Capacity Planning and Schedule Feasibility

Introduction

Master production scheduling (MPS) translates aggregate production intentions into time-phased end-item schedules. An MPS is useful only when the enterprise can execute it. A schedule that satisfies forecast demand but requires more labour, machine time, tooling, storage, or supplier capability than is available is not a feasible plan; it is an aspiration. Capacity planning therefore provides the essential link between demand decisions and operational reality.

At the MPS level, the principal capacity question is not whether every individual work centre can be scheduled in minute-by-minute detail. Rather, it is whether the proposed schedule places an acceptable load on critical resources over the planning horizon. This is the purpose of rough-cut capacity planning (RCCP). RCCP identifies probable overloads early enough for planners to revise the MPS, alter shifts, subcontract, change product mix, or negotiate delivery dates.

This chapter develops three major RCCP approaches: capacity bills, resource profiles, and overall factors. It also examines bottlenecks, finite and infinite capacity assumptions, and the reconciliation of material and capacity constraints.

Learning Objectives

By the end of this chapter, the reader should be able to:

1. Explain the role of capacity planning within the MPS and planning hierarchy.
2. Construct and interpret a capacity bill and a resource profile.
3. Apply the overall factor method and evaluate its limitations.
4. Calculate capacity requirements, utilization, and overloads with consistent units.
5. Distinguish infinite-capacity and finite-capacity scheduling assumptions.
6. Identify bottlenecks and assess alternative capacity responses.
7. Reconcile material availability, capacity availability, and customer due dates.
8. Evaluate the managerial, technological, and governance implications of schedule-feasibility decisions.

1. Capacity Planning in the Planning Hierarchy

Capacity is the maximum sustainable output or workload that a resource can provide during a specified period under stated operating conditions. It is not necessarily the theoretical maximum. Planned maintenance, breaks, changeovers, absenteeism, quality losses, and operating policies reduce the capacity that is practically available.

Capacity decisions occur at several levels:

RCCP is deliberately less detailed than capacity requirements planning (CRP). RCCP uses the MPS and approximate capacity relationships to test feasibility. CRP uses planned order releases, bills of material, routings, operation times, and work-centre calendars to calculate detailed loads. RCCP is consequently faster and more appropriate for testing alternative MPS scenarios, but it cannot replace detailed planning.

A useful distinction is between required capacity and available capacity. Required capacity is the workload generated by the planned production mix. Available capacity is the effective capacity offered by a resource in a period. Feasibility requires that requirements do not exceed available capacity, subject to the organization’s tolerance for queues, overtime, and risk.

For a resource r in period t, the basic load relationship is:

Here, Lᵣ,ₜ is required load on resource r in period t, Qᵢ,ₜ is the quantity of end item i scheduled in period t, and aᵢ,ᵣ is the resource requirement per unit of item i. If aᵢ,ᵣ is measured in machine-hours per unit, Lᵣ,ₜ is measured in machine-hours. The relationship assumes that requirements are additive and that the resource requirement per unit is sufficiently stable for planning purposes.

Capacity utilization is:

where Cᵣ,ₜ is effective available capacity in resource-hours. Utilization above 100% indicates an overload under the stated assumptions. Utilization below 100% does not automatically prove feasibility, because setup interactions, sequence dependence, variability, and material shortages may still prevent execution.

2. Measuring Effective Capacity

A calendar may show 2,400 available machine-hours, but the usable figure may be much lower. A common planning expression is:

where:

• N = number of parallel machines or employees;
• H = scheduled hours per resource per period;
• D = planned operating days or periods, when H is a daily measure;
• A = availability factor, representing time actually available after breakdowns, maintenance, and planned interruptions;
• E = efficiency factor, representing performance relative to the standard rate.

If quality yield is relevant, good output may be represented by:

where Y is the proportion of output meeting quality requirements. These factors must be used carefully. If standard processing times already include normal efficiency losses, applying E again double-counts the loss.

Illustration. A work centre has four machines, each scheduled for 40 hours per week. Availability is 90%, efficiency is 85%, and quality yield is 98%.

Good-capacity equivalent is:

Whether quality yield should be included in capacity depends on the convention used in the planning system. The essential requirement is consistency and transparent definitions.

3. Rough-Cut Capacity Planning Methods

3.1 Capacity Bills

A capacity bill is analogous to a bill of material, but it expresses the capacity required to produce one unit of an end item. It normally lists the critical resources, hours per unit, and sometimes the period offset in which the capacity is consumed.

If the MPS schedules 300 units of P in Week 3, the capacity bill produces 120 fabrication hours, 75 assembly hours, and 30 test hours. The calculation is direct:

For several products, requirements are summed by resource and period. Capacity bills are particularly useful when the product mix is stable and when management knows which resources are strategically critical.

A capacity bill can include setup requirements. If each batch requires s setup-hours and the period contains b batches, total load may be estimated as:

The setup term is valid only when batch counts and setup times are known or reasonably estimated. Ignoring setups is a frequent source of optimistic RCCP results.

3.2 Resource Profiles

A resource profile adds time-phased detail to a capacity bill. It shows not only how much capacity an item requires, but also when that capacity is consumed relative to the MPS due date. A product completed in Week 6 may require fabrication in Week 4, assembly in Week 5, and testing in Week 6. The profile therefore shifts loads backward according to lead times or operation offsets.

Suppose product A requires:

An MPS quantity of 400 units due in Week 6 creates 200 hours on F in Week 4, 120 hours on A in Week 5, and 40 hours on T in Week 6. This approach is more realistic than loading all capacity in Week 6.

Resource profiles are valuable when lead times, queues, or bottleneck locations matter. Their accuracy depends on credible operation offsets. If the offsets are based on ideal lead times while actual queues are long and variable, the profile may shift the problem rather than reveal it.

3.3 Overall Factors

The overall factor (OF) method estimates capacity requirements by applying an aggregate capacity factor to the MPS quantity. For a resource r:

where OFᵣ is the average resource-hours required per unit of the representative product or product family, and MPSₜ is the aggregate scheduled quantity in period t.

For a product mix, a weighted overall factor may be calculated as:

where wᵢ is the assumed proportion of product i in the mix and ∑wᵢ = 1. The estimate then becomes:

The method is simple and quick, making it useful for high-level what-if analysis. However, it conceals product-mix changes. If a high-load product becomes a larger proportion of the schedule, the historical overall factor understates requirements. OF analysis should therefore be supplemented by product-family or resource-profile analysis when mix variability is material.

Comparison of RCCP Methods

4. Worked RCCP Example

Consider an illustrative manufacturer with a four-week MPS for two products. The MPS quantities are:

The capacity requirements per unit are:

Available effective capacity is 100 cutting hours, 70 assembly hours, and 30 test hours per week.

For Week 1:

• Cutting = (100 × 0.40) + (80 × 0.60) = 40 + 48 = 88 hours.
• Assembly = (100 × 0.30) + (80 × 0.20) = 30 + 16 = 46 hours.
• Test = (100 × 0.10) + (80 × 0.15) = 10 + 12 = 22 hours.

Thus, Week 1 is feasible on these aggregate measures. For Week 3:

• Cutting = (80 × 0.40) + (120 × 0.60) = 32 + 72 = 104 hours.
• Assembly = (80 × 0.30) + (120 × 0.20) = 24 + 24 = 48 hours.
• Test = (80 × 0.10) + (120 × 0.15) = 8 + 18 = 26 hours.

Cutting is overloaded by 4 hours, or 4% of available capacity. Its utilization is:

Management might authorize four hours of overtime, move some production to another machine, subcontract part of the work, alter the product sequence, or shift quantities between weeks if customer commitments and material availability permit. The calculation identifies a problem; it does not by itself select the best remedy.

5. Bottlenecks and Capacity Economics

A bottleneck is a resource whose capacity constrains the throughput of the system. The highest-utilization resource is a candidate bottleneck, but utilization alone is not sufficient. A resource at 95% utilization with highly variable arrivals may cause more delivery disruption than a stable resource at 102% temporarily supported by planned overtime.

Bottleneck analysis should consider:

• required load relative to effective capacity;
• variability and queue formation;
• setup and changeover dependence;
• availability of substitutes or alternative routings;
• material synchronization;
• downstream consequences of lost time;
• economic contribution per constrained hour.

A useful prioritization measure is contribution per bottleneck hour:

This measure supports product-mix decisions when demand exceeds bottleneck capacity. It should not be used mechanically: contractual obligations, strategic customers, quality, safety, and future market considerations may override a purely short-term economic ranking.

The theory of constraints emphasizes five broad managerial actions: identify the constraint, exploit it, subordinate other activities to it, elevate its capacity, and repeat the process when the constraint moves. In MPS practice, this means protecting the bottleneck from starvation, avoiding excessive work-in-process before it, and ensuring that upstream schedules do not create loads the bottleneck cannot process.

6. Infinite and Finite Capacity Assumptions

Infinite-Capacity Planning

Under an infinite-capacity assumption, planned orders are generated or loaded without initially restricting them to available resource capacity. The system then reports overloads. This approach is computationally convenient and can reveal the total theoretical requirement. It is common in MRP and early-stage RCCP because it preserves demand and material signals for analysis.

Its danger is interpretive: an MPS may appear valid simply because the software accepted it. An overload report must be treated as a feasibility warning, not as a schedule commitment.

Finite-Capacity Planning

Finite-capacity planning explicitly prevents or reschedules work when a resource’s capacity is exhausted. It considers calendars, operation precedence, queue positions, and sometimes setup sequences. Finite scheduling can create a more executable plan, but it may delay orders, generate material conflicts, or require complex optimization.

Neither assumption is universally superior. Infinite-capacity analysis is useful for exposing strategic capacity gaps. Finite-capacity analysis is necessary when commitments must reflect actual resource limits. A mature planning process uses both: first identify demand-implied requirements, then construct a feasible schedule.

7. Reconciling Material and Capacity Constraints

A feasible production plan must satisfy at least three linked conditions:

1. the required materials and components are available at the appropriate time;
2. the necessary capacity is available at the appropriate time;
3. the resulting completion dates satisfy customer and policy commitments.

Material and capacity planning are interdependent. A component shortage may reduce the load on assembly but does not make the MPS feasible; it merely prevents production for a different reason. Conversely, materials may be available while a critical machine is overloaded.

The reconciliation process can be organized as follows:

1. Load the MPS. Convert scheduled quantities into material and resource requirements.
2. Identify exceptions. Locate material shortages, capacity overloads, and timing conflicts.
3. Classify the constraint. Determine whether the issue is quantity, timing, supplier reliability, labour, equipment, or policy related.
4. Generate alternatives. Consider rescheduling, lot splitting, substitution, overtime, subcontracting, alternate routing, inventory use, or demand negotiation.
5. Recalculate both dimensions. A change that solves a capacity problem may create a material shortage or another bottleneck.
6. Approve a controlled revision. Record assumptions, ownership, effective dates, and customer impact.

Available-to-promise (ATP) calculations must also reflect capacity where capacity is a binding constraint. Material-based ATP that ignores a bottleneck can promise dates the factory cannot meet. In such cases, capable-to-promise logic is more appropriate because it checks material, labour, equipment, and lead-time feasibility together.

8. Detailed Illustrative Case: Meridian Pumps

Background

Meridian Pumps manufactures two industrial pump families: Standard (S) and High-Pressure (H). The MPS team plans four weeks of production. Both products use the same final test cell, but H requires more test time and more machining time.

Problem

The proposed MPS is:

Requirements per unit are:

Effective weekly capacities are 180 machining hours, 100 assembly hours, and 80 test hours.

Analysis

Week 3 loads are:

• Machining = (140 × 0.50) + (100 × 0.80) = 150 hours.
• Assembly = (140 × 0.35) + (100 × 0.45) = 94 hours.
• Test = (140 × 0.20) + (100 × 0.50) = 78 hours.

Week 3 is feasible, although assembly utilization is 94% and test utilization is 97.5%. Week 4 loads are:

• Machining = (200 × 0.50) + (80 × 0.80) = 164 hours.
• Assembly = (200 × 0.35) + (80 × 0.45) = 106 hours.
• Test = (200 × 0.20) + (80 × 0.50) = 80 hours.

Assembly is overloaded by 6 hours, while testing is exactly at 100%. The apparent bottleneck is assembly because it exceeds available capacity, but test has no protection against disruption.

A material review also finds that a seal used in H is available for only 70 units in Week 4. The proposed 80 H units are therefore infeasible on material grounds as well as vulnerable on capacity grounds.

Approach and Solution

The team considers three alternatives:

• authorize six assembly overtime hours and expedite ten H seals;
• reduce Week 4 H production to 70 units and replace the ten units with S, if customer priorities permit;
• shift ten H units from Week 4 to Week 3, but Week 3 test capacity would then be exceeded.

The preferred solution is to produce 70 H units in Week 4, use six assembly overtime hours for the resulting mix, and negotiate or reschedule the remaining ten H units. If ten S units replace ten H units, assembly load falls by 1 hour because each replacement reduces assembly demand by 0.10 hour, but the original 106-hour load becomes 105 hours. Additional action is still required. A more effective option is to move 12 S units to Week 2, provided Week 2 capacity and materials are checked, and use the resulting Week 4 capacity for committed H orders.

Outcome

The case demonstrates that no single calculation resolves the planning problem. The team must reconcile the MPS with resource loads, material availability, customer priorities, and the consequences of moving work between periods. The final approved plan should include the revised quantities, overtime authorization, seal availability, and a named owner for the remaining customer commitments.

Learning Points

1. A resource can be feasible in aggregate but unsafe at high utilization.
2. A material shortage and a capacity overload can occur simultaneously.
3. Product substitution changes both material and capacity requirements.
4. Rescheduling must be iterated through the complete planning model.
5. The best solution is not necessarily the one with the lowest immediate overtime cost.

9. Implementation, Technology, and Governance

Modern advanced planning and scheduling systems can combine MPS, bills of material, routings, calendars, finite scheduling, and supplier data. Digital twins and simulation allow planners to test overtime, alternate routings, batch-size changes, and demand scenarios before committing the plan. Machine-learning models may improve estimates of processing time, breakdown probability, or supplier reliability, but predictions should not replace operational accountability.

Governance is especially important when automated systems alter priorities. Planners should maintain version control, document assumptions, distinguish standard from actual times, and audit overrides. Capacity decisions can affect employee workload, safety, subcontractor conditions, and customer fairness. Overtime must comply with labour law and collective agreements; subcontracting must respect quality, confidentiality, environmental, and supplier-governance requirements.

Recommended controls include:

• approved and periodically reviewed work-centre calendars;
• clear definitions of availability, efficiency, yield, and utilization;
• exception thresholds for overloads and material shortages;
• scenario comparison rather than reliance on one forecast;
• escalation rules for bottlenecks and customer-impacting changes;
• post-period review of planned versus actual capacity consumption.

10. Limitations and Best Practices

RCCP is an approximation. It may omit detailed sequencing, transportation, tool availability, operator skills, rework, stochastic breakdowns, and interaction among resources. Capacity factors can also become obsolete when products, processes, automation, or staffing change.

Best practice is to:

1. use effective rather than theoretical capacity;
2. include setups and major loss factors when material;
3. time-phase requirements using credible resource profiles;
4. segment products when overall factors hide important mix differences;
5. protect bottleneck resources and review high-utilization risks;
6. reconcile capacity with material and supplier constraints;
7. use finite scheduling for execution-critical decisions;
8. maintain a formal feedback loop between planned standards and actual performance.

Chapter Summary

Capacity planning tests whether an MPS can be executed with available resources. Rough-cut capacity planning provides an efficient bridge between aggregate planning and detailed CRP. Capacity bills express resource hours per unit; resource profiles place those requirements in the periods in which work occurs; and overall factors provide rapid aggregate estimates based on representative product mix. Required capacity is calculated by multiplying scheduled quantities by resource requirements and summing by resource and period. Effective capacity must reflect operating calendars, availability, efficiency, and, where appropriate, yield.

Infinite-capacity methods expose the workload implied by demand but permit overloads to appear. Finite-capacity methods restrict the schedule to actual resource limits but may delay orders and conceal the full scale of demand pressure. Bottleneck analysis requires more than identifying the highest utilization: variability, setups, material synchronization, economics, and strategic priorities also matter. Finally, material feasibility and capacity feasibility must be evaluated together. A credible MPS is therefore a controlled compromise among demand, resources, materials, lead times, and customer commitments.

Key Terms

Capacity: The sustainable resource capability available under specified operating conditions.

Effective capacity: Usable capacity after planned losses such as maintenance, availability, and efficiency effects.

Rough-cut capacity planning: Approximate capacity analysis performed at or near the MPS level.

Capacity bill: A time-independent listing of resource requirements per unit of an end item.

Resource profile: A time-phased representation of capacity requirements relative to completion dates.

Overall factor: An average resource requirement applied to aggregate scheduled quantity.

Capacity requirements planning: Detailed calculation of work-centre loads using orders, routings, and operation times.

Infinite capacity: A planning assumption under which work is loaded without initially restricting it to resource limits.

Finite capacity: A planning assumption that prevents or reschedules work when capacity is exhausted.

Bottleneck: A resource that constrains system throughput.

Utilization: Required load expressed as a proportion or percentage of available capacity.

Overload: Required capacity greater than available capacity in a specified period.

Setup time: Time required to prepare a resource for a job, product, or batch.

Material feasibility: Availability of required materials in the required quantities and periods.

Capable-to-promise: A commitment process that considers material and capacity feasibility.

Aggregate planning: Medium- to long-range planning of output, workforce, inventory, and capacity.

Finite scheduling: Detailed sequencing of operations subject to resource, precedence, and calendar constraints.

Throughput: The rate at which the system produces saleable output.

Review Questions

Short Answer Questions

1. What is the purpose of rough-cut capacity planning in the MPS process?
2. Distinguish effective capacity from theoretical capacity.
3. What information does a resource profile add to a capacity bill?
4. Why can a high-utilization resource be risky even when it is not formally overloaded?
5. What is the principal difference between infinite-capacity and finite-capacity planning?

Long Answer Questions

1. Explain the construction and managerial use of capacity bills, resource profiles, and overall factors.
2. Discuss how bottleneck identification should incorporate variability, setups, material synchronization, and economics.
3. Compare the advantages and limitations of infinite-capacity and finite-capacity assumptions.
4. Explain how material and capacity constraints should be reconciled when revising an MPS.
5. Evaluate the governance controls required when advanced planning software or AI supports capacity decisions.

Analytical / Application Questions

1. A work centre has 3 machines, each scheduled for 40 hours per week. Availability is 92% and efficiency is 80%. Calculate effective weekly capacity and explain the assumptions in your calculation.
2. Product A requires 0.4 machine-hours and Product B requires 0.7 machine-hours. If a week’s MPS contains 150 A units and 100 B units, and available capacity is 120 hours, determine the load, utilization, and overload or spare capacity.
3. A proposed MPS has adequate materials but overloads final test by 8 hours and assembly by 4 hours. Develop and justify two alternative responses, explaining how each could affect materials, customer dates, and the location of the bottleneck.

Multiple Choice Questions with Answers

Question 1

What is the primary purpose of rough-cut capacity planning?

A. To sequence every individual job on the shop floor

B. To test whether an MPS places acceptable loads on critical resources

C. To calculate annual depreciation for production equipment

D. To replace the bill of material

Correct Answer: B. To test whether an MPS places acceptable loads on critical resources

Question 2

A resource requires 0.5 machine-hours per unit. An MPS schedules 240 units. What is the required load?

A. 120 machine-hours

B. 240 machine-hours

C. 480 machine-hours

D. 0.5 machine-hours

Correct Answer: A. 120 machine-hours

Question 3

Which RCCP method explicitly places capacity requirements in periods relative to the completion date?

A. Overall factor method

B. Resource profile method

C. Annual budget method

D. Economic order quantity method

Correct Answer: B. Resource profile method

Question 4

A work centre has 100 available hours and a required load of 115 hours. Its utilization is:

A. 15%

B. 85%

C. 100%

D. 115%

Correct Answer: D. 115%

Question 5

What is a major limitation of the overall factor method?

A. It cannot use aggregate quantities

B. It always requires detailed sequencing data

C. It may conceal changes in product mix

D. It excludes all capacity information

Correct Answer: C. It may conceal changes in product mix

Question 6

Under an infinite-capacity assumption, an overload is generally:

A. Automatically eliminated by the system

B. Reported after the workload is loaded

C. Ignored permanently

D. Converted into a material receipt

Correct Answer: B. Reported after the workload is loaded

Question 7

Which condition most directly indicates a capacity overload?

A. Required capacity is less than available capacity

B. Required capacity equals zero

C. Required capacity exceeds available capacity

D. Material inventory is above safety stock

Correct Answer: C. Required capacity exceeds available capacity

Question 8

If four machines each provide 40 scheduled hours, with 90% availability and 80% efficiency, what is effective capacity?

A. 115.2 hours

B. 128.0 hours

C. 144.0 hours

D. 160.0 hours

Correct Answer: A. 115.2 hours

Question 9

Why must material and capacity feasibility be checked together?

A. Material availability always determines machine speed

B. Solving one constraint can create or expose another constraint

C. Capacity planning eliminates the need for inventory records

D. Materials and capacity are unrelated planning dimensions

Correct Answer: B. Solving one constraint can create or expose another constraint

Question 10

Which action is consistent with bottleneck management?

A. Allow the bottleneck to remain starved while upstream work increases

B. Protect and exploit the constrained resource before adding capacity elsewhere

C. Maximize utilization of every resource regardless of demand

D. Ignore setup time at the constrained resource

Correct Answer: B. Protect and exploit the constrained resource before adding capacity elsewhere