What is Capacity Requirements Planning (CRP)?
Capacity Requirements Planning (CRP) is a detailed manufacturing process that calculates the labor and machine hours needed at each work center to execute the orders generated by a Material Requirements Planning (MRP) system. It verifies that scheduled production is feasible given current resources and exposes bottlenecks before they disrupt delivery.
- Detailed feasibility check: CRP converts MRP planned orders into work-center hours by period, so a master schedule cannot quietly overload a single bottleneck.
- Four inputs drive accuracy: Bill of materials, routings, work-center capacity, and lead times set the ceiling on how reliable CRP output can be.
- RCCP and CRP are not the same: Rough-cut capacity planning sanity-checks the master schedule on key resources; CRP runs after MRP and covers every work center.
- Industry-wide capacity sits below trend: U.S. manufacturing capacity utilization averaged 76.6% in 2024, leaving most plants with structural headroom, so CRP gains usually come from balancing load, not adding capacity (Federal Reserve G.17, 2025).
Why CRP exists in the MRP stack
CRP exists because MRP can produce a plan that looks feasible on paper but overloads a single work center in week three. Master scheduling and MRP focus on what to make and when; CRP is the layer that asks whether the people and machines are actually there to do it.
Without CRP, manufacturers discover infeasibility on the shop floor, where late orders and overtime become the only recovery options.
The process keeps capacity and demand in balance, which in turn protects on-time delivery, prevents idle machines, and gives planners early warning when a quarter's mix is heavier on one line than the previous quarter's. This is also where CRP intersects with operating goals: a strategy execution cadence built on quarterly outcomes depends on a feasible production plan underneath it.
The CRP process, step by step
The Capacity Requirements Planning process runs as a continuous cycle with four steps:
- Workload determination. Translate MRP planned and released orders into hours of work, per work center, per time bucket. This step uses parts routings and time standards.
- Capacity assessment. Measure the available capacity at each work center, accounting for shifts, downtime, scheduled maintenance, and labor availability.
- Scheduling. Align the workload with available capacity, identify overloaded periods, and resolve them by re-sequencing orders, moving work to alternate centers, adding overtime, or pushing the MPS back.
- Monitoring. Track actual output against plan, feed variance back into routing standards, and trigger replans when reality drifts from the schedule.
The output is not just a feasible schedule but also a set of lead indicators that surface capacity stress weeks before it shows up as a late shipment.
CRP, RCCP, and MRP: how they fit together
The three planning layers are often confused because they all sit inside MRP II. They run at different levels of detail and at different points in the cycle.
Aspect | Rough-Cut Capacity Planning (RCCP) | Material Requirements Planning (MRP) | Capacity Requirements Planning (CRP) |
|---|---|---|---|
Inputs | Master Production Schedule, key resources | MPS, BOMs, inventory on hand | MRP planned and released orders, routings |
Scope | Bottleneck and key work centers only | All materials and components | Every work center and labor pool |
Question answered | Is the master schedule plausible? | What materials, in what quantity, by when? | Can each work center actually run the load? |
Timing | Before MRP runs | After MPS is approved | After MRP generates orders |
Level of detail | High-level resource profile | Component-level demand | Hour-by-hour work-center load |
RCCP catches infeasibility early and cheaply; CRP catches it in detail before orders hit the floor. Skip CRP and you rely on the MRP plan being correct, which it often is not once routings and downtime are factored in.
The four inputs CRP depends on
CRP output is only as good as four underlying datasets:
- Bill of Materials (BOM). The complete list of materials, components, and sub-assemblies that make up each product.
- Routings. The sequence of operations each product passes through, the work center for each operation, and the standard time per operation.
- Work centers. The defined production resources (machines, lines, cells, labor pools) with their available hours per period.
- Lead time. Procurement, queue, setup, run, and move times that determine how long an order takes from raw material to finished good.
If any one of these is wrong, the entire capacity plan is wrong in the same direction. Most CRP implementations that fail do so because BOMs and routings are stale, not because the algorithm is flawed.
What CRP delivers when it works
When CRP is in place and the input data is clean, it changes how a plant operates rather than just how it plans:
- Higher and steadier resource utilization. Resources such as labor, machinery, and materials run closer to their planned load, reducing waste and idle time.
- More reliable production schedules. Promise dates hold because the schedule was tested against real capacity before it was committed.
- Cost control. Overproduction, last-minute overtime, and expedited freight drop because problems surface in the plan rather than on the floor. This is where CRP feeds directly into cost reduction programs that target gross margin.
- Better capital decisions. Capacity stress over multiple quarters becomes visible, so investment in additional shifts, equipment, or outsourcing is evidence-based rather than reactive.
Across U.S. manufacturers, 49% rely on their ERP for production planning and scheduling (NetSuite, 2024). CRP discipline is now embedded in the system most plants already operate.
CRP in modern manufacturing
With the rise of Industry 4.0 and smart manufacturing, CRP has shifted from a periodic batch calculation to a near-real-time process. IoT sensors push live work-center status into the planning system, and machine-learning models predict capacity drift before it shows up in actuals.
Modern CRP systems can flag a likely bottleneck several days ahead and propose re-sequencing options on the fly. Cloud deployment also makes it practical for multi-plant manufacturers to view consolidated load across sites and rebalance work without waiting for a quarterly review.
Where CRP rollouts typically break
The benefits assume the implementation succeeds. In practice, four failure modes account for most of the disappointment:
- Stale routings and BOMs. CRP relies heavily on accurate master data. If standard times reflect an old machine setup or an old shift pattern, the plan is wrong on day one. Data hygiene is a key performance indicator for the planning team itself, not just the shop floor.
- Integration gaps. Bolting CRP onto an existing ERP and MRP environment is complex; broken integrations produce silent errors that planners cannot trace.
- Resource constraints during peaks. When demand spikes, no amount of planning creates capacity that does not exist. CRP exposes the constraint; closing it is a capital decision.
- Change management. Adopting a new CRP workflow requires change management strategies so planners trust the system's output and stop maintaining a parallel spreadsheet.
Using CRP in your planning cycle
Most plants do not need a more sophisticated CRP system; they need cleaner inputs and a tighter loop between CRP output and operational decisions. A practical sequence:
- Audit BOMs and routings before chasing algorithm improvements.
- Run CRP weekly during the production cycle and treat any work center above 90% planned load as an open risk.
- Tie CRP variance to a small set of lag indicators (on-time delivery, schedule attainment, overtime hours) so capacity stress becomes a metric the leadership team sees.
- Use a quarterly review of capacity headroom to inform outcome-based planning decisions on shifts, hiring, and capex.
