Skip to content
AnrilX

Supply chain & operations

The question that crosses four modules

Whether an order is at risk depends on production, procurement and the warehouse at the same time. Each of those has its own screens, and none of them answers the question on its own.

What gets in the way

Exposure is never inside one module

Whether a production order is at risk depends on what it needs and on whether that has arrived: demand on one side, goods movements and open purchase orders on the other. Every screen shows you half of it, and joining the halves is a spreadsheet somebody rebuilds each week.

The expensive problems are absences

Stock that stopped moving. A location nobody has counted. A piece of equipment quietly generating a notification a month. Nothing alerts on a thing that fails to happen, so these are found when somebody goes looking, which is to say, late.

Every figure is an aggregate nobody stores

Behind plan, failure frequency, ageing, first-pass yield; none is a field. Each is a calculation over a window, which is why the answer usually arrives as a number one person computed and everybody else takes on trust.

Where you start, and what you get

Questions you can ask on day one

  • “Which orders are at risk from a material that has not arrived?”
  • “Where is stock sitting that nothing has ordered in six months?”
  • “Which equipment costs more in downtime than it would to replace?”
  • “Which defects keep coming back from the same supplier?”
  • “Where is production running behind plan?”
  • “Which work centres are the constraint this month?”
  • “What moved yesterday, and where did it go?”

What you get back

  • Cross-module questions answered as one figure, with each side naming the entity it came from
  • Absences found deliberately (stock without movement, equipment without a count) rather than stumbled on
  • Aggregates computed in code, so the same question asked twice returns the same number
  • A saved operations view that re-reads live SAP every time it opens

Four modules, one question

“Which orders are at risk” is the question operations actually asks, and it is unanswerable inside any one module. The demand sits with the production order. Whether it has been satisfied sits in goods movements and open purchase orders. What happens if it slips sits with the customer commitment, in sales.

Each of those has a perfectly good screen. None of them has the answer, so the answer gets assembled by hand, usually on a Monday, usually by the same person, and usually just late enough that the useful window has closed.

Absence is the hard shape

Most operations reporting is about presence: what is held, what was made, what moved. Your system does that well.

The costly questions are the inverse. Stock nothing has ordered in six months. A machine that has generated nine notifications this year when the fleet average is two. A location that has not been counted since last spring.

You cannot filter for a row that does not exist. Each of those questions has to be constructed (what exists, against what happened in a window, and the difference), which is precisely the query that never gets written because it takes a specialist an afternoon and nobody has budgeted an afternoon for a hunch.

Why the arithmetic being in code matters here

Behind plan is planned minus confirmed. Ageing is last movement against today. Yield is passed over inspected. Every one of those is a calculation, and every one has a version that is subtly wrong: the right denominator, the wrong window, the previous month that had data rather than the previous calendar month.

AnrilX computes them in code rather than leaving them to a model, which means the same question asked twice returns the same number. That is a low bar, and it is the bar a figure has to clear before anybody schedules production against it.

One of these questions, answered

Which production orders are at risk from a material that has not arrived?

Twenty-three orders, covering 4,180 units due in the next fourteen days. Nineteen of them wait on the same component: TG-2207, whose purchase order is eleven days past its confirmed date. Fixing one receipt clears most of the exposure.

Order · material short Units at risk
19 orders · TG-2207 3,410
3 orders · TG-0915 620
1 order · TG-3302 150
23 orders 4,180 units

Source: A_ProductionOrder_2 and A_ProductionOrderComponent · open reservations with no goods movement · joined to I_PurchaseOrderHistory on Material for the receipt position · scheduled finish within 14 days

Assumed: at risk means the component is short against the reservation, not that stock is merely low. An order with enough on hand is excluded even where the purchase order is late.

Before a pilot

What it will not do for this role

The limits that bear on these questions specifically.

  • It reads the chain as your SAP records it. Where a step happens in a system that is not SAP, that step is a gap in the answer and is named as one rather than bridged.

  • Lead times are measured between document dates, not between physical events. Where your process records a movement late, the measurement inherits that lag and the answer names the dates it used.

  • A question crossing modules can only follow links SAP actually models. Where two documents are related in practice but not in the data, the platform reports the break rather than inferring it from matching dates and quantities.

Questions people ask us

Can it really answer across production and procurement at once?

Yes, and that is the point. It reads component demand against what has actually been received, and names the entity each half came from, so a risk figure can be checked rather than taken on trust.

Can it find stock that has stopped moving?

It reads what is held against what moved in a window you choose and reports the difference. It reports that as stock without recent movement rather than labelling it obsolete, which is a business judgement rather than a data one.

Does it replace our planning or maintenance systems?

No. It answers questions across what those systems have already recorded. It does not reschedule, re-plan or raise work, beyond a governed write that a person approves first.

Bring the question your reports cannot answer

Thirty minutes against a live SAP system we provide: no access to yours, nothing to set up. If it cannot answer, you find that out in half an hour rather than three months into a pilot.