below.
What service centres use today, and what they can build instead
| Registers and Excel | Packaged ERP | Application built with Pentoggle | |
|---|---|---|---|
| Coil inventory | Register with weights | Tonnage by grade | Coil-wise with grade, dimensions, heat, cost and location |
| Cost per coil | Averaged | Weighted average or standard | Actual per coil, with replacement cost alongside |
| Slitting and cut planning | Planned by the production head | Not modelled | Mult combinations planned against the order mix |
| Yield and scrap | Reconciled at stock take | Standard scrap percentage | Per coil, with edge trim and end scrap separated |
| Mill test certificates | Filed and photocopied | Attachment at best | Held against the coil and carried to every dispatch |
| Toll processing | Challan book | Subcontract module, rarely used | Customer material tracked separately with challans |
| Customer reservations | Promised verbally | Stock allocation at best | Coils reserved against a customer or an order |
What a service centre can build
Each of these can be built separately or combined. Most centres start with coil-wise inventory.
Coil register
Every coil with supplier, grade, thickness, width, weight, heat number, mill certificate, purchase cost, arrival date and physical location.
Balance tracking
Weight consumed and remaining as a coil is processed across several jobs, so a part-used coil remains findable and accountable.
Order book and reservations
Customer orders with grade and size requirements, and coils reserved against them.
Slitting and cut-to-length planning
Mult combinations planned against the order mix, with expected edge trim and end scrap for each plan.
Production recording
Output by size and weight per job, with actual trim, end scrap and rejection.
Yield reporting
Saleable output against input weight, by coil, by plan and by line.
Mill test certificate handling
Certificates held against the coil and issued automatically with every dispatch that contains material from it.
Toll processing
Customer-owned coils tracked separately with inward and outward challans and material reconciliation.
Dispatch
Bundles and weights, with the certificate, the weighment and the documentation each load needs.
Margin against replacement
Realisation per tonne compared to the current cost of replacing the material, alongside the conventional margin.
Your inventory is a position, whether you meant it or not
A service centre holds material for weeks between purchase and sale. Steel prices move in that window. That makes every coil on the floor a position in a commodity, regardless of whether anybody in the business thinks of themselves as trading.
The consequence shows up in reported margin. A coil bought at a high price and sold after the market softens produces a poor margin no matter how efficiently it was processed, and the production team gets asked why. A coil bought before a rise produces an excellent margin that nobody in operations earned. Managing the business on margin against purchase cost means the operational signal is buried under a price signal, and the two get confused month after month.
Measuring alongside replacement cost separates them. If you sold at a price that would still be profitable were you buying the coil today, the processing business is sound. If not, the margin you reported came from the purchase, and it will not repeat when stock is replenished at current prices.
This is not an argument for a different accounting method, since your books are your books. It is an argument for a second view, maintained operationally, that tells you which part of your result came from buying well and which from running well.
The slitting plan decides the yield
A coil of a given width is slit into a combination of narrower strips. Whatever cannot be accommodated becomes edge trim, and the tail of the coil becomes end scrap. Both leave as scrap at scrap rates, having been bought at coil rates.
The choice of which widths to run together is where yield is decided, and it is made before the line starts. Planning one order at a time nearly guarantees waste, because a single order rarely combines into the coil width neatly. Planning against the order book allows combinations that fill the width, and lets a planner decide whether to hold an order for a day so it can run with something that complements it.
There is a purchasing consequence too. A centre that records trim by coil width and size combination learns which coil widths suit its actual order pattern, which is a better basis for buying than availability and price alone.
The gains here are not marginal in a business where material is the overwhelming majority of cost and processing margins are thin.
The certificate has to follow the steel
Customers in construction, automotive and general engineering ask for the mill test certificate covering the material they received. The heat number on the certificate has to correspond to the material in the bundle, and the connection is made in your plant when the coil is processed.
If coils are tracked as tonnage rather than as objects, that connection breaks. Two coils of the same grade and size, processed on the same day, produce bundles that nobody can attribute afterwards. The certificate issued is then the one that seems right, which is a problem if the customer ever compares.
Holding the certificate against the coil, and carrying the coil identity into every job and dispatch it feeds, makes the certificate an output rather than a search. It also handles the harder case, which is a customer coming back months later with a bundle tag and a question.
Why building this is now practical
A centre running a slitter and a cut-to-length line has never been able to justify custom software. A development team, a specification document and a six month build were never going to be recovered on processing margins measured in hundreds of rupees per tonne.
That has changed. With Pentoggle you describe how your centre runs, including your lines and their capabilities, your grades and sizes, how you plan cuts and how you handle certificates and toll work, and get a working application. When you add a line, start doing toll processing for a new customer, or want replacement cost tracked alongside margin, you describe the change and the application updates. Most centres start with coil-wise inventory, because everything else on this page depends on it existing.
Why service centres choose Pentoggle
Coils as objects, not tonnage
Grade, dimensions, heat, cost, balance and location held per coil.
Certificates that follow the material
Issued from the coil identity rather than searched for at dispatch.
Works alongside Tally
Pentoggle handles the yard and the lines. Your accounting stays where your CA already works.
Planning against the order book
Cut combinations chosen for yield rather than one order at a time.
Changes in days
A new line or a new toll processing arrangement does not become a three month project.
The one number that runs a service centre
Realisation per tonne against replacement cost.
Conventional margin tells you what happened. This tells you whether it will happen again. If your selling prices still leave a processing margin when the material is valued at what it would cost to buy today, the business is working. If they do not, the reported profit came from the timing of a purchase, and the next cycle will look different.
Track it monthly by grade, alongside yield. The two together answer the only question that matters at this level: are we buying well, running well, or coasting on a price movement that has already happened.
Ready to build steel service centre software?
You are not only selling steel. You are selling the price you bought it at.