Consumables Inventory Control for Stone Fabricators

A stone shop can survive almost any equipment problem for a day. A saw that goes down gets worked around, a CNC that throws an alarm gets rescheduled, a truck that will not start gets replaced with a rental. What a shop cannot work around is running out of mixing tips at four in the afternoon with three tops seamed and waiting. Consumables are the smallest line on the purchase ledger and the most common cause of unplanned downtime, because they are the only items that get used up completely and silently, without a warning light.

The reason consumables get managed badly is that they feel too small to systematize. Nobody wants to run a stockroom for sandpaper. But the cost of a stockout is never the price of the missing item; it is the crew standing around, the install that slips to next week, the customer who now has a story to tell, and the emergency freight bill to fix it. A basic control system for consumables takes a few days to set up and a few minutes a week to run, and it is one of the highest-return administrative changes available to a fabrication shop.

Which consumables actually stop production

The first job is to separate the items that halt work from the items that merely annoy people. A shop stocks hundreds of consumable lines, and most of them can be missing for a week without consequence. A much shorter list will stop a department cold. Ranking that list honestly, department by department, is what turns inventory control from a filing exercise into a production tool.

On the saw and CNC side, the critical items are cutting tools: bridge saw blades in the sizes you actually run, profiling and milling tooling, core bits for faucet and fixture holes, and the bushings, flanges, and arbor hardware that let you change them. Tooling hardware is the sleeper item on that list. A shop with a spare blade and no correct bushing is in exactly the same position as a shop with no blade at all, and the hardware is rarely tracked because it is not consumed on a predictable curve.

On the finishing side, the critical items are polishing pads across the full grit sequence and the backer pads or holders that drive them. Missing a single grit in the middle of a sequence is worse than being short at either end, because there is no way to skip it without leaving a scratch pattern the next step cannot remove. Any shop that stocks pads by total count rather than by grit will eventually discover this at the wrong moment.

On the assembly side, the critical items are the adhesive system and everything that goes with it: cartridges or knife-grade adhesive, hardener, the correct static mixing tips for those cartridges, color pigments if you tint, and the dispensing gun itself. Mixing tips are the classic stockout because they are single-use, cheap, easy to under-order, and completely non-substitutable. Structural reinforcement belongs in the same group: fiberglass mesh, rodding material, epoxy for rodding, and the abrasives used to prepare a seam.

Finally there is the safety and compliance group, which stops production for a different reason. Respirator cartridges and filters, hearing protection, cut-resistant gloves, and eye protection are consumables, and running out of them means either an unprotected crew or a stopped line. Given that OSHA sets a respirable crystalline silica permissible exposure limit of 50 micrograms per cubic meter as an eight-hour time-weighted average with an action level of 25, and a noise permissible exposure limit of 90 dBA with a hearing conservation action level of 85 dBA, these items are not optional stock.

Building a working reorder system

Min, max, and the reorder point

A reorder point is the stock level at which you place an order so that the replenishment arrives before you run out. The standard calculation is average usage over the lead time, plus a safety stock buffer. Usage comes from your own history, lead time comes from your supplier, and safety stock covers the variability in both. The important discipline is that all three numbers get written down and reviewed, rather than living in the head of whoever normally orders.

Maximum stock is the other half. Without a ceiling, a shop that has been burned by a stockout overcorrects and ties up cash in a corner full of pads that will be superseded before they are used. A sensible maximum is roughly double the reorder point, which keeps about one replenishment cycle of cover on hand above the trigger. Items with shelf life need a tighter ceiling than items without.

Lead time is where most reorder points are wrong, because shops use the lead time they were quoted rather than the one they experience. Track actual receipt dates against order dates for a quarter and use the real number, including the slow weeks. Imported tooling, custom-bond blades, and anything shipping by ocean freight deserve a longer buffer than a domestic stock item. Where a supplier has been unreliable, the honest response is either a larger buffer or a second source.

Two-bin kanban and bin labeling

For high-volume, low-cost items, the simplest reliable system is two-bin kanban. Each item lives in two containers. The crew draws from the first bin until it is empty, then starts on the second and puts the empty bin, with its label, into a reorder location. The label is the order signal. When the replacement arrives it goes into the empty bin, which goes behind the one in use. No counting, no software, no judgment call.

Two-bin works because the bin quantity is the reorder point. Size each bin to cover usage over the lead time plus buffer, and the system self-regulates. It fails only when someone opens both bins at once or when nobody moves the empty bin to the reorder rack, both of which are visible failures that a supervisor can correct on a walk through the shop.

Labeling is what makes the whole system survivable. Every bin should carry the item description, the part or reference number, the supplier, the reorder quantity, and the bin location code. Photographs on the label help enormously with items that look alike, particularly pads of adjacent grits and mixing tips of different diameters. A label that only says pads is not a label.

A worked example

The table below shows how the arithmetic works using the standard formula: reorder point equals average weekly usage multiplied by lead time in weeks, plus safety stock, with maximum stock set at twice the reorder point. The usage and lead time figures shown are illustrative placeholders to demonstrate the method. Substitute your own measured usage and your own verified supplier lead times before using any of it.

Shelf life, storage, and tracking usage by job

Not every consumable is stable on a shelf. Adhesives, polyester and epoxy resins, hardeners, cartridge products, sealers, and color enhancers all have a manufacturer-stated shelf life, and most of that stated life assumes storage inside a specified temperature range. Buying a year of adhesive to capture a volume discount is a false economy if part of it cures in the drum or loses reactivity before it is used. Shelf-life items need a maximum stock level driven by expiry rather than by cash.

First in, first out is the only rotation rule that works, and it needs a physical layout that makes it the easy option. Load new stock from the back of a shelf and pick from the front, or use gravity-fed racking where practical. Date every container on receipt with a marker, in large writing, on a surface a person can read without lifting the item. Shops that rely on a printed manufacturer date code end up not rotating, because nobody wants to hunt for a code in a dim storeroom.

Temperature is the storage variable that catches shops out, particularly those with unconditioned storerooms or with product living on a service truck. Resin systems, cartridge adhesives, and many sealers have both a storage range and a working range published by the manufacturer, and exceeding either changes cure behavior. Freezing is as damaging as overheating for many water-based products. A conditioned cabinet for chemistry is inexpensive relative to the value of what goes in it, and it also protects the accuracy of your color matching.

Tracking consumable usage by job is where inventory control turns into cost control. If a shop knows roughly how many blade inches, pad sets, cartridges, and mixing tips a typical kitchen consumes, it can price work from data instead of instinct, and it can spot the jobs where consumption ran far above normal. Those outliers are where the money is: a stone that eats tooling, a crew member using three tips per seam, a machine out of adjustment, or a slab that produced a remake nobody logged.

The tracking does not need to be elaborate to be useful. A tick sheet at the issue point, listing job number and item, run for one month, will tell most shops more about their consumable cost than a year of accounting reports. If you already run job-costing software, tie consumable issues to the job number. The point is attribution: an item drawn against a job is a cost you can analyze, and an item drawn against nothing is shrinkage you will only see at year end.

Cycle counting and keeping the system honest

An annual physical count is a compliance exercise, not a management tool. It happens once, it takes a Saturday, it is done by tired people, and the variances it uncovers are twelve months stale by the time anyone sees them. Cycle counting replaces it with a small, continuous count: a handful of items every week, chosen so that fast-moving and high-value lines get counted often and slow, cheap lines get counted rarely.

The value of cycle counting is not the count. It is that a variance found this week can still be explained. Someone remembers the rush job, the box that went out on the truck, the pads that were scrapped. Explanations lead to fixes: a bin that needs a label, an issue point that needs a sign-out sheet, a supplier who ships short. An annual count produces a number nobody can act on.

A workable rotation is to count the short list of production-stopping items weekly or every other week, the moderate-value items monthly, and everything else quarterly. Ten minutes with a clipboard covers most of it. The counts should be done by someone other than the person who orders, for the same reason that shops separate the person who cuts from the person who inspects.

Supplier relationships belong in the same conversation. Consolidating consumables with fewer suppliers usually improves lead time consistency and simplifies reordering, while single-sourcing a production-stopping item concentrates risk. The practical compromise is a primary supplier for volume and a qualified alternate for the items on the stop-production list, with the alternate's part numbers already cross-referenced in your bin labels so nobody has to research them during a shortage.

Finally, review the system on a schedule rather than after a crisis. Once a quarter, pull the stockout log, the cycle count variances, and the actual lead times, and adjust reorder points and bin quantities accordingly. Usage changes as the work mix changes, and a shop that starts doing more porcelain or more large-format work will burn through a different tooling profile than it did the year before. A reorder point set once and never revisited is a reorder point that is quietly wrong.

The last habit worth building is writing down the stockouts. Every time production waits on a consumable, log the item, the date, the hours lost, and what it cost to expedite. That log is the business case for the whole system, and it converts an argument about tying up cash in inventory into a comparison of two real numbers.

A reorder system only works if the items behind it are consistent from order to order. Standardizing on a known tooling line makes usage predictable, which makes reorder points accurate: build your stop-production list around diamond blades and polishing pads you can reorder without requalifying, and keep chemistry on a tight expiry-driven ceiling by ordering adhesives and sealers in quantities matched to real usage. Dynamic Stone Tools supports US fabrication shops with consistent stock and repeatable part numbers.