Warehouse Network, Location and Capacity Strategy

Introduction

Warehouse strategy is a long-term design problem concerning the structure, ownership, location, size, technology and operating role of facilities through which inventory moves. It is not simply a matter of finding the cheapest building. A warehouse network determines the speed with which customers can be served, the amount of inventory required, exposure to disruption, transport emissions, regulatory compliance and the organisation’s ability to scale.

The central strategic problem is a trade-off. More warehouses may reduce outbound delivery distance and improve responsiveness, but they usually increase facility costs, duplicated inventory and managerial complexity. A centralised network can exploit scale and pool demand uncertainty, but it may increase delivery distance and vulnerability to a single major disruption. The appropriate design therefore depends on customer-service requirements, product characteristics, demand geography, supply structure, risk appetite and the economics of transport and inventory.

This chapter develops a framework for deciding warehouse ownership, number, location, capacity, centralisation, outsourcing, cross-docking and broader network configuration. It combines strategic reasoning with quantitative models and managerial judgement.

Learning Objectives

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

1. Explain the strategic role of warehouse networks in supply-chain performance.
2. Compare centralised, decentralised, regional, hybrid and hub-and-spoke configurations.
3. Evaluate warehouse ownership and outsourcing alternatives.
4. Apply location, capacity and total-cost models with appropriate assumptions.
5. Analyse the effects of facility count on transport, inventory, service and risk.
6. Design cross-docking and postponement arrangements where operationally suitable.
7. Incorporate resilience, sustainability, technology, governance and regulatory requirements into network decisions.

1. The Warehouse as a Network Node

A warehouse performs more than storage. Its roles may include receiving, quality inspection, inventory buffering, order picking, consolidation, break-bulk, postponement, returns processing, repair, value-added packaging and customer fulfilment. The same physical building may perform different roles for different product families.

A network consists of nodes and links. Nodes include suppliers, plants, ports, warehouses, fulfilment centres, stores and customers. Links include road, rail, air, sea and information flows. A facility decision must therefore be evaluated as part of the complete network rather than in isolation.

The principal performance dimensions are:

A warehouse network should be aligned with the supply-chain operating model. Stable, predictable demand may support efficient centralisation. Highly time-sensitive, customised or geographically dispersed demand may justify regional facilities. E-commerce often requires facilities designed for individual-order fulfilment rather than pallet storage alone.

2. Strategic Network Configurations

2.1 Centralised network

A central warehouse or distribution centre serves a broad market. It generally provides scale economies in material handling, information systems, labour supervision and inventory pooling. Demand variability from different markets may partially offset, reducing aggregate safety stock. Its weaknesses include greater average outbound distance, potential transport dependence and high consequences if the facility fails.

2.2 Decentralised network

Several local warehouses serve defined territories. This can reduce delivery time and last-mile distance, support local market knowledge and improve continuity when one facility is unavailable. However, inventory is duplicated, fixed costs rise and stock imbalances become more likely.

2.3 Regional network

Regional distribution centres occupy an intermediate position. They are useful when customer geography is broad, delivery promises are strict and transport economics do not favour one national facility. Regionalisation should not be assumed to be optimal merely because it appears balanced; the number and boundaries of regions require analysis.

2.4 Hub-and-spoke network

A central hub consolidates inbound flows and transfers inventory or shipments to smaller spokes. It can exploit line-haul scale while retaining local service capability. The design is effective when shipment consolidation is valuable and transfer time is acceptable. It is less suitable when the hub becomes a severe bottleneck or when products require immediate direct fulfilment.

2.5 Hybrid and postponement networks

A hybrid network may centralise slow-moving or high-value items and decentralise fast-moving items. It may also centralise generic inventory and complete final configuration near the customer. This postponement approach delays product differentiation, reducing the number of finished-stock variants held at each location.

A useful strategic principle is that network differentiation should follow product and customer requirements. A single network architecture for every stock-keeping unit is often inefficient.

3. The Economics of the Number of Warehouses

The number of facilities affects several cost categories simultaneously:

• inbound transport and transfer costs;
• outbound transport and last-mile costs;
• facility rent, rates, maintenance and utilities;
• labour and material-handling equipment;
• information-system and management costs;
• cycle stock and safety stock;
• obsolescence, damage and shrinkage;
• taxes, duties and compliance costs;
• disruption, recovery and carbon costs.

The basic network objective can be represented as:

where TC is total annual network cost; C_F is facility cost; C_T is transport cost; C_I is inventory cost; C_H is handling and operating cost; C_R is risk or expected disruption cost; and C_E is relevant environmental or externality cost. The formulation is a decision framework rather than a universal accounting identity: each component must be defined consistently and double counting must be avoided.

Facility and transport trade-offs often produce a U-shaped total-cost curve. At very few facilities, transport costs may be high. At very many facilities, fixed cost, duplicated inventory and management complexity may dominate. The minimum of the curve is not necessarily the best strategic choice if service or resilience constraints are violated.

3.1 Inventory pooling and safety stock

Centralisation can reduce safety stock because independent demand variation is pooled. Under a simple periodic-review approximation, safety stock for a single location is:

where SS is safety stock in units, z is the standard-normal service factor, and σ_LT is the standard deviation of demand during lead time. The equation assumes that demand during lead time is approximately normally distributed and that the selected service measure is a cycle-service probability.

If demand is independent across n identical locations, each with standard deviation σ, aggregate standard deviation is:

If all demand were pooled centrally, the standard deviation of total demand would be approximately σ√n, rather than nσ. Thus, pooled safety stock grows more slowly than the sum of independently positioned safety stocks. This is the statistical basis of risk pooling. The result weakens when demands are positively correlated, lead times differ, products are location-specific, or service constraints require local stock.

For two demand streams with standard deviations σ₁ and σ₂ and correlation ρ, the pooled standard deviation is:

Here −1 ≤ ρ ≤ 1. Positive correlation reduces the benefit of pooling; negative correlation increases it. Units on both sides are demand units, so the equation is dimensionally consistent.

3.2 Square-root approximation

A commonly used approximation states that total safety stock under decentralisation may scale approximately with the square root of the number of locations, provided demand is independent and locations are otherwise comparable. This is a conceptual approximation, not a substitute for SKU-level simulation. It should not be applied to deterministic cycle stock, capacity requirements or strongly correlated demand without qualification.

4. Warehouse Location Analysis

Location analysis normally proceeds from broad screening to detailed site evaluation.

4.1 Strategic screening

The first stage identifies feasible countries, regions or metropolitan areas. Relevant criteria include customer density, supplier and port access, road and rail quality, labour availability, land cost, utility reliability, taxation, customs arrangements, climate exposure, political stability and future development plans.

A location that is inexpensive today may be strategically poor if congestion, labour scarcity or land constraints make expansion impossible. Conversely, a more expensive location may be justified by reliable service and lower total logistics cost.

4.2 Centre-of-gravity method

A basic location-screening tool uses weighted geographic coordinates:

Here x_i,y_i are the coordinates of demand or supply point i, and w_i is its weight, such as annual shipment volume, mass or transport spend. The result is a geometric screening point, not automatically a feasible site. It assumes a continuous plane and simplified transport relationships; real roads, terrain, borders and congestion may substantially change the answer.

Worked example. Three markets have coordinates and annual shipment weights as follows:

Total weight is 400. Therefore:

x^* = (100×10 + 200×40 + 100×20)/400 = 27.5

y^* = (100×20 + 200×30 + 100×60)/400 = 35

The screening point is (27.5, 35). The organisation must next test actual sites near this point for road access, land, labour, planning permission, flood risk and operating cost.

4.3 Factor-rating method

When both quantitative and qualitative factors matter, a weighted factor-rating method can be used:

where Score_j is the score of site j, α_k is the weight of criterion k, and s_jk is the site’s rating on that criterion. Weights should sum to one and ratings must use a consistent scale. The method structures judgement but does not eliminate subjectivity. Sensitivity analysis should test whether small changes in weights reverse the ranking.

4.4 Discrete location models

For a set of candidate sites and customer zones, a simplified capacitated facility-location model is:

subject to:

Here F_j is the fixed annual cost of opening site j, y_j indicates whether it opens, c_ij is unit allocation or transport cost from site j to zone i, q_ij is quantity supplied, d_i is demand in zone i, and K_j is site capacity. The model assumes known demand, additive costs and feasible allocation. Service constraints, multiple products, inventory, carbon and disruption scenarios can be added, but complexity increases rapidly.

5. Capacity Strategy

Capacity must be assessed in both physical and operational terms. A building may have sufficient floor area but insufficient dock doors, picking faces, charging capacity, labour, systems throughput or yard space.

Required storage capacity can be estimated as:

The equation is valid only when units are defined consistently and the buffer is explicitly tied to uncertainty, unusable space and operating policy. Peak inventory should reflect replenishment cycles, seasonality, safety stock, inbound variability and planned promotions—not merely average inventory.

For labour or equipment, utilisation is:

A utilisation of 100% is not normally a practical target because variability creates queues. Sustainable utilisation depends on process variability, labour flexibility and service requirements. High average utilisation can therefore produce poor throughput and long order delays.

Capacity options include:

• owning a purpose-built facility;
• leasing a standard facility;
• using flexible overflow space;
• adding shifts or temporary labour;
• redesigning storage and picking methods;
• outsourcing selected activities;
• sharing capacity with partners;
• postponing final configuration.

A robust plan distinguishes base capacity, surge capacity and recovery capacity. It also identifies trigger points for expansion, such as sustained utilisation, service failure or forecasted demand thresholds.

6. Ownership, Outsourcing and Governance

6.1 Ownership alternatives

An owned warehouse provides control over layout, labour practices, systems, security and strategic capability. It may be attractive when volumes are stable, the activity is core, specialised infrastructure is required or long-term asset control is valuable. Its disadvantages are capital commitment, demand risk, lower flexibility and responsibility for compliance and maintenance.

Leasing reduces initial capital expenditure and may improve flexibility, although lease obligations can still be long-term. A third-party logistics provider (3PL) offers external assets, expertise and potentially variable cost. A 4PL or lead logistics provider may coordinate several providers and transport modes, but governance and accountability become more complex.

6.2 Make-or-buy logic

The decision should compare relevant total cost and strategic consequences, not warehouse rent alone. Relevant considerations include:

• avoidable internal costs;
• provider price and pass-through charges;
• transition and integration costs;
• switching and termination costs;
• service-level penalties;
• data ownership and cybersecurity;
• labour and regulatory exposure;
• business continuity and provider financial health;
• loss or retention of process knowledge.

Outsourcing is particularly suitable where specialist scale, variable demand or geographic reach is important. It is less attractive when the operation contains distinctive customer knowledge, sensitive products, proprietary processes or a requirement for direct strategic control.

Contracts should specify measurable service-level agreements covering inventory accuracy, order cycle time, fill rate, damage, loss, labour compliance, data availability, audit rights, disaster recovery, subcontracting and exit assistance. Governance should include joint performance reviews and transparent root-cause analysis rather than relying only on penalty clauses.

7. Cross-Docking and Flow-Through Design

Cross-docking transfers received goods to outbound vehicles with little or no storage. It can reduce dwell time, handling and inventory, especially for predictable, high-volume products. Typical forms include pre-distribution cross-docking, where final destinations are known before receipt, and post-distribution cross-docking, where goods are allocated after arrival.

Cross-docking requires accurate advance shipping notices, synchronised appointments, compatible packaging, reliable transport, high inventory visibility and disciplined exception management. It is not simply “zero inventory.” Goods may wait in staging areas, and safety stock may remain elsewhere in the network.

Cross-docking is appropriate when demand is sufficiently predictable, inbound and outbound schedules can be synchronised, product identification is accurate and the cost of rapid coordination is lower than the cost of storage. It is risky for uncertain demand, fragile products, long inbound delays or highly variable order profiles.

8. Resilience, Risk and Sustainability

A low-cost network can be strategically weak if it has a single point of failure. Risk analysis should consider natural hazards, fire, cyberattack, labour disruption, supplier failure, port closure, energy interruption, political events and sudden demand changes.

Expected disruption cost can be represented simply as:

where p_s is the probability of disruption scenario s and L_s is the loss under that scenario. The probabilities should be understood as decision estimates, not precise facts. Loss may include lost contribution, emergency freight, recovery cost, penalties and reputational effects. Scenario analysis is often more useful than a single expected value because severe low-probability events may require specific safeguards.

Resilience choices include dual sourcing, alternative transport routes, reciprocal warehouse capacity, strategic buffer stock, geographically separated facilities, backup power, cyber controls and tested business-continuity plans. Redundancy has a cost, so the decision should reflect the criticality of products and the consequences of service failure.

Sustainable network design considers transport emissions, building energy, refrigerants, packaging, reverse logistics, land use and local traffic. A centralised network may reduce building duplication but increase delivery distance. The correct comparison requires system-wide measurement, not a narrow facility metric. Carbon should be measured using a stated boundary and consistent activity data.

9. Technology and Analytical Enablement

Warehouse-management systems provide inventory visibility, task control, slotting support, labour tracking and integration with transport and enterprise systems. Warehouse-control systems coordinate automation such as conveyors, sorters and autonomous mobile robots. Digital twins and discrete-event simulation can test alternative layouts, order profiles and capacity policies before physical investment.

AI can support demand forecasting, labour planning, slotting, anomaly detection, route prediction and predictive maintenance. However, model outputs depend on data quality and governance. Managers should monitor forecast bias, explainability, cybersecurity, privacy, model drift and the risk of automating a flawed process. AI should augment network judgement rather than obscure accountability.

Technology selection should follow the operating requirement. Automation is not automatically economical: high fixed cost, product variability, maintenance dependence and low utilisation can undermine the business case. A staged, modular architecture is often more robust than a technically ambitious system that cannot adapt.

10. Decision Process and Performance Measures

A disciplined network study can follow these steps:

1. Define customer segments, service promises, products and planning horizon.
2. Gather demand, supply, transport, inventory, cost, capacity and risk data.
3. Establish current-state performance and data-quality limitations.
4. Generate centralised, regional, outsourced, cross-dock and hybrid alternatives.
5. Model total cost, service, capacity, carbon and risk under base and stress scenarios.
6. Test assumptions through sensitivity and break-even analysis.
7. Select a preferred design and define implementation stages and contingencies.
8. Establish governance, metrics and review triggers.

Useful measures include order-cycle time, on-time-in-full delivery, fill rate, inventory accuracy, cost per order, cost per unit, dock-to-stock time, pick productivity, capacity utilisation, damage, shrinkage, carbon per shipment and recovery time after disruption. No single measure is sufficient. For example, high utilisation may coexist with late orders, and low warehouse cost may conceal expensive emergency transport.

11. Illustrative Case: Northstar Medical Supplies

Background

Northstar Medical Supplies is an illustrative case involving a distributor serving hospitals and pharmacies across a large country. It operates one national distribution centre near the main port. Demand has grown, customers increasingly require next-day delivery, and a major highway serving the facility has experienced repeated closures.

Problem

The national facility offers inventory pooling and low fixed cost, but average delivery distance is high. A proposed second regional facility would improve service but duplicate safety stock and require new systems and labour. Management must decide whether to retain the national model, add a regional site, outsource regional fulfilment or create a hybrid network.

Analysis

The team separates products into three groups. Emergency and fast-moving products require high availability and short lead time. Slow-moving, high-value products benefit strongly from centralised pooling. Temperature-sensitive products require validated storage, monitoring and contingency power.

The team constructs a capacitated facility-location model using hospital demand zones, candidate sites, transport rates, facility costs and service constraints. It then adds scenarios for highway closure, power interruption and a 25% demand surge. Qualitative assessment covers regulatory licensing, labour availability, cold-chain capability, data integration and provider resilience.

Approach and solution

The selected design is hybrid. The national facility retains slow-moving products, regulated inventory and central planning. A regional facility near the largest hospital cluster holds fast-moving emergency stock and provides late cut-off fulfilment. A qualified 3PL is contracted for overflow and disaster recovery, but Northstar retains inventory ownership, quality authority and customer data governance.

The network uses cross-docking for scheduled hospital replenishment where orders are confirmed before inbound receipt. Generic medical kits are assembled regionally, while final labels and customer-specific documentation are postponed until order release. The contract includes temperature excursions, inventory accuracy, recovery time, audit rights and exit support.

Outcome

The design increases fixed operating cost but reduces emergency freight and improves next-day coverage. Inventory is not decentralised uniformly: only selected high-volume items are regionalised. The backup 3PL is treated as resilience capacity rather than as the default lowest-cost provider. The result illustrates that network strategy is a portfolio of differentiated flows, not necessarily a choice between one warehouse and two identical warehouses.

Learning points

1. Product segmentation can reveal that different items require different network positions.
2. Service and resilience constraints may justify a cost increase.
3. Outsourcing can be used selectively without surrendering strategic control.
4. Scenario analysis is essential when a facility or transport route is critical.
5. Cross-docking and postponement depend on information and process reliability.

12. Challenges and Limitations

Network models may create false precision when demand forecasts, transport rates or disruption probabilities are uncertain. Average costs can hide peak-period congestion and service failures. Centre-of-gravity methods ignore actual road networks. Factor ratings may reflect managerial bias. Optimisation models may omit implementation difficulty, labour relations or political constraints.

A further limitation is path dependence. Existing leases, systems, labour skills and customer contracts restrict feasible choices. Transition risk can be material: inventory must be transferred, systems tested, staff trained and service maintained during the change. Therefore, a theoretically optimal long-term network may be inferior to a staged design with controlled migration.

Ethically, organisations must consider worker safety, fair labour practices, surveillance, community traffic, environmental justice and responsible treatment of outsourced workers. Regulatory requirements may cover food safety, pharmaceuticals, hazardous materials, customs, data protection, accessibility and building standards. Compliance should be designed into location and facility decisions rather than added after site selection.

Best-Practice Principles

• Design the network from customer and product requirements, not from available buildings alone.
• Evaluate total cost, service, risk, carbon and flexibility together.
• Use pooling benefits carefully and recognise demand correlation.
• Separate base, surge and recovery capacity.
• Validate model results with operational experts and site visits.
• Use sensitivity analysis for demand, transport rates, service targets and disruption assumptions.
• Outsource activities with clear governance, data rights and exit provisions.
• Treat cross-docking as a synchronisation system, not merely a storage-reduction technique.
• Segment products and flows where a single network policy creates avoidable compromise.
• Review the network periodically as demand, technology, regulation and risk change.

Chapter Summary

Warehouse network strategy determines how inventory and information are positioned between supply and demand. Its decisions include the number and location of facilities, ownership structure, capacity, centralisation, outsourcing, cross-docking and the configuration of transport and inventory flows.

Centralisation can reduce duplicated safety stock through risk pooling and exploit scale economies, while decentralisation can improve responsiveness and reduce delivery distance. The preferred balance depends on demand geography, correlation, customer promises, product characteristics, risk and sustainability. Centre-of-gravity and factor-rating methods support preliminary screening; capacitated facility-location models support more rigorous comparison, but their assumptions must be tested.

Capacity strategy must include storage, labour, docks, systems, yard space and surge capability. Ownership and outsourcing decisions require analysis of relevant total cost, control, flexibility, knowledge, compliance and provider risk. Cross-docking and postponement can accelerate flow and reduce finished-goods inventory when information and transport coordination are reliable.

A strategically sound network is not necessarily the cheapest under average conditions. It is a network that delivers acceptable cost and service while remaining resilient, scalable, governable and sustainable under realistic uncertainty. The strongest decisions combine quantitative models, scenario analysis, operational knowledge and explicit managerial judgement.

Key Terms

Warehouse network: The connected set of facilities and transport links used to position and move inventory.

Centralisation: Concentrating inventory or activities in relatively few facilities.

Decentralisation: Distributing inventory or activities across multiple geographically separated facilities.

Risk pooling: Reduction in aggregate variability achieved by combining demand or inventory requirements.

Distribution centre: A facility designed primarily for receiving, storing, processing and dispatching goods.

Centre of gravity: A weighted geographic screening method for identifying a potential facility area.

Factor rating: A weighted scoring technique for comparing sites against quantitative and qualitative criteria.

Facility-location model: An analytical model that selects facility sites and allocates demand subject to costs and constraints.

Capacity buffer: Additional capacity held to accommodate uncertainty, peaks or disruption.

Cross-docking: Direct transfer of inbound goods to outbound flows with limited storage.

Postponement: Delaying final product differentiation until demand or customer information is available.

3PL: A third-party logistics provider performing logistics activities for a client.

4PL: A lead logistics integrator coordinating multiple logistics resources or providers.

Service level: A defined performance commitment, such as availability or delivery time.

Network resilience: The ability to withstand, adapt to and recover from disruption.

Expected disruption cost: Probability-weighted loss associated with possible disruption scenarios.

Total cost to serve: The complete cost of supplying a customer, including facility, transport, inventory and handling effects.

Slotting: Assigning products to warehouse locations to improve handling and picking performance.

Digital twin: A data-enabled model used to represent and analyse a physical operation.

Reverse logistics: The movement and management of returns, repairs, recycling or disposal flows.

Review Questions

Short Answer Questions

1. Why can a centralised warehouse reduce safety stock?
2. State two limitations of the centre-of-gravity method.
3. What operational conditions are necessary for effective cross-docking?
4. Distinguish between base capacity and recovery capacity.
5. Identify four issues that should be addressed in a 3PL contract.

Long Answer Questions

1. Compare centralised, decentralised, regional and hybrid warehouse networks. Explain the circumstances in which each may be appropriate.
2. Explain how warehouse ownership and outsourcing affect cost, control, flexibility and risk.
3. Describe a rigorous process for selecting a warehouse location and explain the role of quantitative and qualitative methods.
4. Discuss how inventory pooling, demand correlation and service requirements influence the number of warehouses.
5. Explain how resilience and sustainability should be incorporated into warehouse network design.

Analytical / Application Questions

1. A company is considering one national warehouse versus three regional warehouses. Design a decision model that includes facility cost, transport, inventory, service constraints and disruption scenarios. State the data and assumptions required.
2. Using the centre-of-gravity formulas, calculate the screening location for four demand zones of your choice and critically assess why the calculated point may not be a feasible final site.
3. Design a hybrid network for a business with fast-moving, slow-moving, temperature-sensitive and customised products. Justify the location, ownership, capacity and fulfilment role assigned to each product group.

Multiple Choice Questions with Answers

Question 1

Which major advantage is most directly associated with centralising inventory?

A. Lower average delivery distance to every customer

B. Risk pooling and reduced duplication of safety stock

C. Elimination of all disruption risk

D. Guaranteed higher warehouse utilisation

Correct Answer: B. Risk pooling and reduced duplication of safety stock

Question 2

In the centre-of-gravity method, the weighting variable may represent:

A. Only building height

B. Annual shipment volume or transport demand

C. The number of warehouse employees at a candidate site

D. The legal age of the facility

Correct Answer: B. Annual shipment volume or transport demand

Question 3

If two independent and otherwise identical demand streams are pooled, their aggregate standard deviation is proportional to:

A. 2σ

B. σ/2

C. σ√2

D. σ²

Correct Answer: C. σ√2

Question 4

Which condition most supports cross-docking?

A. Highly uncertain demand and unreliable inbound schedules

B. Accurate information and synchronised inbound and outbound flows

C. Large quantities of obsolete inventory

D. A requirement for long-term storage of every product

Correct Answer: B. Accurate information and synchronised inbound and outbound flows

Question 5

In a capacitated facility-location model, the binary variable y_j commonly indicates whether:

A. Customer i places an order

B. Demand is normally distributed

C. Candidate facility j is opened

D. A shipment is damaged

Correct Answer: C. Candidate facility j is opened

Question 6

Which is a potential disadvantage of outsourcing warehouse operations?

A. Access to specialist scale

B. Variable capacity options

C. Loss of direct control and provider dependency

D. Reduced need for internal warehouse expertise

Correct Answer: C. Loss of direct control and provider dependency

Question 7

A warehouse with sufficient floor area but inadequate dock doors demonstrates that capacity is:

A. Only a question of storage volume

B. Multidimensional and operationally constrained

C. Independent of throughput

D. Always fixed by building rent

Correct Answer: B. Multidimensional and operationally constrained

Question 8

Expected disruption cost is calculated by:

A. Adding all scenario losses without probabilities

B. Multiplying the largest loss by the smallest probability

C. Summing probability-weighted scenario losses

D. Dividing facility cost by inventory value

Correct Answer: C. Summing probability-weighted scenario losses

Question 9

Postponement primarily seeks to:

A. Increase finished-goods variety at every warehouse

B. Delay final differentiation until demand information is available

C. Eliminate the need for customer data

D. Replace all transport with storage

Correct Answer: B. Delay final differentiation until demand information is available

Question 10

Which governance provision is especially important when using a 3PL for critical products?

A. A prohibition on measuring performance

B. No audit or exit rights

C. Defined service levels and business-continuity requirements

D. Unlimited unilateral price changes

Correct Answer: C. Defined service levels and business-continuity requirements