Inventory optimization for manufacturers
Put the right inventory in the right place. Set stock targets that reflect real demand variability and real supplier lead times, find the excess, and protect service while releasing working capital.
Too much of the wrong stock, too little of the right
Most operations carry more inventory than they need and still run short. Both happen for the same reason: stock targets were set from judgement at some point in the past and have not been revisited since, while demand patterns, suppliers and lead times have all moved.
The symptoms are familiar. Expedited freight to cover a shortage on an item that is technically overstocked across the business. A warehouse holding twelve months of something that sells twice a year. A safety stock number nobody can explain the origin of. An annual write-off that everyone treats as a cost of doing business.
Inventory optimization replaces judgement-set numbers with targets derived from your own demand history and receipt records, item by item, location by location.
Reorder points, lead time and safety stock
Three numbers do most of the work. The reorder point triggers replenishment early enough that stock arrives before the safety level is reached. Safety stock absorbs the variation that remains.
| Level | What it is | What it is derived from |
|---|---|---|
| Reorder point | The stock level that triggers a new order | Average demand over the lead time, plus safety stock |
| Safety stock | Buffer against demand and supply variation | Demand variability, lead-time variability, target service level |
| Order quantity | How much to order when triggered | Order cost, holding cost, supplier minimums and price breaks |
What we deliver
Working outputs your planners use, not a one-off analysis.
- A reorder point and safety stock table for every item and location, with the target service level each is set to.
- An ABC/XYZ classification of your catalogue, with a recommended review policy for each of the nine groups.
- An excess and obsolescence list ranked by the capital it would release, with the reason each item appears.
- Measured lead times taken from your receipt history, compared against what suppliers quote.
- A rebalancing plan where you hold stock in more than one location.
- A monthly refresh that recalculates the targets as new history arrives.
How it works
-
Pull the history
Sales or consumption history, the item master, and purchase receipts. Exports are fine. We check them for the problems that distort stock calculations — duplicate item codes, unit-of-measure changes, one-off bulk orders that should not count as demand.
-
Classify the catalogue
ABC by value, XYZ by predictability. This decides how much attention each item deserves and which calculation method suits its demand pattern.
-
Measure real lead times
From order date to receipt date in your own records, including the variability. This is usually the single biggest correction to existing targets.
-
Set targets to a service level you choose
You decide what service level each class should achieve. We show the inventory cost of each choice so the trade-off is explicit rather than implied.
-
Hand over and automate
Targets load into your ERP or planning sheet, the refresh runs on a schedule, and your team gets the documentation and the training to adjust it.
What you need to get started
Three exports and a conversation.
- Item-level sales or consumption history, ideally two to three years.
- Your item master, with whatever attributes exist.
- Purchase order and receipt history, so lead times can be measured.
Spreadsheets and ERP exports are fine. If any of this does not exist yet, that is what the data engineering service is for, and the assessment will say so.
Typical outcomes
Described qualitatively. The size of the effect depends entirely on where you start.
- Capital released from items that were not protecting service.
- Better availability on the items that cause the most disruption when they run out.
- Fewer expedited orders, because the reorder point triggers early enough.
- A documented, reviewable basis for every stock target, instead of a number inherited from a previous planner.
Common questions
How is safety stock actually calculated?
The standard approach multiplies a service-level factor by the variability of demand over the replenishment lead time. In practice the hard part is not the formula — it is getting honest lead times and recognising that demand for many industrial items is intermittent, which breaks the normal-distribution assumption the textbook formula rests on. We use methods suited to each demand pattern and document which applies where.
What is ABC/XYZ analysis?
ABC classifies items by value or volume, so you know where the money is. XYZ classifies them by how predictable their demand is. Combining them gives nine groups, and each deserves a different policy. A high-value, highly predictable item should be managed tightly; a low-value, erratic one should not consume a planner's attention at all.
Will this just tell us to hold less stock?
No. It almost always says hold less of some things and more of others. The usual finding is that total inventory is roughly right but badly distributed — capital tied up in slow items while the ones that cause stockouts are under-covered. Rebalancing is where most of the value is.
Our lead times are unreliable. Does that break it?
It is the reason to do this. Lead-time variability is a direct input to safety stock. If a supplier quotes four weeks and delivers anywhere between three and nine, the stock you need to maintain service is materially higher than the quoted figure suggests. We measure actual lead times from your receipt history rather than using the quoted ones.
How often do the targets need to be reviewed?
Monthly or quarterly for most operations, and automatically where we can. Setting targets once and leaving them is the most common reason inventory drifts. We build the recalculation into the pipeline so it happens without anyone remembering to do it.
Find out what your data can already tell you
A free 30-minute data readiness assessment. We look at what you have, tell you what it can support today, and name the first thing worth building. No preparation needed.