RCCP vs Finite Capacity Scheduling: Why a Green Capacity Plan Still Produces a Late Schedule

The monthly capacity check came back green. The critical resources absorbed the load, the master schedule was approved, and the sales commitments followed. Six weeks later the same lines are behind, and the team rebuilds the sequence every morning. Nobody miscalculated, which is exactly what makes it hard to explain in a steering meeting.

Rough-Cut Capacity Planning (RCCP) is a capacity validation technique that verifies whether a Master Production Schedule (MPS) is achievable across a limited set of critical resources. It works from a bill of resources and reasons in aggregated time periods, usually weekly or monthly buckets. It is a gross capacity technique: it does not consider scheduled receipts or on-hand inventory, and it is normally restricted to key resources only.

Finite capacity scheduling answers a different question entirely. It does not calculate how many hours a period requires. It determines in what order, on which resource, and at what time each operation runs, under the full set of real shop-floor constraints.

Both are correct. They simply do not measure the same thing.

What is Rough-Cut Capacity Planning (RCCP)?

RCCP sits at a specific point in the MRP II hierarchy. S&OP sets aggregate volumes, the MPS breaks them into items and periods, RCCP tests whether that MPS is supportable, and MRP then calculates material requirements. It runs upstream of any execution logic, at a moment when production orders do not yet exist.

The mechanism is deliberately simple. A bill of resources links each item, or each product family, to an hours consumption on the resources considered critical. The system multiplies MPS quantities by those ratios, accumulates the result per period, and compares the resulting load against declared capacity.

The output is not a plan. It is a verdict on supportability, resource by resource and period by period, and it functions as a negotiation instrument between commercial and operations. When an overload appears, the levers are well known: move quantities within the master schedule, add a shift, open overtime, subcontract, or bring in an additional machine.

That is the full value of RCCP, and also its boundary. It informs commitment decisions, not sequencing decisions.

What is finite capacity scheduling, and how is it different from a capacity check?

A capacity check answers "how many hours does this need, and do I have enough." Finite capacity scheduling answers "in what order, on which machine, at what time." This is not a difference in precision. It is a difference in kind.

Finite capacity scheduling produces a directly executable artifact: an operation-level sequence with dates and times, across every constrained resource, respecting shift calendars, operator qualifications, routing dependencies, sequence-dependent setup times and material availability at the same moment. Every task in the output can be released as it stands.

A green RCCP never promised that, and structurally could not. The distinction reads more clearly dimension by dimension.

RCCP vs finite capacity scheduling: what actually differs?

This distinction is a sharper version of the line that separates production planning from detailed scheduling. RCCP is the control instrument of the upper layer. It does not reach down into the lower one.

Why does a valid RCCP still produce an infeasible shop-floor schedule?

Because accumulated hours over a period structurally hide everything that depends on the order of operations. Six causes come up again and again.

  • Sequence-dependent setup times. Time lost in changeover depends on which product ran before. That information does not exist inside an aggregated bucket, where setups are at best averaged into the bill of resources ratios.
  • Resources outside the scope. RCCP covers the critical resources identified at a point in time. A resource judged non-critical can become the real constraint in a given week, depending on mix.
  • Smoothing inside the bucket. Forty hours of load in a forty-hour week assumes perfectly even flow, with no queueing between operations and no upstream dependency.
  • Simultaneous multi-resource constraints. One operation may require a machine, a qualified operator, a tool and a utility at the same instant. Aggregated load treats those resources separately.
  • Routing dependencies. Queue, cooling or curing time between two operations shifts real flow without ever appearing in a cumulative load figure.
  • Actual material availability. RCCP is a gross capacity technique that considers neither scheduled receipts nor on-hand inventory.

Each of these gaps is small on its own. Accumulated across a high mix and a bottleneck that moves with the load, they are enough to turn a supportable plan into an unmanageable week.

When is RCCP enough, and when do you need finite capacity scheduling?

RCCP on its own remains appropriate where the production environment is stable and predictable:

  • narrow product mix, consistent from period to period
  • few changeovers, or changeovers of broadly uniform duration
  • a single, known bottleneck that stays in the same place
  • long decision horizon, with few interruptions
  • delivery commitments made by family rather than by order

Finite capacity scheduling becomes necessary as soon as those conditions break down:

  • high mix, short runs, significant week-to-week variability
  • sequence-dependent setups that are material to output
  • a bottleneck that shifts with the weekly mix
  • intra-day replanning that has become routine
  • order-level delivery dates that commit the customer
  • planners rebuilding the sequence every morning in a spreadsheet outside the system

That last symptom is the most diagnostic. When a parallel spreadsheet becomes the shop floor's source of truth, the gap between the control layer and the execution layer is no longer being closed by a tool. It is being closed by a person.

How do RCCP and finite capacity scheduling work together?

The two are not alternatives. They occupy two floors of the same decision chain, and performance depends on the quality of the handover between them.

  1. S&OP sets aggregate volumes by family across the tactical horizon.
  2. The MPS breaks those volumes into items, quantities and periods.
  3. RCCP tests the MPS against critical resources and triggers the arbitration if an overload appears, before any commitment is made.
  4. Finite capacity scheduling converts the validated MPS into a dated sequence under the full constraint set, and becomes the plan the shop floor executes.
  5. Execution feedback, through MES or shop-floor reporting, recalibrates both the running sequence and the bill of resources ratios used by RCCP.

Step 5 is the one most often missing. Without execution feedback, bill of resources ratios stay frozen on old assumptions and the capacity check quietly loses its predictive value. The bridge between the tactical plan and the schedule is not a file transfer. It is a loop.

One scoping point worth stating plainly: demand forecasting feeds S&OP at the top of this chain. It is never an output of scheduling, which works on orders rather than probabilities.

What does this mean for your planning architecture?

For most manufacturers the problem is not choosing between RCCP and finite capacity scheduling. It is the model break between them. The capacity check lives in the ERP, the real sequence lives somewhere else, and the two rest on capacity assumptions that drift apart without anyone measuring it. The division of responsibility between APS, ERP and MES is where that break usually shows up first.

MangoGem APS Optimizer addresses exactly this point by combining planning and scheduling in a single system. It supports Master Production Scheduling, which typically operates at a rough-cut, long-term planning level, while its finite capacity scheduling handles detailed and precise execution. The capacity verdict and the executable sequence then come from the same constraint model.

That continuity changes the nature of the upstream check. When sequence-dependent setups, operator qualifications, multi-resource constraints and real calendars are already modeled, validating the master schedule stops being an approximation built on ratios. It becomes a test against what the plant can actually run.

The constraint model is parametric and configured without custom development, so planning and production teams can evolve the rules at the pace of the plant. Integration with ERP and MES sustains the loop described above, which means execution feedback informs both the replanning and the capacity assumptions behind it.

 

Frequently Asked Questions

1. Is RCCP the same as Capacity Requirements Planning (CRP)?
No. RCCP validates master production schedules upstream, against critical resources only. CRP validates the plans produced by MRP and checks capacity across all resources required by the material plan. CRP is more detailed than RCCP, but it remains a load calculation, not a schedule.

2. Can an APS replace RCCP?
An APS can produce both the aggregated capacity check and the detailed schedule from a single model, which removes the data break between the two layers. The decision itself remains: choosing between moving volume, opening a shift or subcontracting is a management call that the tool informs rather than makes.

3. Why does RCCP ignore changeover time?
Because it reasons in cumulative load per period, with no notion of sequence. A sequence-dependent setup only exists in relation to the preceding product, and that information has no representation at bucket level.

4. Does finite capacity scheduling require a demand forecast?
No. Scheduling works on firm orders and resource constraints. Demand forecasting is an input to the planning chain, upstream of S&OP and the MPS, never an output of scheduling.

5. Can we keep RCCP in the ERP and add an APS only for scheduling?
This is a common and workable configuration. The point to watch is consistency of capacity assumptions between the two systems. Without a mechanism to recalibrate bill of resources ratios from actual execution, the gap between the upstream verdict and downstream feasibility widens silently.

 

To know more about RCCP vs Finite Capacity Scheduling: www.mangogem.com.