Finite vs. infinite capacity scheduling in Business Central

If your Business Central production schedule looks right in the morning and wrong by lunch, the problem usually isn’t your planners. It’s an assumption baked into standard planning: infinite capacity.

What “infinite capacity” really means

When Business Central calculates order dates through MRP, it treats each work center as if it can run everything asked of it at the same time. It schedules backward or forward from a due date and hands you a start date — without checking whether the machine is already busy with three other jobs in that window.

That’s not a bug. Infinite-capacity planning is fast, and it’s exactly what you want for material planning: it tells you what to buy and make, and roughly when. The trouble starts when people treat those dates as an executable schedule. They aren’t. They’re a materials answer being asked to do a scheduling job.

Where it breaks on the floor

Three failures show up again and again:

• Overloaded machines. Five jobs are all “scheduled” onto the same cell for Tuesday. Reality only fits two. The other three silently slip, and nobody decided which.
• Material blind spots. A job is scheduled to start Monday, but a key component doesn’t land until Thursday. The plan never knew, so the job stalls in progress instead of being sequenced around the shortage.
• A plan that drifts from reality. Because the schedule was never feasible, the floor improvises. Within hours the office plan and the shop floor disagree — and every downstream promise is now guesswork.

What finite-capacity scheduling changes

Finite-capacity scheduling starts from the opposite assumption: a machine can do one thing at a time. Instead of stamping a date and hoping, it places each operation into an actual open slot on an actual resource, in sequence, respecting shift hours and the jobs already there.

That single constraint changes the output from a forecast into a plan you can run:

• No double-booking. If two jobs want the same machine at the same time, one of them moves. The trade-off is made explicitly, by the schedule, not accidentally, by whoever shouts loudest.
• The plan is executable. The start dates it produces are dates the floor can actually hit, because the capacity to hit them was checked before the date was assigned.
• Sequence becomes a lever. Once you’re placing real operations on real machines, you can optimize the order — shortest jobs first, earliest due dates first, or minimizing changeover between jobs.

Adding material awareness

Finite capacity fixes the machine side. The other half is material. A truly trustworthy schedule won’t place a job before the parts to build it are available — it treats the earliest material-availability date as a hard floor, not a suggestion. Capacity tells you whether the machine is free; material tells you whether the job can even start. You need both to get a date worth promising.

The bottom line

Standard Business Central planning answers “what should we make and buy?” extremely well. It was never designed to answer “what can we actually run today, on these machines, with the material we actually have?” That’s a finite-capacity, material-aware question — and answering it is the difference between a schedule that describes a wish and one that describes the day.

Pacer Production Scheduler adds exactly that layer inside Business Central: finite-capacity, material-aware scheduling that produces a plan your shop floor can follow. See how it works →