APS vs MRP: when your ERP isn't enough anymore

The symptom you see every Monday

Monday morning planning meeting. On the table: the MRP printout from your ERP, ten spreadsheets, two handwritten notes. You decide what to launch. On Wednesday a key customer pushes back a delivery; on Thursday another order becomes urgent. Monday's printout is already worthless.

It isn't a people problem. It's a tool problem.

What your ERP's MRP actually does

Classic Material Requirements Planning:

• takes the demand (orders + forecasts);
• explodes the bills of materials;
• proposes production and purchase orders;
• at infinite capacity: it assumes resources are always available.

It's a genius logic from the 1970s — when lots were big, mixes small, and setups negligible. It still works in many contexts, but it has a specific blind spot: it doesn't tell you whether what you planned is feasible.

What an APS does

An APS works at finite capacity. It doesn't just say what to produce: it says when, on which resource, in what order. It considers:

• the real hours available on each work center;
• sequence constraints (setups, color changes, tooling);
• actual on-hand stock of components and WIP;
• commercial priorities and customer SLAs;
• operators and required skills, when relevant.

The output isn't a Monday printout. It's a living plan that recalculates in minutes whenever something changes.

The concrete signals you need an APS

It's not a philosophical choice. You make it when you see two or more of these together:

1. Promised dates are unreliable. When sales calls production to "see if we can make it", they're not using the system — they're doing back-of-envelope math.
2. WIP grows but on-time delivery doesn't. You're pushing material onto the floor hoping "something will start". Classic infinite-capacity symptom.
3. Rush orders are 30%+ of what you launch. When urgent is the rule, you're not planning: you're firefighting.
4. Sequence changes cost a lot. Setups, color changes, tooling changes — MRP doesn't see them, APS does.
5. Average utilization is "high" but bottlenecks exist. MRP tells a comforting lie, APS shows the constraint.

What actually changes after a well-run APS rollout

Not brochure numbers — numbers we've seen on real projects in Italian SMEs between 30 and 300 employees:

• OTD from 70% to 85-92% in 6-9 months.
• Lead times cut by 25-40%, mostly because you launch less and better.
• Average WIP down 15-25%.
• The slowest machines — the "bottlenecks", the ones that limit the throughput of the whole line — produce 10-20% more in the same time, because setups and sequences are planned better.
• Replanning time: from half a day to minutes when an urgent order shifts the picture.

These numbers don't come from the software. They come from the fact that, for the first time, the planner has a tool that matches the factory's reality.

What no one talks about

An APS is not plug-and-play. The things that make the difference between a winning project and a shelfware one:

• Master data quality. Wrong BOMs, lump-sum routing times, off-scale nominal capacities — APS amplifies hidden errors, it doesn't forgive them. The first real month of a serious project is almost always data cleanup.
• ERP / MES / WMS integration. Without real shop-floor feedback, APS plans blind.
• Planner team adoption. If the planner keeps overriding the system "because he knows", the tool will never sync with the factory. You need an onboarding and trust journey, not a one-week training.

When you do NOT need an APS

For professional honesty: if you make few SKUs, in large lots, with trivial setups, for customers who accept loose dates, your ERP's MRP is enough. Spending on an APS in that context is the opposite mistake — and you'll see it 12 months in, with "why doesn't anyone use it?".

The question to ask yourself

"How many planning decisions, in a week, are made on the phone or standing in front of a spreadsheet?"

If the answer is "many", the problem isn't your people's willingness. It's that they don't have a tool that matches the reality they manage.

And when they finally do, they stop being firefighters. They start being planners.