Capacity, Location and Facility Planning

Introduction

Capacity, location, and facility planning are long-term operations decisions that determine how much an organization can produce, where it will operate, and how its physical resources will support customers and supply-chain partners. These decisions are difficult to reverse, require substantial capital, and influence cost, responsiveness, resilience, sustainability, and competitive positioning.

Capacity planning concerns the volume and timing of resources required to meet demand. Location planning determines the geographic position of plants, warehouses, service outlets, and distribution centers. Facility planning integrates these choices with process design, technology, transportation networks, labor availability, regulatory requirements, and risk.

A business analytics perspective is particularly valuable because these decisions involve uncertain demand, multiple objectives, trade-offs, and large data sets. Analysts use forecasts, scenario analysis, break-even models, queuing analysis, decision trees, factor-rating methods, center-of-gravity calculations, and optimization models to support managerial judgment.

Learning Objectives

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

1. Explain capacity concepts and compare major capacity strategies.
2. Distinguish design, effective, and actual capacity and calculate utilization and efficiency.
3. Evaluate economies of scale, demand uncertainty, and capacity cushions.
4. Analyze capacity alternatives using break-even analysis and decision trees.
5. Explain how capacity affects queues, waiting time, and service quality.
6. Apply factor-rating, center-of-gravity, and transportation-optimization approaches to location decisions.
7. Assess strategic, ethical, environmental, and technological issues in facility planning.

1. Capacity as a Strategic Decision

Capacity is the maximum sustainable output or service potential of a resource, facility, process, or system over a specified period. It must always be expressed with a time basis and a unit of output: for example, 10,000 units per month, 250 customer visits per day, or 40 pallets per hour.

Three capacity concepts are especially important:

• Design capacity is the theoretical maximum output under ideal conditions.
• Effective capacity is the realistic maximum output after allowing for product mix, maintenance, breaks, setup time, quality procedures, and other operating constraints.
• Actual output is what the system really produces.

Two related measures are utilization and efficiency:

Utilization compares output with the theoretical maximum, whereas efficiency compares output with the practical operating maximum. If a factory has design capacity of 12,000 units per month, effective capacity of 10,000 units, and actual output of 8,500 units, then utilization is 70.8% and efficiency is 85.0%.

These ratios must use the same period and output unit. A facility may show high utilization but low efficiency if its design capacity is unrealistic or if effective capacity has been poorly estimated.

Capacity Decisions at Different Levels

Capacity decisions occur at several levels:

Strategic capacity should be aligned with the organization’s competitive priorities. A low-cost manufacturer may favor large, highly automated facilities, while a rapid-response retailer may value multiple smaller facilities close to customers.

2. Capacity Strategies and Capacity Cushion

Demand rarely matches capacity perfectly. Organizations therefore choose how and when to add capacity.

Lead, Lag, and Match Strategies

A lead strategy adds capacity before demand is fully present. It supports rapid growth and short response times but creates a risk of idle resources and excess fixed cost.

A lag strategy adds capacity only after demand has materialized. It reduces the risk of overinvestment but can cause lost sales, long queues, overtime, and customer dissatisfaction.

A match strategy adds capacity in smaller increments as demand develops. It balances risk but requires frequent decisions and may sacrifice some economies of scale.

A related concept is the capacity cushion, the amount by which capacity exceeds expected demand:

For example, if expected demand is 8,000 units per month and effective capacity is 10,000 units, the cushion is 20%. The formula assumes that demand and capacity are measured over the same period and in comparable units.

The appropriate cushion depends on demand variability, service expectations, supply risk, expansion lead time, and the cost of unused capacity. Hospitals, emergency services, and telecommunications networks generally require larger cushions than stable, make-to-stock operations.

Economies and Diseconomies of Scale

Economies of scale occur when average cost falls as output increases. Causes include spreading fixed costs over more units, specialization, volume purchasing, and more efficient equipment. Diseconomies of scale arise when size increases coordination difficulty, congestion, bureaucracy, transportation distance, or managerial complexity.

Average cost can be expressed as:

A larger facility is not automatically superior. The relevant comparison includes construction, labor, inventory, transportation, quality, coordination, risk, and future flexibility. An apparently low-cost centralized facility may increase delivery distances and vulnerability to disruption.

3. Demand Uncertainty and Capacity Flexibility

Capacity planning should not rely solely on a single-point forecast. Forecasts contain error because of changing customer preferences, competitor behavior, economic conditions, seasonality, and technological disruption.

Useful analytical practices include:

• developing high, medium, and low demand scenarios;
• calculating forecast-error ranges and prediction intervals;
• identifying capacity triggers, such as utilization above 85% for several periods;
• using flexible labor, modular equipment, outsourcing, and temporary space;
• evaluating the cost of excess capacity against the cost of shortage;
• testing plans with sensitivity analysis and simulation.

Flexibility has several forms. Volume flexibility allows output to rise or fall. Mix flexibility allows the system to produce different products. Routing flexibility permits work to move through alternative resources. Expansion flexibility allows future growth without abandoning current investments.

Capacity plans should distinguish committed capacity from scalable capacity. Committed capacity is difficult to change, whereas scalable capacity can be added through overtime, temporary labor, contract manufacturing, cloud services, or leased space.

4. Break-Even Analysis for Capacity Alternatives

Break-even analysis compares alternatives whose fixed and variable costs differ. For a single product, total cost is:

where:

• TC = total cost;
• F = fixed cost for the period;
• v = variable cost per unit;
• Q = quantity produced.

If two alternatives have fixed costs F₁, F₂ and unit variable costs v₁, v₂, their cost-indifference volume is:

This equation applies when the denominator is nonzero and both alternatives are being compared over the same period, product definition, and relevant range. If the alternatives also have different revenues, profit should be compared rather than cost alone.

Worked example. A company can retain a manual process or install automated equipment. The manual alternative has fixed cost of $120,000 and variable cost of $18 per unit. Automation has fixed cost of $360,000 and variable cost of $10 per unit.

Below 30,000 units, the manual process has lower cost. Above 30,000 units, automation has lower cost, assuming the stated costs and capacities remain valid.

If the alternatives have different maximum capacities, quality levels, lead times, or strategic benefits, break-even analysis is only a starting point. Managers should add relevant nonfinancial criteria and test demand scenarios.

5. Decision Trees and Capacity Investment Under Uncertainty

A decision tree represents sequential decisions and uncertain events. A square denotes a decision node; a circle denotes a chance node. The analyst calculates expected monetary value (EMV) by multiplying each outcome by its probability and summing the results:

where p_i is the probability of outcome i and V_i is its value. Probabilities should sum to 1, and monetary values should be measured on the same basis, usually present value if outcomes occur at different times.

Worked example. A firm may build a large facility for $5 million or a small facility for $3 million. If demand is high, the large facility produces a net benefit of $8 million before construction cost; if demand is low, it produces $2 million. Probabilities are 0.6 and 0.4. The small facility produces $5 million in high demand and $2.5 million in low demand, before its construction cost.

Large-facility EMV:

Small-facility EMV:

On expected monetary value alone, the small facility is preferred. However, the large facility may be favored if the organization places a high value on strategic growth, resilience, or avoiding future expansion delays. Decision trees do not eliminate judgment; they make assumptions and trade-offs visible.

6. Capacity, Queues, and Service Performance

In service systems, capacity affects waiting time as well as output. A queue forms when demand temporarily arrives faster than service can be provided. Even when average capacity exceeds average demand, variability can produce waiting.

For a simple single-server model, the utilization rate is:

where λ is the average arrival rate and μ is the average service rate, both measured in customers per unit of time. A stable basic queue requires ρ < 1. As ρ approaches 1, expected waiting time increases sharply.

For an M/M/1 queue—Poisson arrivals, exponentially distributed service times, one server, stable steady-state operation—the expected number in the system is:

and the expected time in the system is:

Little’s Law provides a broader relationship:

Here, L is the average number in the system, λ is throughput or arrival rate, and W is average time in the system. Little’s Law applies to a stable system over a suitable observation period, even when the detailed probability assumptions of M/M/1 are not met.

Worked example. A service desk receives 18 customers per hour and can serve 24 per hour. Thus ρ = 18÷24 = 0.75. Under the M/M/1 assumptions:

The result includes both waiting and service time. The model is a useful approximation, not a guarantee. Multiple servers, finite queues, scheduled arrivals, priority rules, and non-exponential service times require other models or simulation.

Capacity decisions should therefore consider both average utilization and variability. Operating a service system at 98% utilization may appear efficient but can produce unacceptable waits and poor recovery from disruptions.

7. Facility Location: Strategic Logic and Data

Location decisions concern the placement of manufacturing plants, warehouses, retail outlets, hospitals, offices, and service centers. Important drivers include:

• proximity to customers and suppliers;
• transportation cost and delivery time;
• labor availability, skills, wages, and productivity;
• land, utilities, taxes, and construction cost;
• infrastructure and digital connectivity;
• political, legal, and regulatory conditions;
• environmental exposure and climate risk;
• community impacts and quality of life;
• access to complementary firms and innovation ecosystems.

The best location depends on the organization’s strategy. A high-service e-commerce network may favor many regional facilities, while a bulk-material producer may locate near raw materials. A location should be evaluated as part of the network rather than as an isolated site.

Factor-Rating Method

The factor-rating method combines qualitative and quantitative criteria. The analyst assigns each factor a weight, rates each location, and calculates a weighted score:

where S_j is the score for location j, w_i is the weight of factor i, and r_ij is the rating of location j on factor i. Weights should sum to 1, and ratings must use a consistent scale.

The tie demonstrates that the method does not automatically produce a unique answer. Ratings may be subjective, and results should be tested with sensitivity analysis. A site that ranks highly under one set of weights may rank poorly when customer proximity becomes more important.

Center-of-Gravity Method

The center-of-gravity method identifies a geographic point that minimizes weighted distance under simplified assumptions. For demand points with coordinates (x_i,y_i) and shipment weights w_i, the coordinates are:

Coordinates must use a consistent system, and shipment weights must represent comparable volume, tonnage, or trips. The method assumes transportation cost is approximately proportional to volume and distance and usually ignores roads, terrain, congestion, taxes, and fixed facility costs. The calculated point is a screening location, not necessarily a feasible site.

Worked example. Three markets have coordinates and monthly demand as follows: Market 1 (2, 8), 400 units; Market 2 (8, 4), 300 units; Market 3 (6, 10), 200 units.

The estimated point is approximately (4.89, 7.11). Analysts must then check whether land, roads, zoning, and infrastructure are available near that point.

Transportation and Network Optimization

When several facilities can serve several markets, the transportation model can minimize total shipping cost:

subject to supply constraints:

and demand constraints:

with:

Here, x_ij is the quantity shipped from facility i to market j, c_ij is unit transportation cost, s_i is facility supply capacity, and d_j is market demand. If total supply equals total demand, inequalities may be written as equalities. Fixed facility-opening costs, service-level requirements, integer decisions, and carbon limits require an expanded mixed-integer model.

Optimization provides a structured answer to a defined objective. It does not make inaccurate data, unrealistic constraints, or omitted risks disappear.

8. Integrated Facility Planning

Capacity and location should be evaluated together. A facility with low production cost may be unattractive if it is far from customers or vulnerable to disruption. Conversely, several small sites may improve responsiveness but increase fixed cost, duplicated inventory, and managerial complexity.

A robust planning process typically follows these steps:

1. Define the service promise, products, markets, and planning horizon.
2. Forecast demand using baseline and alternative scenarios.
3. Convert demand into capacity requirements by period and process stage.
4. Identify feasible capacity strategies and expansion triggers.
5. Screen locations using legal, technical, environmental, and social constraints.
6. Compare costs, service, risk, sustainability, and flexibility.
7. Optimize flows where data quality supports quantitative modeling.
8. Stress-test the preferred plan against disruption and demand changes.
9. Establish governance, performance measures, and review points.

Relevant performance measures include capacity utilization, efficiency, unit cost, order cycle time, on-time delivery, average queue time, inventory, carbon emissions, labor turnover, and return on invested capital. Measures should not encourage local optimization—for example, maximizing plant utilization while creating excessive inventory or late deliveries.

9. Illustrative Case: NorthStar Medical Devices

Background

NorthStar Medical Devices is an illustrative manufacturer of diagnostic cartridges. Demand is growing in three regions: North, Central, and South. The current plant is near capacity, and management must decide whether to expand it, build a second plant, or outsource part of production.

Problem

The current plant has design capacity of 1.2 million cartridges annually and effective capacity of 1.0 million. Expected demand next year is 900,000 units, but scenario analysis estimates 700,000 units under low demand and 1.3 million under high demand. Medical customers require 98% on-time delivery, and stockouts are costly.

Analysis

The current facility’s expected utilization is:

Its expected efficiency is:

Although design utilization appears comfortable, effective capacity leaves only 100,000 units of practical headroom. The high-demand scenario creates a shortfall of 300,000 units.

The team considers three alternatives:

• Expand the existing site: low coordination cost and strong quality control, but limited room for future expansion.
• Build a regional second plant: higher fixed cost, but shorter delivery distances and better resilience.
• Outsource overflow: flexible and fast to activate, but with risks involving quality, intellectual property, and supplier dependence.

A factor-rating exercise gives the regional site a higher score for customer proximity and resilience, while the existing site scores higher for labor experience and quality systems. A center-of-gravity calculation places the candidate regional facility near the Central market, but zoning analysis eliminates the mathematically ideal parcel. A transportation model then compares feasible parcels using actual road distances and capacity limits.

Approach and Solution

NorthStar selects a staged plan: improve effective capacity at the current plant through setup reduction and preventive maintenance, contract a qualified supplier for a limited overflow volume, and reserve land near the Central market for a modular second plant. The expansion trigger is defined as two consecutive quarters with forecast demand above 950,000 units and a projected service risk above the target.

Supplier qualification includes process validation, audit rights, traceability, cybersecurity controls, and a contingency source. The company also evaluates the carbon and labor implications of each alternative rather than selecting solely on unit cost.

Outcome

The staged plan avoids immediate overbuilding under the low-demand scenario while preserving a credible response to high demand. It also separates reversible decisions—supplier qualification and process improvement—from less reversible decisions such as land acquisition and construction.

Learning Points

1. Design capacity can conceal practical constraints.
2. A capacity cushion should reflect uncertainty and service requirements.
3. Location analysis should move from geographic screening to feasible-site and network analysis.
4. Staged investment can preserve flexibility.
5. Quality, resilience, ethics, and sustainability belong in the decision model, not as afterthoughts.

10. Risks, Ethics, Sustainability, and Technology

Facility decisions affect employees, communities, suppliers, customers, and the environment. Ethical analysis should consider worker safety, fair labor standards, displacement, pollution, water use, noise, traffic, data privacy, and equitable access to essential services. A low-cost location should not be selected because regulation is weak or vulnerable workers have limited bargaining power.

Environmental assessment may include energy use, greenhouse-gas emissions, packaging, transport distance, waste, water stress, and climate exposure. Total cost of ownership should include expected disruption, insurance, adaptation, and remediation costs.

Digital technologies improve planning through geographic information systems, real-time transport data, digital twins, demand sensing, optimization solvers, and simulation. Artificial intelligence can identify demand patterns and generate scenarios, but models require representative data, validation, monitoring, and human accountability. Automated recommendations may reproduce historical bias—for example, systematically deprioritizing communities whose data are incomplete. Sensitive location and labor data must be governed through access controls, privacy safeguards, documentation, and auditability.

Contemporary facility planning increasingly emphasizes resilience: multiple suppliers, geographic diversification, backup capacity, renewable energy, cyber protection, and rapid recovery procedures. Resilience can increase cost in normal conditions, but its value becomes evident when disruption probabilities and consequences are considered.

Best-Practice Checklist

Before approving a capacity or location decision, managers should ask:

• Are demand scenarios, forecast error, and seasonality explicit?
• Are design, effective, and actual capacity clearly distinguished?
• Have queues and variability been analyzed rather than average utilization alone?
• Are fixed, variable, transportation, inventory, quality, and disruption costs included?
• Have qualitative factors and stakeholder effects been documented?
• Are optimization assumptions, constraints, and data sources transparent?
• Has the recommendation been tested under alternative weights and scenarios?
• Is there a staged implementation plan with measurable triggers?
• Are safety, labor, environmental, privacy, and governance requirements satisfied?

Chapter Summary

Capacity planning determines the volume and timing of resources needed to serve demand. Design capacity represents theoretical potential, effective capacity reflects realistic operating conditions, and actual output is observed production. Utilization and efficiency compare output with design and effective capacity, respectively.

Lead, lag, and match strategies express different approaches to investment timing. Capacity cushions protect service performance under uncertainty, while economies of scale must be balanced against diseconomies, transportation, coordination, and resilience risks. Break-even analysis identifies cost-indifference volumes, and decision trees evaluate sequential choices under uncertain demand.

Capacity affects queues and customer experience. Utilization approaching 100% can cause waiting times to rise sharply, and Little’s Law links average inventory or customers in a system with throughput and time. Location decisions require strategic, financial, geographic, regulatory, environmental, and social analysis. Factor-rating supports structured comparison, center-of-gravity provides a geographic screening point, and transportation optimization allocates flows subject to supply and demand constraints.

The strongest plans integrate capacity, location, network flows, uncertainty, sustainability, and governance. Analytics improves transparency and consistency, but managerial judgment remains necessary because models depend on assumptions, data quality, and the values embedded in objectives and constraints.

Key Terms

Actual output: The quantity actually produced or served during a specified period.

Capacity: The maximum sustainable output or service potential of a resource or system.

Capacity cushion: The proportion by which effective capacity exceeds expected demand.

Center of gravity: A weighted geographic estimate based on demand locations and shipment volumes.

Design capacity: The theoretical maximum output under ideal conditions.

Diseconomies of scale: Rising average cost caused by excessive size or complexity.

Effective capacity: Realistic maximum output after normal operating constraints.

Efficiency: Actual output divided by effective capacity.

Economies of scale: Declining average cost as output increases over a relevant range.

Factor rating: A weighted scoring method for comparing locations or alternatives.

Lead capacity strategy: Adding capacity before demand fully materializes.

Lag capacity strategy: Adding capacity only after demand has been demonstrated.

Little’s Law: The relationship L = λW between system size, throughput, and time.

Queuing: The waiting of entities for service or processing.

Utilization: Actual output divided by design capacity.

Match capacity strategy: Adding capacity incrementally as demand develops.

Transportation model: An optimization model allocating shipments from sources to destinations.

Capacity flexibility: The ability to adjust volume, mix, routing, or expansion over time.

Review Questions

Short Answer Questions

1. Distinguish design capacity, effective capacity, and actual output.
2. What is the difference between a lead and a lag capacity strategy?
3. Why can high utilization create service problems?
4. State two assumptions of the center-of-gravity method.
5. What does a factor-rating score represent?

Long Answer Questions

1. Explain how economies and diseconomies of scale influence facility-size decisions.
2. Compare break-even analysis and decision-tree analysis for capacity planning.
3. Explain the relationship between utilization, variability, and queuing performance.
4. Describe a systematic process for evaluating alternative facility locations.
5. Discuss how sustainability, ethics, and resilience should be incorporated into facility planning.

Analytical / Application Questions

1. A facility has design capacity of 50,000 units per month, effective capacity of 42,000 units, and actual output of 36,000 units. Calculate utilization and efficiency, then interpret the results.
2. Two process alternatives have costs TC_A = $80,000 + $24Q and TC_B = $200,000 + $18Q. Find the cost-indifference volume and recommend an alternative for demand of 15,000 units, assuming no other differences.
3. A service center receives 30 customers per hour and serves 40 per hour through one server. Under M/M/1 assumptions, calculate utilization and expected time in the system, and explain why the result should be treated as an approximation.

Multiple Choice Questions with Answers

Question 1

A plant has design capacity of 12,000 units per month and actual output of 9,000 units. What is utilization?

A. 60%

B. 75%

C. 90%

D. 133%

Correct Answer: B. 75%

Question 2

Which capacity strategy adds resources before demand is fully realized?

A. Lag strategy

B. Lead strategy

C. Match strategy

D. Shutdown strategy

Correct Answer: B. Lead strategy

Question 3

If expected demand is 8,000 units and effective capacity is 10,000 units, what is the capacity cushion?

A. 10%

B. 20%

C. 25%

D. 80%

Correct Answer: B. 20%

Question 4

Two alternatives have fixed costs of $100,000 and $220,000, with variable costs of $20 and $14 per unit, respectively. At what volume are costs equal?

A. 10,000 units

B. 20,000 units

C. 30,000 units

D. 40,000 units

Correct Answer: C. 30,000 units

Question 5

In a stable single-server queue, which condition is required by the basic model?

A. λ = 0

B. λ > μ

C. ρ > 1

D. ρ < 1

Correct Answer: D. ρ < 1

Question 6

Which method calculates a weighted geographic point using coordinates and shipment volumes?

A. Factor rating

B. Center of gravity

C. Decision tree

D. Break-even analysis

Correct Answer: B. Center of gravity

Question 7

In a factor-rating model, what should the location-factor weights generally sum to?

A. 0

B. 0.5

C. 1.0

D. 100.0

Correct Answer: C. 1.0

Question 8

What does the transportation model primarily determine?

A. Employee compensation

B. Shipment allocations between facilities and markets

C. Product design specifications

D. Forecast confidence intervals

Correct Answer: B. Shipment allocations between facilities and markets

Question 9

A service system has arrival rate 18 customers per hour and service rate 24 customers per hour. What is utilization?

A. 0.25

B. 0.50

C. 0.75

D. 1.33

Correct Answer: C. 0.75

Question 10

Which statement best describes Little’s Law?

A. Average cost equals fixed cost divided by output

B. System size equals throughput multiplied by time in the system

C. Utilization equals design capacity divided by actual output

D. Break-even volume equals fixed cost multiplied by variable cost

Correct Answer: B. System size equals throughput multiplied by time in the system