What is the difference between CRP and RCCP?

Rough-cut capacity planning runs before MRP, tests the master production schedule against a few key resources, and is fast and coarse. Capacity requirements planning runs after MRP, tests every work center in detail hour by hour, and is slower and far more precise. RCCP asks whether the plan is roughly feasible; CRP asks whether the released schedule is actually runnable.

What is a work-center load profile?

It is CRP's main output: a period-by-period chart for each work center showing required hours against the capacity line. Bars above the line are overloads where orders demand more hours than the center can supply; bars below are idle capacity. The profile turns a vague sense that a resource is busy into a specific, sized problem a planner can level by moving work or adding capacity.

Does CRP replace MRP?

No. CRP works after MRP, not instead of it. MRP decides what to make and when but assumes infinite capacity; CRP takes those orders and checks whether the work centers actually have the hours to do them. The two are complementary: MRP handles materials and timing, CRP handles the capacity reality MRP ignores, and together they produce an executable plan.

Why do CRP load profiles sometimes mislead?

Almost always because the time standards are wrong. A standard set years ago, before a process or machine changed, understates or overstates how long work really takes, so the load profile looks fine while the floor falls behind, or shows a false overload. The arithmetic is sound; the inputs are stale. Keeping routings and standards current is what makes CRP trustworthy.

How do you fix an overloaded work center in CRP?

By leveling the load. Options include pulling some of the overloaded period's work into a nearby period with idle capacity, adding a shift or overtime, offloading operations to an alternate work center, or adjusting the master production schedule so the demand is realistic. CRP does not choose the fix; it shows the exact place and size of the overload so the planner can act.

What Is Capacity Requirements Planning (CRP)?

Q: What is capacity requirements planning (CRP)?

Capacity requirements planning is the detailed process of converting MRP orders into the exact hours of work they demand from each work center in each period, using routings and time standards. It runs after MRP and produces a load profile per work center that lays required hours against available hours, so overloaded and idle resources show up before the schedule reaches the floor.

Q: What inputs does CRP need?

Three: the orders from MRP, both released shop orders and planned orders; routings, which name the sequence of operations and the work center for each; and time standards, the setup and run time per operation. CRP multiplies quantity by run time, adds setup, assigns the hours to the right work center and period, and sums the load. The output is only as good as those routings and standards.

How CRP turns MRP orders into hours of load on every work center using routings and time standards, and how it differs from rough-cut capacity planning.

Capacity requirements planning (CRP) is the detailed process of converting the orders from an MRP run into the exact hours of work they will demand from each work center in each time period, using routings and time standards, so overloaded resources show up before the schedule hits the floor.

Material requirements planning tells a plant what to make and when, but it assumes every machine has unlimited hours. CRP is the reality check that MRP skips. It takes the same orders, walks each one through its routing, and adds up the hours they will pile onto each work center, week by week, so a planner can see, in advance, which resources are overloaded and which are starving. This post defines CRP, shows the inputs it needs, contrasts it with rough-cut capacity planning, and explains what a load profile tells you. It is educational, not vendor advice, and names no products.

What is capacity requirements planning (CRP)?

Capacity requirements planning is the process of determining, in detail, the labor and machine capacity a set of production orders requires, translating planned and released orders into hours of work per work center per period using parts routings and time standards. It runs after MRP has generated the orders, and its job is narrow but essential: turn "here is what to make" into "here is exactly how many hours each work center must supply, and when," then compare that demand against each work center's available hours.

The output is a load profile for every work center, a period-by-period picture of required hours against available hours. Where required exceeds available, you have an overload the plan cannot honor as written; where required falls well short, you have idle capacity. CRP does not fix these; it exposes them, early enough that a planner can act, by moving work, adding a shift, offloading to another center, or pushing the master schedule back, before the shortfall becomes a missed shipment. It is the detailed, work-center-level counterpart to broader capacity planning.

CRP multiplies each order by the standard hours in its routing, sums the result per work center per period, and lays it against available hours. The comparison is the whole point: it makes overloads visible before they reach the floor.

What inputs does CRP need?

CRP needs three inputs, and the quality of all three decides whether the output is trustworthy. The first is the set of orders from MRP: both released shop orders already on the floor and planned orders MRP intends to release. The second is routings, the defined sequence of operations each product passes through, naming the work center for each operation. The third is time standards, the setup and run time per operation that turn a quantity into hours.

Multiply an order's quantity by the run time in its routing, add the setup, assign it to the named work center in the right period, and repeat for every order and operation, and you have the load. This is arithmetic, not magic, which is exactly why it is only as good as the standards behind it. If a routing says an operation takes six minutes and it really takes nine, CRP will understate the load by half again and hand the planner a rosy picture that the floor quietly disproves. Accurate, maintained routings and standards are the price of a CRP you can trust.

What is the difference between CRP and rough-cut capacity planning?

CRP and rough-cut capacity planning (RCCP) both check capacity, but at different times, different levels of detail, and against different plans. RCCP runs before MRP, at the master-schedule level, testing the master production schedule against a handful of key resources or bottlenecks to see whether the plan is roughly feasible. It is fast, coarse, and meant to catch big infeasibilities early, so no one commits to a master schedule the plant plainly cannot support.

CRP runs after MRP, at the shop-order level, testing every work center in detail, hour by hour and period by period. It is slower and far more precise, and it catches the specific overloads RCCP is too coarse to see. Think of RCCP as the sanity check on the plan and CRP as the detailed check on the orders that plan produced. A plant often uses both: RCCP to keep the master schedule honest, CRP to make the released schedule executable.

Same goal, two altitudes. RCCP is the coarse pre-check that keeps the master schedule realistic; CRP is the detailed post-check that makes the released orders buildable. Most plants run both.

What does a work-center load profile tell you?

A load profile is CRP's main product: a bar-by-period chart for each work center showing required hours against the capacity line. Bars above the line are overloads, periods where the orders demand more hours than the center can supply, and they are warnings that something must give. Bars below the line are underloads, idle capacity you could fill by pulling work forward or that signals you are carrying more resource than the plan needs.

The value is that it turns a vague worry into a specific, actionable picture. Instead of "the mill seems busy," a planner sees "the mill is 140 percent loaded in week 3 and 60 percent in week 5," which points straight at the fix: level the load by pulling some week-3 work into week 5. This load-leveling is the same smoothing instinct behind heijunka and disciplined lean manufacturing applied to the capacity plan. CRP does not decide the fix, but it gives the planner the exact place and size of the problem.

How do you run capacity requirements planning?

CRP follows a consistent sequence, whether it runs inside a planning system or a planner works it through by hand:

Take the orders from MRP. Gather all planned and released orders that will consume capacity in the horizon you are checking.
Pull each order's routing. For every order, list the operations and the work center each one runs on.
Apply the time standards. Multiply quantity by run time and add setup time to convert each operation into hours of work.
Assign hours to periods. Place each operation's hours in the correct time bucket based on when the order is scheduled.
Sum load per work center. Total the hours demanded of each work center in each period to build its load profile.
Compare against available capacity. Lay the load against each center's available hours to find overloads and idle time.
Level and re-plan. Resolve overloads by moving work, adding capacity, offloading, or adjusting the master schedule, then re-run to confirm the plan is now feasible.

Dimension	Rough-cut capacity planning (RCCP)	Capacity requirements planning (CRP)
When it runs	Before MRP	After MRP
Plan it tests	Master production schedule	Planned and released orders
Resources checked	A few key resources or bottlenecks	Every work center
Level of detail	Coarse, fast	Detailed, hour by hour
Question it answers	Is the plan roughly feasible?	Is the released schedule runnable?

What do the standards and data say?

Context from standards bodies and primary sources:

CRP, RCCP, routings, and time standards are defined in the planning body of knowledge maintained by the Association for Supply Chain Management (ASCM/APICS) which frames CRP as the detailed capacity check that follows MRP and RCCP as the higher-level check on the master schedule.
Academic operations treatments lay out the same structure: a university chapter on capacity planning describes translating open and planned orders into hours of work by work center by period using routings and standard times.
The base of resources to load is large: the Bureau of Labor Statistics counts roughly 13 million manufacturing jobs in the United States, spread across work centers whose real hours CRP is meant to respect.

The practical takeaway: CRP is arithmetic on routings and standards, so its honesty rises and falls with how current and accurate those two data sets are kept.

Where CRP lives or dies: the data underneath

CRP is only as trustworthy as the routings and standards it multiplies, and in many plants those are the most neglected data in the system. Time standards set years ago, before a process changed or a machine was rebuilt, quietly lie about how long work really takes, and a load profile built on them looks reassuring while the floor drowns. The failure is never the arithmetic; it is that the standard said six minutes and the job takes nine. Harmony is an AI-native layer that connects machines, software, and paperwork into one operational layer, with no rip-and-replace, so the signals CRP depends on, real run times, real setup durations, real work-center availability, become one current record instead of standards frozen in a spreadsheet. AI search returns cited answers across those records, so a planner can ask what an operation actually took over the last month or how much time a work center really loses to changeovers and downtime, then feed a true number back into the plan. It is the same paper-to-digital move Harmony makes elsewhere in the plant (see the CLS case study), and Harmony's digital workflows keep the load profile connected to what the floor is actually doing. It pairs directly with higher-level capacity planning above it and with the master production schedule it ultimately protects, since a detailed load check is only as sound as the standards and the plan feeding it.