Foundations of Capacity Requirement Planning

Introduction

Capacity Requirement Planning (CRP) is the process of determining whether an organisation has sufficient productive capability to meet the workload generated by its production plan. It connects what the business intends to produce with the machines, labour, tools, facilities, and time actually required to produce it. CRP is therefore not merely a scheduling technique. It is a resource-feasibility discipline that exposes overloads, underutilisation, bottlenecks, and the timing of capacity-related decisions.

In manufacturing, a feasible material plan is not necessarily a feasible production plan. A bill of material may show that all required components are available, while the relevant work centre lacks the hours, skills, tooling, or maintenance availability needed to process them. CRP addresses this gap. It converts planned orders into time-phased capacity requirements and compares those requirements with available capacity.

CRP is especially important in environments characterised by product variety, uncertain demand, shared resources, long lead times, and expensive equipment. Its application extends from traditional job shops and repetitive assembly to process industries, maintenance operations, healthcare services, logistics, and digital production systems.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Explain the purpose, scope, and evolution of CRP.
2. Distinguish capacity planning from production planning, scheduling, and production control.
3. Define work centres, resources, loads, capacity units, and planning horizons.
4. Calculate basic available capacity and capacity requirements under stated assumptions.
5. Interpret capacity profiles and identify overloads, bottlenecks, and managerial responses.
6. Evaluate the limitations of conventional CRP and the role of contemporary digital systems.

1. The Meaning and Purpose of CRP

Capacity is the capability of a resource or system to perform work during a specified period under specified operating conditions. It is not always equal to theoretical machine speed. Capacity depends on working time, staffing, maintenance, breaks, setup losses, quality losses, product mix, and operating policies.

CRP is the medium- and short-term process of translating planned production into resource requirements. A typical CRP analysis answers four questions:

• What work is expected to arrive at each work centre?
• When will that work be required?
• How much capacity is available during each period?
• What action is needed when requirements and availability do not match?

The central comparison is between capacity required and capacity available. For a work centre and time period, the basic utilisation relationship is:

where U_t is utilisation in period t, R_t is required capacity, and A_t is available capacity. Both R_t and A_t must use the same unit, such as standard machine-hours or labour-hours. If utilisation exceeds 1, the work centre is theoretically overloaded; if it is below 1, unused capacity exists. In practice, a target below 100 percent is often necessary because variability, urgent work, breakdowns, and quality problems require protective capacity.

The capacity balance may also be expressed as:

where B_t is the capacity balance. A negative balance indicates an overload, while a positive balance indicates available spare capacity. These relationships are diagnostic rather than complete decision rules. A positive balance at one work centre cannot compensate automatically for a shortage at another unless work can be transferred, subcontracted, or rerouted.

Scope of CRP

CRP commonly covers:

• machines, labour, tools, fixtures, inspection equipment, and storage or handling resources;
• planned orders, released orders, rework, maintenance, and sometimes engineering or service work;
• capacity requirements by day, week, shift, or other planning bucket;
• finite or infinite capacity assumptions;
• corrective actions such as overtime, subcontracting, alternate routing, lot splitting, rescheduling, and investment.

CRP is narrower than strategic capacity planning but more detailed than aggregate production planning. It normally operates at the work-centre level and uses the production plan, master production schedule, material requirements plan, routings, work-centre calendars, and standard times as inputs.

2. Position of CRP within Production Management

Production management contains several related but distinct planning and control activities. Confusing them leads to poor decisions because each activity answers a different question.

Capacity planning versus production planning

Production planning determines the intended quantity and timing of output. Capacity planning tests whether the resources needed to achieve that plan are available. Production planning is therefore plan-oriented, whereas capacity planning is feasibility-oriented. The two processes are iterative: an infeasible production plan may require revised quantities, different timing, or additional resources.

Capacity planning versus scheduling

Scheduling assigns jobs to particular resources and positions them in a sequence. CRP normally aggregates or time-phases the workload sufficiently to reveal capacity problems without necessarily determining the exact sequence of every job. Scheduling must consider sequence-dependent setups, priorities, due dates, dispatching rules, and real-time disruptions. CRP supplies a capacity feasibility framework; scheduling converts that framework into an executable order of work.

Capacity planning versus production control

Production control monitors execution and responds to deviations. It records actual output, downtime, scrap, labour attendance, material shortages, and job progress. CRP is primarily prospective, whereas production control is primarily corrective and operational. Actual performance data should nevertheless update CRP standards and assumptions.

3. Evolution of Capacity Requirement Planning

Early production systems relied heavily on visual planning, supervisor experience, dispatch lists, and manual load charts. These methods can work in stable, low-variety environments but become difficult when many orders share resources.

The development of Material Requirements Planning (MRP) created a systematic way to calculate time-phased material needs from a master production schedule and bills of material. However, early MRP systems often assumed that the necessary capacity was available. CRP emerged to test the production plan against routings, operation times, and work-centre calendars.

The progression can be summarised as follows:

1. Manual load analysis: planners used boards, charts, and experience to compare work with available hours.
2. Closed-loop MRP: material plans were linked with capacity checks and feedback from the production floor.
3. Manufacturing Resource Planning (MRP II): materials, capacity, labour, finance, and management policies were integrated.
4. Enterprise resource planning (ERP): production data became connected with purchasing, sales, finance, maintenance, and distribution.
5. Advanced planning and scheduling: finite-capacity algorithms, optimisation, constraint management, and scenario analysis became more prominent.
6. Digital and cyber-physical planning: real-time machine data, cloud platforms, digital twins, predictive analytics, and artificial intelligence support more dynamic capacity decisions.

This evolution has not eliminated the basic CRP problem. It has increased the speed, scale, and data richness of the analysis. The reliability of an advanced system still depends on accurate routings, realistic standards, disciplined transaction recording, and appropriate managerial judgement.

4. Core Elements of a CRP Model

4.1 Work centres

A work centre is a production resource or group of similar resources for which capacity is planned and measured. It may be one machine, a group of machines, a labour pool, an inspection area, a maintenance crew, or a department with common capability.

The level of aggregation is a managerial choice. If a work centre is defined too broadly, important bottlenecks disappear inside departmental averages. If it is defined too narrowly, data maintenance becomes burdensome and planners may focus on insignificant fluctuations. A useful work-centre definition usually reflects shared capability, common calendars, similar cost or performance characteristics, and a meaningful managerial responsibility.

A work centre record commonly includes:

• resource identity and description;
• number of machines or operators;
• operating calendar and shifts;
• efficiency and utilisation assumptions;
• setup, run, queue, and move-time information;
• alternative resources and routing rules;
• maintenance and planned downtime;
• cost rates and performance measures.

4.2 Resources

Resources are the means required to perform work. They include equipment, direct labour, indirect labour, tools, dies, fixtures, inspection capacity, energy, and specialised facilities. A CRP model must identify whether resources are substitutable. Two machines may have similar nominal rates but different capabilities, quality performance, or tooling requirements.

Resource calendars are particularly important. A machine may be physically installed for 168 hours in a seven-day week but available for less because of planned shifts, holidays, maintenance, breaks, cleaning, and changeover policies.

4.3 Routings and operation times

A routing specifies the sequence of operations required to make an item. It normally identifies the work centre, setup time, processing time, queue or move assumptions, and sometimes inspection requirements.

A common operation-time model is:

where T_i is total time for operation i, S_i is setup time, q_i is quantity processed, and P_i is processing time per unit. The result has units of time. This equation assumes one setup for the lot and a constant processing time per unit. If multiple setups, batch transfers, rework, or sequence-dependent changeovers occur, additional terms are required.

For example, a lot of 120 units requires 45 minutes of setup and 3 minutes per unit. Total time is 405 minutes, or 6.75 hours:

T = 45 + (120 × 3) = 405 minutes.

If the routing record incorrectly omits setup time, the capacity requirement is understated by 45 minutes. In high-mix manufacturing, such omissions can create systematic overloads.

4.4 Loads

A load is the amount of work assigned or expected to arrive at a resource during a specified period. Loads may be expressed as hours, standard minutes, labour-hours, machine-hours, units, kilograms, litres, or another capacity unit.

A work-centre load for a period can be represented as:

where R_t is total required capacity in period t, T_jt is the capacity requirement contributed by job or operation j, and n is the number of relevant operations. The summation is valid only when all requirements are expressed in comparable units and assigned to the same period and resource.

Load may be classified as:

• planned load: generated by planned orders;
• released load: generated by orders authorised for production;
• actual load: work already performed or recorded;
• finite load: limited by explicit resource availability;
• infinite load: calculated without preventing assignments that exceed capacity.

4.5 Capacity units

A capacity unit is the measurement used to compare work requirements with resource availability. Common units include machine-hours, labour-hours, standard hours, units per shift, tonnes per day, and batches per week.

The unit should reflect the constraint. Machine-hours are suitable when machine time is limiting; labour-hours may be more appropriate when skilled labour is scarce. Physical units can be misleading when products have different processing times. For example, producing 100 simple parts and 100 complex parts does not imply equal capacity consumption.

4.6 Available capacity

Available capacity is the productive time or output that a resource can provide during a planning period. A simple time-based expression is:

where N is the number of equivalent resources, H_t is scheduled hours per resource, S_t is the fraction of scheduled time available after planned non-operating periods, E_t is expected efficiency, and Q_t is the expected quality or yield factor. Each factor except N is dimensionless, so A_t has units of hours.

This formulation assumes that the factors are defined consistently and that their multiplicative use is justified. Some organisations instead subtract downtime categories directly. Double-counting must be avoided. For instance, if efficiency already includes minor stoppages, those stoppages should not be deducted again.

Worked example. A work centre has two equivalent machines. Each is scheduled for 40 hours per week. Planned breaks and maintenance leave 90% scheduled availability, expected efficiency is 85%, and expected yield is 98%. Then:

A = 2 × 40 × 0.90 × 0.85 × 0.98 = 59.98 hours, approximately 60.0 hours.

If the planned load is 63 hours, the nominal overload is approximately 3 hours and utilisation is 63 ÷ 59.98 ≈ 1.05, or 105%.

4.7 Time fences and planning horizons

A planning horizon is the future period covered by a plan. CRP commonly uses several horizons:

• strategic horizon: capacity investments, facilities, technology, and workforce capability;
• tactical horizon: aggregate staffing, subcontracting, shifts, and major maintenance;
• operational horizon: order release, overtime, sequencing, and daily balancing;
• execution horizon: real-time dispatching and response to disruptions.

A time fence limits the degree to which plans may be changed. Near-term periods may be frozen to protect material commitments and production stability, while later periods remain flexible. A frozen zone improves execution discipline but can make short-term overloads harder to correct. A completely flexible plan may appear feasible on paper while creating instability, expediting, and repeated rescheduling.

5. Infinite- and Finite-Capacity Planning

Under infinite-capacity planning, every required operation is loaded onto its preferred resource regardless of whether the resulting load exceeds availability. This approach is useful for identifying the size and timing of potential overloads and for preserving the requirements implied by the production plan.

Under finite-capacity planning, assignments are constrained by resource availability. A job may be delayed, moved to an alternative resource, split, or rejected from the current period. Finite planning produces a more executable schedule but requires more detailed data and decision rules.

Neither method is universally superior. Infinite loading is valuable for diagnosis and medium-term scenario analysis. Finite loading is valuable for execution-oriented planning. In practice, organisations often use infinite loading for an initial CRP profile and finite-capacity scheduling for critical work centres and near-term periods.

6. Interpreting Capacity Profiles

A capacity profile displays required and available capacity by work centre and time period. A typical interpretation distinguishes:

• overload: requirement exceeds available capacity;
• underload: substantial capacity remains unused;
• bottleneck: a resource whose limited capacity restricts system throughput;
• temporary peak: a short-lived excess that may be handled by timing or overtime;
• structural shortage: a recurring excess requiring process, workforce, or investment action.

An overload is not automatically a bottleneck. A work centre may show high utilisation but have an alternative resource, flexible routing, or low system impact. Conversely, a work centre with moderate utilisation may be a bottleneck if it controls a critical component or has no substitute.

Corrective actions should be evaluated in terms of cost, lead time, quality, risk, and system-wide consequences. Options include:

1. rescheduling orders within allowable due-date limits;
2. using overtime or additional shifts;
3. transferring work to qualified alternative resources;
4. splitting lots or overlapping operations;
5. reducing setup time through SMED or improved tooling;
6. improving yield, uptime, or standard work;
7. subcontracting selected operations;
8. changing product mix or customer priorities;
9. adding equipment, labour, or training;
10. redesigning the product or process.

The preferred response is not necessarily the one that removes the largest local overload. It should improve the flow and delivery performance of the complete system.

7. An Applied Case: Meridian Precision Components

This is an illustrative case. Meridian Precision Components produces two families of industrial valves. Its weekly master production schedule requires 80 units of Product A and 50 units of Product B. Both products use the same CNC turning centre, but their routings differ.

Background and problem

Product A requires 0.30 machine-hours per unit and Product B requires 0.45 machine-hours per unit. Each weekly lot requires one setup: 2.0 hours for A and 3.0 hours for B. The work centre has two machines, each scheduled for 40 hours. Planned maintenance and breaks reduce availability to 90%, and expected operating efficiency is 85%.

Analysis

Required capacity is:

• Product A: 2.0 + (80 × 0.30) = 26.0 hours.
• Product B: 3.0 + (50 × 0.45) = 25.5 hours.
• Total requirement: 51.5 hours.

Available capacity is:

A = 2 × 40 × 0.90 × 0.85 = 61.2 hours.

The resulting utilisation is:

U = 51.5 ÷ 61.2 ≈ 0.842 or approximately 84.2%.

The initial profile therefore appears feasible, with about 9.7 hours of expected capacity balance. However, a second work centre performs a specialised heat-treatment operation. It has only 42 available hours, while the two products require 0.35 and 0.50 hours per unit respectively, plus 1.5 hours of setup per product family:

R = (1.5 + 80 × 0.35) + (1.5 + 50 × 0.50) = 56.5 hours.

Heat treatment is overloaded by 14.5 hours. It is the likely bottleneck, even though machining has spare capacity.

Approach and solution

Meridian evaluates three responses. First, it considers overtime, but the specialised furnace requires certified supervision and overtime would cover only 8 hours. Second, it qualifies an approved external supplier for 10 hours of heat treatment. Third, it reduces setup time through a family-based sequence, saving 2 hours per week.

The combined effect reduces the internal requirement to 54.5 hours and transfers 10 hours externally, leaving 44.5 internal hours. This is still slightly above the 42-hour internal capacity, so the company also moves 3 hours of non-urgent Product B work to the following week, subject to customer approval. The revised internal requirement becomes 41.5 hours, which is feasible.

Outcome and learning points

The case demonstrates that CRP must examine every critical work centre rather than relying on total plant capacity. It also shows why setup data, alternative resources, supplier capability, customer priorities, and time fences matter. Capacity balancing is a cross-functional decision involving production, purchasing, quality, sales, maintenance, and finance.

8. Data Quality, Governance, and Professional Judgement

CRP is only as credible as its master data. Important data risks include obsolete routings, inaccurate standard times, missing setup requirements, incorrect calendars, unrecorded maintenance, unrealistic efficiency factors, and failure to record rework. Data governance should assign ownership, approval authority, revision control, and periodic audit responsibilities.

Managers should distinguish standard capacity from politically desired capacity. Inflating efficiency assumptions to make a plan appear feasible transfers risk to the shop floor. Similarly, reporting utilisation above 100% as normal may conceal chronic understaffing and encourage unsafe work practices.

Ethically, capacity decisions should consider worker safety, fatigue, quality, environmental obligations, and fair treatment of suppliers. Overtime and subcontracting are not neutral technical actions. Excessive overtime can increase accidents and defects; rushed subcontracting can create traceability and compliance risks.

9. Technology and Contemporary Developments

Modern ERP and manufacturing execution systems can integrate orders, routings, calendars, labour records, maintenance events, and actual production feedback. Advanced planning and scheduling systems may apply finite-capacity logic, constraint programming, heuristics, or mathematical optimisation. Digital twins can test alternative shifts, routings, and maintenance policies before implementation.

Artificial intelligence can support demand sensing, anomaly detection, predictive maintenance, estimated processing times, and scenario generation. However, AI does not remove the need for explainability, data validation, human approval, and constraint awareness. A model trained on historically delayed schedules may reproduce poor priorities rather than improve them. Governance should therefore specify data lineage, decision rights, performance monitoring, and override procedures.

Contemporary CRP is also expanding beyond machine hours. Organisations increasingly model energy availability, carbon limits, skilled-labour certifications, warehouse space, charging infrastructure for electric equipment, and supply-chain capacity. This broader view is appropriate when these resources constrain throughput.

10. Limitations and Best Practices

CRP models simplify reality. Standard times may not reflect learning, fatigue, variability, or product-specific complexity. Aggregate loading may ignore sequencing effects. A nominally available machine may be unusable because the required tool is unavailable. Capacity is also partly endogenous: methods improvement, maintenance quality, workforce learning, and product mix change the effective capability of a resource.

Best practices include:

• define work centres at a level that supports decisions;
• use consistent capacity units and calendars;
• include setup, maintenance, inspection, and realistic yield losses;
• separate planned, released, and actual work;
• validate standards using shop-floor data;
• protect critical resources with appropriate capacity buffers;
• use scenarios rather than a single supposedly precise forecast;
• identify system bottlenecks, not merely local overloads;
• coordinate CRP with MRP, scheduling, maintenance, purchasing, and sales;
• review exceptions by materiality and due-date risk;
• periodically audit master data and model assumptions.

Chapter Summary

Capacity Requirement Planning translates planned production into time-phased resource requirements and compares those requirements with available capacity. Its purpose is to test feasibility, reveal overloads and underloads, support corrective action, and coordinate production decisions across functions. CRP differs from production planning, which determines intended output; scheduling, which assigns sequence and timing; and production control, which manages execution and deviations.

The essential CRP concepts are work centres, resources, routings, operation times, loads, capacity units, available capacity, planning horizons, and time fences. Capacity may be analysed using infinite or finite assumptions. Basic calculations must use consistent units and realistic calendars, while setup, efficiency, yield, maintenance, and alternative resources require explicit treatment.

The evolution from manual load charts to MRP, ERP, advanced planning, and digital systems has increased analytical capability but has not replaced managerial judgement. Accurate master data, ethical capacity policies, cross-functional governance, and attention to system bottlenecks remain fundamental. Effective CRP balances feasibility, delivery, cost, quality, safety, and long-term capability.

Key Terms

Capacity: The productive capability of a resource or system during a specified period.

Capacity balance: Available capacity minus required capacity for a period.

Capacity unit: The unit used to measure resource capability or workload, such as machine-hours.

Capacity Requirement Planning (CRP): The process of comparing time-phased capacity requirements with available resource capacity.

Finite-capacity planning: Planning that prevents assignments from exceeding specified resource availability.

Infinite-capacity planning: Loading work without initially restricting assignments by resource limits.

Load: The quantity of work assigned or expected at a resource during a period.

Master production schedule: A time-phased plan for producing specific end items.

Planning horizon: The future period covered by a planning process.

Routing: The sequence of operations and resources required to produce an item.

Setup time: Time required to prepare a resource before processing a lot or product.

Time fence: A boundary limiting changes to plans within a specified period.

Utilisation: Required capacity divided by available capacity.

Work centre: A machine, group of machines, labour pool, or department treated as a capacity-planning unit.

Bottleneck: A resource whose limited capability constrains system throughput.

Standard time: The planned time allowed for completing an operation under defined conditions.

Available capacity: Capacity that can realistically be provided during a planning period.

Production control: Monitoring and corrective management of work as production occurs.

Review Questions

Short Answer Questions

1. What is the primary purpose of Capacity Requirement Planning?
2. Define a work centre and give two examples.
3. How does CRP differ from detailed scheduling?
4. What is the difference between infinite- and finite-capacity planning?
5. Why must capacity requirements and available capacity use compatible units?

Long Answer Questions

1. Explain the relationship among aggregate production planning, master production scheduling, MRP, CRP, scheduling, and production control.
2. Discuss the evolution of CRP from manual load analysis to digital and cyber-physical planning systems.
3. Explain the major elements of a CRP model, including work centres, routings, loads, calendars, and capacity units.
4. Analyse the managerial alternatives available when a work centre is overloaded.
5. Critically evaluate the limitations of CRP and propose practices for improving its reliability.

Analytical / Application Questions

1. A work centre has three machines, each scheduled for 36 hours per week. Planned availability is 92%, efficiency is 80%, and yield is 97%. Calculate available capacity and utilisation if the planned load is 72 hours. State your assumptions.
2. A routing requires a 1.5-hour setup and 4 minutes per unit for a lot of 180 units. Calculate total capacity requirement and explain how the result would change if two setups were required.
3. A plant has adequate total capacity but one specialised inspection centre is overloaded every week. Analyse why total plant capacity is insufficient evidence of feasibility and recommend a response sequence.

Multiple Choice Questions with Answers

Question 1

Which statement best describes the primary purpose of CRP?

A. To determine annual sales revenue

B. To compare time-phased resource requirements with available capacity

C. To select suppliers solely on price

D. To define the sequence of customer deliveries

Correct Answer: B. To compare time-phased resource requirements with available capacity

Question 2

A work centre has required capacity of 48 hours and available capacity of 60 hours. What is its utilisation?

A. 0.80 or 80%

B. 1.25 or 125%

C. 12%

D. 108 hours

Correct Answer: A. 0.80 or 80%

Question 3

Which activity primarily determines the intended quantity and timing of end-item production?

A. Production control

B. Detailed dispatching

C. Master production scheduling

D. Preventive maintenance

Correct Answer: C. Master production scheduling

Question 4

A routing requires 2 hours of setup and 0.25 hours per unit for 100 units. What is the total operation time?

A. 25 hours

B. 27 hours

C. 50 hours

D. 102 hours

Correct Answer: B. 27 hours

Question 5

Which of the following is a capacity unit?

A. Machine-hours

B. Customer satisfaction

C. Selling price

D. Product colour

Correct Answer: A. Machine-hours

Question 6

What is a key characteristic of finite-capacity planning?

A. It ignores resource limitations

B. It prevents assignments from exceeding specified availability

C. It eliminates the need for routings

D. It measures only material quantities

Correct Answer: B. It prevents assignments from exceeding specified availability

Question 7

Why can a work centre with moderate utilisation still be a bottleneck?

A. It may control a critical operation with no qualified alternative

B. Moderate utilisation always means zero output

C. Utilisation is unrelated to capacity

D. Bottlenecks are defined only by labour cost

Correct Answer: A. It may control a critical operation with no qualified alternative

Question 8

Which data problem most directly understates CRP requirements in a high-mix environment?

A. Including setup time accurately

B. Using approved alternative routings

C. Omitting sequence or product-family setup time

D. Recording planned maintenance

Correct Answer: C. Omitting sequence or product-family setup time

Question 9

A work centre has 2 machines, 40 scheduled hours each, 90% availability, and 85% efficiency. What is its available capacity before yield losses?

A. 61.2 hours

B. 68.0 hours

C. 72.0 hours

D. 80.0 hours

Correct Answer: A. 61.2 hours

Question 10

Which response is most appropriate when a recurring overload indicates a structural capacity shortage?

A. Ignore the overload because total plant capacity is adequate

B. Consider capability improvement, additional resources, or qualified subcontracting

C. Delete the routing operation

D. Increase the reported efficiency without operational change

Correct Answer: B. Consider capability improvement, additional resources, or qualified subcontracting