Demand, Capacity and Space Planning

Introduction

Facilities and plant layout planning begins with a deceptively simple question: what must the facility be capable of producing, and what resources must it contain to do so reliably? The answer cannot be obtained from current demand alone. Managers must translate market forecasts into production requirements, equipment quantities, labour schedules, utility loads, storage areas, circulation space and future expansion provisions. These decisions are interdependent: excessive capacity raises fixed cost and obsolescence risk, while inadequate capacity creates queues, overtime, missed deliveries and loss of market share.

Demand, capacity and space planning therefore forms the analytical bridge between operations strategy and physical facility design. It connects the sales and operations plan to process technology, workforce planning, material flow, building services and capital investment. At postgraduate level, the central concern is not merely calculating capacity, but evaluating uncertainty, capacity cushions, bottlenecks, scalability, resilience and the economic consequences of alternative layouts.

Learning Objectives

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

1. Distinguish demand, design capacity, effective capacity and actual output.
2. Convert forecasts and production plans into resource and space requirements.
3. Calculate utilisation, efficiency, capacity cushions and equipment requirements.
4. Identify bottlenecks and evaluate their effect on system throughput.
5. Incorporate variability, maintenance, yield, changeovers and product mix into capacity decisions.
6. Compare expansion alternatives using operational, financial and strategic criteria.
7. Develop a defensible capacity-and-space plan for a real or illustrative facility.

1. Demand as the Starting Point

Demand is the quantity customers are willing and able to purchase during a specified period. A facility plan normally uses several demand measures:

• Independent demand: demand originating outside the production system, such as customer orders or market forecasts.
• Dependent demand: demand derived from a parent product, such as components required for an assembly schedule.
• Average demand: expected volume over a planning horizon.
• Peak demand: a high-volume period that may determine required capacity.
• Takt demand: the rate at which output must be completed to match customer demand.

A forecast is not a fact; it is an estimate with error. A robust plan therefore uses scenarios or probability distributions rather than a single deterministic number. Important demand descriptors include volume, product mix, timing, seasonality, service-level requirements and geographical distribution.

If available production time is measured in minutes per period and demand is measured in units per period, the required production rate is:

where r is the required rate in units per unit of time, D is demand in units, and T is available calendar or planning time. The units must be consistent. For example, 12,000 units per month over 20 operating days of 8 hours gives T = 9,600 minutes per month, so the required rate is 1.25 units per minute, or one unit every 0.8 minutes.

In a repetitive flow environment, takt time is commonly expressed as:

where A is net available production time and D is customer demand during the same period. Net time excludes planned breaks, meetings and other non-production intervals. Takt time is a demand signal, not necessarily the processing time of every operation.

2. Capacity Concepts and Measures

Capacity is the maximum output a system can produce under stated conditions. It is meaningful only when the time period, product definition, operating conditions and quality standard are specified.

2.1 Design and effective capacity

Design capacity is the theoretical maximum output under ideal conditions. Effective capacity is the realistic maximum after allowing for product mix, preventive maintenance, breaks, setups, quality procedures and other planned losses. Actual output is what the facility really produces.

Two widely used measures are:

Utilisation is appropriate when comparing actual output with theoretical potential. Efficiency is appropriate when assessing performance against realistic operating capability. Neither measure by itself proves that a facility is well managed. High utilisation can increase queues and reduce resilience; moderate utilisation may be rational where demand is volatile or service speed is strategically important.

The capacity cushion is the capacity deliberately held above expected demand:

The selected capacity measure must be stated. A cushion based on effective capacity is usually more informative for operational planning than one based on design capacity. A facility facing uncertain demand, long replenishment lead times or severe service penalties generally requires a larger cushion than a stable, make-to-stock operation.

2.2 Capacity in mixed-product systems

A facility producing several products cannot use units alone unless products consume equal resources. A better approach is to calculate workload in standard hours:

where Dᵢ is demand for product i in units and sᵢ is the standard processing time per unit in hours. If quality yield is less than 100%, required gross production is:

If a product requires 0.4 hours per unit and 2,000 good units are required at a 95% yield, gross production is 2,000 ÷ 0.95 = 2,105.26 units, normally rounded upward to 2,106 units. The required processing time is approximately 842.4 hours.

Setup time must be added when it is material:

Here nᵢ is the number of setups for product or campaign i, and qᵢ is setup time per changeover. This formulation assumes additive processing and setup times and does not itself account for overlapping operations.

3. Translating Demand into Capacity Requirements

3.1 Equipment quantity

For a work centre, the basic number of machines required is:

where W is required workload per period and Cₑ is effective capacity per machine per period. Since a fraction of a machine cannot normally be installed, the result is rounded upward:

Effective capacity may be estimated as:

where H is scheduled hours, s is nominal processing rate in units per hour, a is availability, p is performance efficiency, and y is first-pass yield. These factors are dimensionless proportions between zero and one. Their use assumes that losses can reasonably be represented multiplicatively and that the measurement period is common.

Worked example. A machine is scheduled for 2,000 hours per year. Its nominal rate is 10 units per hour, availability is 0.90, performance is 0.95 and yield is 0.98. Thus:

Cₑ = 2,000 × 10 × 0.90 × 0.95 × 0.98 = 16,758 units/year

If annual demand is 31,000 units, two machines provide 33,516 units of effective capacity. The resulting utilisation against effective capacity is approximately 92.5%. This may be financially attractive but operationally fragile if demand or downtime increases.

3.2 Labour requirements

Labour planning must distinguish direct labour, indirect support and skills. A simple direct-labour estimate is:

The number of employees depends on paid hours, attendance, training, breaks, overtime policy and shift pattern. A skill matrix is essential where one employee cannot substitute for another. Labour capacity is therefore not merely a headcount; it is a time-and-competence resource.

3.3 Utilities and support services

Capacity plans should estimate electricity, compressed air, water, steam, chilled water, ventilation, information systems and waste-handling requirements. For a utility with a load profile, peak demand is often more important than average consumption because distribution equipment is sized for peak or coincident load. Diversity factors may reduce the combined load when equipment does not operate simultaneously, but such reductions require evidence rather than optimism.

4. Bottlenecks and Flow Capacity

A bottleneck is a resource whose available capacity constrains the throughput of the entire system. In a serial process with no inventory losses, system capacity is approximately the minimum capacity of its stages:

where Cᵢ is the capacity of stage i. In real systems, yield losses, blocking, starving, rework and product routing complicate this relationship.

A bottleneck should be analysed using workload, variability and queue behaviour, not only nominal rates. Improving a non-bottleneck may produce no increase in total throughput. Typical responses include reducing setup time, adding parallel equipment, improving reliability, reallocating work, redesigning the product, subcontracting, or changing the sequence of operations.

Bottleneck management should follow a disciplined logic: identify the constraint, exploit it using existing resources, subordinate other activities to it, elevate its capacity through investment, and then reassess because the constraint may move. This logic prevents local efficiency from being mistaken for system improvement.

5. Space Planning: From Workload to Floor Area

Space planning translates capacity decisions into a physical programme. Required space normally includes:

1. Processing and equipment areas.
2. Material receiving, inspection and storage.
3. Work-in-process and finished-goods storage.
4. Maintenance, tooling and utilities.
5. Offices, welfare, training and control rooms.
6. Aisles, pedestrian routes, emergency access and safety clearances.
7. Future expansion and flexibility zones.

A useful conceptual model is:

The calculation is not a universal engineering formula because circulation and support requirements depend on codes, equipment, fire protection, vehicle type, process risk and layout philosophy. Net equipment area should include the machine footprint plus operating, maintenance, loading and safety clearances. Storage area should be based on inventory policy and handling method rather than current average inventory alone.

For pallet storage, a preliminary estimate may be made as:

The occupancy factor is less than one because aisles, access restrictions and uneven slot utilisation prevent every nominal location from being continuously usable. The estimate must then be converted into rack, floor and aisle area using the selected storage system. Peak inventory, not average inventory, should be used when service continuity is critical.

5.1 Relationship diagrams and flow analysis

Space planning should begin with process data: from-to flows, relationship ratings, adjacency needs, safety separation, material-handling equipment and sequence constraints. A high-volume flow may justify close adjacency, whereas hazardous materials may require separation despite high flow. The best layout is therefore not necessarily the one with the shortest distance; it is the one that balances flow, safety, flexibility, visibility, maintainability and total cost.

Figure: The iterative translation of demand into capacity and space requirements.

6. Variability, Risk and Scalability

Capacity decisions are made under uncertainty. Sources include forecast error, absenteeism, breakdowns, supplier disruption, quality variation, demand surges and product introductions. A plan based on average demand may fail when variability creates queues or when several resources experience adverse events simultaneously.

Useful approaches include scenario analysis, sensitivity analysis, simulation, real-options reasoning and staged investment. For example, a base case may represent expected demand, a high case may represent the 90th percentile, and a low case may represent a recession scenario. The decision should show which assumptions change the preferred alternative.

Scalability is the ability to increase or decrease capacity without disproportionate cost or disruption. Modular equipment, standardised utilities, expandable buildings, cross-trained labour, postponement and flexible automation can improve scalability. Dedicated high-speed equipment may have lower unit cost at high volume but poor adaptability. Flexible equipment may cost more per unit but preserve strategic options.

Expansion alternatives commonly include:

A sound decision compares total cost of ownership, not only purchase price. Relevant factors include capital, labour, energy, maintenance, material handling, quality, inventory, disruption during construction, depreciation, residual value and risk exposure.

7. Integrated Planning Procedure

A professional capacity-and-space study can follow these steps:

1. Define the planning horizon, service promise, product families and units of analysis.
2. Validate demand scenarios, including mix, seasonality and forecast uncertainty.
3. Convert demand into a master production or operations schedule.
4. Calculate net available time by shift, resource and period.
5. Determine workload, including processing, setups, yield losses and rework.
6. Identify bottlenecks and calculate required equipment and labour by work centre.
7. Estimate utilities, storage, material-handling and support requirements.
8. Develop alternative block layouts and test flow, safety and adjacency constraints.
9. Evaluate capacity cushions, expansion options, cost and risk.
10. Stress-test the preferred plan against demand, downtime and mix scenarios.
11. Specify governance: assumptions, owners, review dates and triggers for expansion.

Capacity planning should be linked to sales and operations planning, master scheduling, material requirements planning and finite-capacity scheduling. A plan that ignores procurement lead times or supplier capacity may create a theoretically adequate factory that cannot operate in practice.

8. Illustrative Case: Orion Medical Devices

Background

Orion Medical Devices is an illustrative manufacturer of sterile diagnostic cartridges. Demand is forecast at 48,000 good cartridges per year. The product passes through moulding, automated assembly and final inspection. The plant operates 2,000 scheduled hours per year. The company is considering whether to install one additional assembly cell or lease space for a second production area.

Problem

The forecast is uncertain, yield is not perfect, and the existing building has limited room for expansion. Management must determine whether current resources are adequate and which investment is scalable.

Analysis

Annual good demand is 48,000 units. At a first-pass yield of 96%, required gross production is:

48,000/0.96 = 50,000 units

The assembly operation requires 2.0 minutes per gross unit, including normal minor stoppages. Its annual workload is:

50,000 × 2.0/60 = 1,666.7 hours

One assembly cell has 2,000 scheduled hours, availability of 0.88 and planned performance of 0.95. Its effective time is 1,672 hours, before any further product-mix adjustment. It is therefore almost fully loaded. If demand rises by 10%, gross requirement becomes 55,000 units and workload becomes 1,833.3 hours, exceeding effective capacity.

Moulding has two machines with combined effective capacity of 70,000 units. Inspection has effective capacity of 60,000 units. Assembly is the system bottleneck under the base case. Adding inspection capacity would not increase throughput.

Approach and solution

Orion compares three alternatives: overtime, a modular second assembly cell, and leased external space. Overtime can cover the base case temporarily but provides little resilience and increases fatigue risk. A modular cell requires capital but can be installed within the existing building if obsolete storage racks are removed and vertical storage is introduced. Leasing external space provides expansion capacity but increases material-handling distance and validation complexity.

The recommended first step is the modular cell, installed in phases. The company should also reduce setup and minor stoppage losses, establish a cross-training programme and reserve a defined expansion zone. A trigger for the next investment is sustained forecast demand above 52,000 gross units or effective assembly utilisation above 85% for three consecutive planning cycles.

Outcome and learning points

The case demonstrates that capacity must be calculated using good demand, yield, workload and effective time together. It also shows why bottleneck identification should precede space investment. A flexible expansion zone can have option value even when it is not immediately productive. Finally, a high utilisation figure may signal vulnerability rather than excellence when the forecast and downtime are uncertain.

9. Governance, Safety and Contemporary Technology

Space and capacity decisions must comply with building regulations, fire codes, occupational health requirements, accessibility rules, environmental permits and industry-specific validation. In regulated sectors, changes to layout may require documented risk assessment, cleaning validation, traceability controls or requalification. Emergency exits, hazardous-material segregation, ergonomic access and maintenance clearances must never be traded away for nominal capacity.

Digital tools can support this work. Enterprise systems provide demand and production data; manufacturing execution systems provide actual performance; computer-aided facility-layout tools visualise alternatives; digital twins and discrete-event simulation test queues, routing and resource interactions. AI may assist with demand sensing, anomaly detection and scenario generation, but its outputs require governance. Planners should document data provenance, monitor model drift, protect confidential information and retain human accountability for capital and safety decisions.

Contemporary facilities increasingly emphasise decarbonisation, resilience and worker-centred design. Energy capacity, renewable generation, electrification, heat recovery, water reuse and climate-risk exposure should be considered alongside conventional production metrics. Automation should be assessed for total system effects: it may reduce direct labour while increasing maintenance, software dependency, cybersecurity exposure and skill requirements.

Challenges and Limitations

Capacity calculations are approximations. Standard times may be outdated, demand scenarios may be correlated, and availability and quality factors may not be independent. The formula for effective capacity can therefore overstate or understate reality if its factors are poorly measured. Nominal area calculations also omit behavioural effects such as congestion, informal storage and workarounds. Queueing can rise sharply as utilisation approaches 100%, so a mathematically feasible plan may be operationally unacceptable.

Other limitations include sunk-cost bias, pressure to justify a predetermined investment, underestimation of construction disruption and failure to include support functions. Managers should expose assumptions, use ranges, validate data with operators and revisit the plan as demand, technology and regulation change.

Best-Practice Principles

• Plan from demand scenarios, not a single forecast.
• Use consistent time bases and units.
• Separate design capacity, effective capacity and actual output.
• Include yield, setups, maintenance, breaks and changeovers.
• Protect the bottleneck and avoid local optimisation.
• Size space for safe operation, maintenance and material flow.
• Reserve expansion options where uncertainty is significant.
• Evaluate total cost, risk and reversibility.
• Validate calculations with operators, engineers and finance specialists.
• Establish measurable triggers for review and expansion.

Chapter Summary

Demand, capacity and space planning converts uncertain market requirements into an integrated physical and operational plan. Demand must be described by volume, mix, timing and service expectations. Capacity analysis distinguishes design, effective and actual capacity and uses utilisation, efficiency and capacity cushion measures appropriately. Workload calculations should include processing time, setups, yield losses and product mix. Equipment and labour requirements follow from workload divided by effective resource capacity, with upward rounding where resources are discrete.

Bottleneck analysis is essential because system throughput is constrained by the limiting resource. Space planning extends beyond machine footprints to include operating clearances, storage, material handling, utilities, support functions, safety routes and expansion allowances. Because forecasts and operating conditions are uncertain, scenarios, sensitivity analysis, simulation and staged investment improve decision quality. The most defensible plans integrate financial, operational, safety, environmental and strategic considerations and establish explicit governance triggers for revision.

Key Terms

Demand: Quantity customers are willing and able to purchase in a specified period.

Independent demand: External customer demand not derived from another item.

Dependent demand: Demand calculated from a parent product or process requirement.

Design capacity: Theoretical maximum output under ideal conditions.

Effective capacity: Realistic capacity after planned operating losses.

Actual output: Output actually achieved during a period.

Utilisation: Actual output divided by design capacity.

Efficiency: Actual output divided by effective capacity.

Capacity cushion: Capacity held above expected demand.

Takt time: Net available production time divided by customer demand.

Workload: Resource time required to produce the planned output.

Yield: Proportion of input that becomes acceptable output.

Bottleneck: Resource that constrains total system throughput.

Scalability: Ability to change capacity without disproportionate cost or disruption.

Capacity expansion: Action that increases available production capability.

Net functional area: Space directly required by equipment or activities.

Circulation area: Space for movement, access and material handling.

Digital twin: Digital representation used to analyse or simulate a physical system.

Capacity trigger: Defined condition that initiates review or investment.

Review Questions

Short Answer Questions

1. Distinguish design capacity from effective capacity.
2. Why is takt time not the same as processing time?
3. Define a capacity cushion and explain its purpose.
4. What is a bottleneck?
5. Name four categories of space that should be included in a facility plan.

Long Answer Questions

1. Explain how a demand forecast is translated into equipment and labour requirements.
2. Discuss the strengths and limitations of utilisation and efficiency as performance measures.
3. Explain how yield, setup time and product mix affect capacity calculations.
4. Compare modular expansion, subcontracting and a new facility as capacity strategies.
5. Describe an integrated procedure for developing a capacity-and-space plan.

Analytical / Application Questions

1. A facility must produce 36,000 good units annually at a 90% yield. Each unit requires 1.5 minutes, and the resource has 1,800 scheduled hours with 85% availability and 95% performance. Calculate gross production, workload, effective capacity in hours and the number of resources required.
2. A three-stage process has effective capacities of 52,000, 47,000 and 55,000 units per year. Identify the bottleneck and explain two actions that could increase system throughput.
3. A manager proposes using average inventory to size storage despite highly seasonal demand. Critically evaluate this proposal and recommend a more defensible planning method.

Multiple Choice Questions with Answers

Question 1

Which measure compares actual output with effective capacity?

A. Design capacity

B. Utilisation

C. Efficiency

D. Capacity cushion

Correct Answer: C. Efficiency

Question 2

A facility has demand of 12,000 units and net available time of 9,600 minutes. What is the required production rate?

A. 0.80 units per minute

B. 1.25 units per minute

C. 8.00 units per minute

D. 21.60 units per minute

Correct Answer: B. 1.25 units per minute

Question 3

If good output required is 2,000 units and yield is 95%, approximately how many units must be produced?

A. 1,900

B. 2,000

C. 2,050

D. 2,106

Correct Answer: D. 2,106

Question 4

In a serial process, which resource normally determines system capacity?

A. The resource with the highest cost

B. The resource with the lowest effective capacity

C. The resource with the largest floor area

D. The resource with the greatest labour count

Correct Answer: B. The resource with the lowest effective capacity

Question 5

Which item is normally excluded from net machine operating area?

A. Maintenance clearance

B. Operator access

C. Safety clearance

D. Corporate headquarters parking

Correct Answer: D. Corporate headquarters parking

Question 6

A machine has 2,000 scheduled hours, availability 0.90, performance 0.95 and yield 0.98. What is its effective time approximately?

A. 1,672 hours

B. 1,800 hours

C. 1,862 hours

D. 1,960 hours

Correct Answer: A. 1,672 hours

Question 7

What is the main purpose of a capacity cushion?

A. To eliminate all fixed costs

B. To hold capacity above expected demand for uncertainty and responsiveness

C. To ensure utilisation is always 100%

D. To replace preventive maintenance

Correct Answer: B. To hold capacity above expected demand for uncertainty and responsiveness

Question 8

Which action is most directly associated with protecting a bottleneck?

A. Adding space to a non-bottleneck storage area only

B. Allowing the bottleneck to remain idle during breaks

C. Reducing setup and downtime at the bottleneck

D. Increasing inspection capacity while assembly remains constrained

Correct Answer: C. Reducing setup and downtime at the bottleneck

Question 9

Why should peak inventory often be used for storage planning?

A. Average inventory is always zero

B. Peak conditions determine whether required service and access can be maintained

C. Peak inventory is always cheaper than average inventory

D. Storage space cannot be expanded

Correct Answer: B. Peak conditions determine whether required service and access can be maintained

Question 10

Which feature most improves scalability?

A. A layout that permanently occupies all available expansion space

B. Equipment designed only for one product and one volume

C. Modular equipment and reserved expansion zones

D. Ignoring utility capacity until demand occurs

Correct Answer: C. Modular equipment and reserved expansion zones