Spare Parts and Inventory Across Sites and Technician Vans

Spare Parts and Inventory Across Sites and Technician Vans

In part five we built cost control on four streams, and about one of them we said something quickly: parts are issued against the work order, so their cost lands on the right project. The statement is correct, but in most contracting and maintenance companies it is not executable, because the part itself has no known location before it is installed.

This is the single biggest hole in the sector: a company that knows its main warehouse balance precisely, but does not know what sits in the site container, what is in the van of a technician who has been on the road for three years, or what happened to the part removed last week. The stock exists — it is simply not known. And the difference between existing and known is the difference between an asset and a deferred loss.

Inventory here is not one warehouse

Most systems assume a central warehouse with goods in and goods out. The reality of a maintenance company is different: the main warehouse is one of several stock locations, and the one that consumes the most is the one moving down the road.

There are four real locations in this sector:

• The main warehouse. The receiving point from suppliers and the source that feeds every other location.
• The site container or project store. Stock held close to the work, usually dedicated to one contract, and it needs an owner by name rather than by job title.
• The technician van. The most important location and the hardest, because it moves, it consumes daily, and usually nobody asks about its balance until month end.
• The technician's personal tool custody. Tools and small parts handed over once and treated afterwards as if they had evaporated, when in fact they are company money.

Every one of these needs the same minimum: a clear name, a named owner, a balance per item, and recorded movements in and out. That is the whole difference between a stock location and a black hole — nothing more elaborate is required.

And one rule here corrects the most commonly wrong report in the sector: moving a part from the warehouse into a technician's van is not an issue, it is a transfer between two locations. The issue happens the moment the part is installed on an asset under a work order. Anyone who treats the transfer as an issue reads inflated cost in the month they stocked their vans and a false low in the month those parts were actually consumed — then prices future work off a curve with no relationship to reality.

Issuing against the work order: the point that settles everything

Every part that leaves stock to be installed must leave against a work order number, not against a supervisor's verbal request or a client's name. That single procedure buys four things at once:

• The cost lands on the correct contract or project at the moment of issue, not two months later in a reconciliation.
• The asset history becomes complete: this unit had this component replaced on this date — which is what later settles the repair-versus-replace argument we covered in part four.
• The payment application rests on a document: completed work and installed materials evidenced by number, rather than a measurement assembled from memory at month end.
• Warranty becomes claimable, because you know when the part was installed, in which asset, and from which supplier.

The objection from the field is valid and must be solved rather than ignored: a technician at a site with an urgent breakdown will not open a screen and search for an item by name. The answer is not to soften the discipline but to shorten the procedure — a barcode or QR code on the item so it is issued with one scan from a phone, the van's own item list shown first because it is what gets used, and offline capability that syncs later exactly as we arranged in part three. Any stock procedure that adds more than a minute to a technician on site will be worked around, and a workaround means the black hole is back.

Then there is the half of the movement that always gets forgotten: the removed part. Every component taken out of an asset has one of three destinations, and it must be recorded — scrapped, repaired and returned to stock as a refurbished item clearly distinguished from new, or returned to the supplier under warranty. Companies that skip this pay for parts the warranty would have covered, and end up storing refurbished items nobody trusts because nobody knows where they came from.

One simple operational rule closes the loop: a work order does not close before its consumed parts are recorded. If it closes with no parts, the cost went to general expenses — and part five showed exactly where general expenses end up.

Minimum levels and reorder points: the parts that stop work

Inventory control does not mean treating every item with equal attention, which is the fastest way to exhaust a team for no return. Items fall into three practical categories, each with a different policy:

• Critical parts. Their absence stops the client's operation or breaches a time commitment in the contract. These are held regardless of carrying cost, because the price of not having one is not the price of the part — it is a delay penalty and a contract's reputation.
• Fast-moving parts. Filters, belts, electrical consumables. Individually cheap and extremely expensive when out of stock, because a stockout generates an entire repeat visit.
• Slow-moving, expensive parts. Ordered when needed rather than stocked, covered instead by a known supplier and an agreed lead time.

A reorder point is not a number written by instinct. It comes from three inputs: actual consumption over a period, the supplier's real lead time rather than the promised one, and safety stock covering the variability of both. Imported parts need a wider buffer, because their lead time is longer and less predictable, and conformity and customs procedures can add days nobody counted.

The most repeated mistake here is writing minimum levels once at go-live and never revisiting them. Consumption is not static: a new contract starts, a site ends, an equipment model is replaced — and levels set a year ago become the reason you stock what nobody uses and run out of what is used daily. Reviewing levels against actual consumption is part of running operations, not a one-off project.

That leaves the line no management enjoys: dead stock. Parts bought for a project that finished, or for equipment since replaced, still sitting on the shelf. This is not inventory, it is sleeping capital that keeps aging, and it needs an explicit decision — use it on another contract, return it to the supplier where the agreement allows, sell it, or write it off and recognise the effect. Leaving it in reports at full value makes the balance sheet prettier and the decision worse.

Cycle counting instead of one long night at year end

The full annual stock count is a procedure inherited from another era: it halts work for a day or more, accumulates a full year of errors into one batch, then corrects them with an adjustment entry while nobody can explain the variance, because its cause happened eight months ago and has been forgotten.

The practical alternative is cycle counting: counting a small slice of items every week instead of everything once, more frequently for critical and fast-moving items and less for slow ones. Work never stops, and a variance is discovered while it is days old rather than a year old.

The real purpose of counting is not to correct the number on the screen — it is to learn why it differs. Recurring variances have known causes: parts issued without recording under the pressure of an urgent breakdown, transfers between locations never documented, unit-of-measure confusion between a metre and a piece, or a receipt recorded at a quantity other than what actually arrived. Each cause has a procedural fix, and settling for a balance adjustment means the same variance returns next month at the same size.

Technician vans get counted too, and this is the most neglected point in the sector. A quick recurring count of van stock — preferably on a published schedule rather than as a surprise inspection — turns the van from a black hole into a controlled location, and protects the technician from an accusation they cannot disprove without a record.

From consumption to purchasing: closing the loop

Controlled inventory is worth little if it stops at the warehouse door. The loop closes when a genuine shortage raises a purchase request by itself: an item hits its reorder point and the system proposes the order at the right quantity, instead of the whole thing depending on a storekeeper noticing or a late call from a technician on site.

Between request and payment, four steps cannot be skipped:

• A supplier file holding prices, real lead times and a performance record: on-time delivery, share delivered conforming, and how quickly warranty claims are handled. Choosing a supplier on price alone costs more than it saves the day a late delivery stalls a contract.
• An approved purchase order before receipt, because buying without an order is the door through which expenses enter with no job number.
• A recorded goods receipt with quantity and condition, and a clear rejection of what does not conform rather than accepting it and discovering it on site.
• Three-way matching between the purchase order, the goods receipt and the supplier invoice before payment. This step alone prevents two common losses: paying for a quantity that never arrived, and paying twice for the same delivery.

On the regulatory side, the supplier invoice must be a tax invoice meeting the e-invoicing requirements of the Zakat, Tax and Customs Authority for input VAT deduction to be valid. Imported spare parts are subject to conformity and safety requirements under the Saudi Standards, Metrology and Quality Organization, and that is not paperwork trivia — it is a component of lead time that belongs in your safety stock calculation. Companies supplying government entities carry an additional layer of local content requirements in procurement, which affect supplier selection before price does.

One simple indicator measures the health of the whole loop: the share of emergency purchases. Every rush order at a higher price for faster delivery is usually evidence of a badly set reorder point, not of an energetic procurement team. When that share falls, inventory has started to be managed.

The Origami view

When we sit with a maintenance company, our question about inventory is not about systems. It is a number: how many items are in your busiest technician's van right now? Almost every time the answer is unknown, alongside a firm conviction that it is under control because the technician is trustworthy. The issue is not trust. It is that what is not measured is not managed, and an employee's honesty does not substitute for a record when the payment application or the warranty claim comes due.

As a technology company serving the Saudi contracting and maintenance sector, four rules apply in every delivery: start with critical and fast-moving items only and never attempt to catalogue everything from day one, because that is what kills projects during data entry; issuing against a work order is mandatory with no exceptions from the first week; moving stock into a van is a transfer, not an issue; and every technician's van is a stock location with a named owner. A fifth rule comes from the field: every extra second imposed on a technician on site buys a percentage of workaround, and control that is not followed equals no control at all.

Where this leaves you

Spare parts are not a side inventory topic in this sector — they are the condition that makes cost control real. Treat every site container and every van as a stock location with a balance and an owner, issue every part against a work order, set reorder points for the items whose absence stops work, count regularly to find the cause rather than to fix the number, and let consumption drive purchasing. Once that loop closes, a large share of repeat visits caused by nothing more than a missing part simply stops happening, and the project cost you read in part five becomes a number you can trust.

What remains is bringing all of it into one place that can be read in minutes. Part seven, the last, covers the manager's dashboard: the short list of indicators you actually run operations on, what action each one should trigger, where AI genuinely helps today and where it is still marketing, and how the series closes back on the seven leaks we started from.

Sources

• Zakat, Tax and Customs Authority — e-invoicing requirements, conditions for input VAT deduction on supplier invoices, and import procedures.
• Saudi Standards, Metrology and Quality Organization — conformity and safety requirements for products and imported spare parts, and their effect on lead times.
• Saudi Organization for Chartered and Professional Accountants — standards adopted in the Kingdom, including inventory measurement and the recognition of obsolescence and write-downs.
• Local Content and Government Procurement Authority — local content requirements in supply and contracting with government entities.
• Saudi Vision 2030 — the national direction on supply chain efficiency and operating spend in contracting and maintenance.