Global Logistics, Inventory and Supply Network Design

Introduction

International sourcing converts a commercial requirement into a cross-border flow of materials, information, documents, money and risk. The physical movement of goods is only one part of the problem. A sourcing decision also determines inventory exposure, customs obligations, lead-time variability, packaging requirements, insurance needs, carbon emissions, working capital and the resilience of the wider supply network.

Global logistics therefore concerns the integrated design and control of transportation, storage, inventory, packaging, trade documentation, intermediaries and logistics service providers. Its central managerial challenge is to provide the required service level at an acceptable total cost while remaining compliant, resilient and responsible.

A low unit purchase price may be economically unattractive if it requires large inventories, expensive air freight, frequent quality failures or costly emergency shipments. Conversely, a higher-cost supplier located closer to the market may reduce total landed cost and improve responsiveness. The relevant decision is not “Which supplier has the lowest price?” but “Which supply network creates the best risk-adjusted total value?”

Learning objectives

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

1. Select transport modes and routes using cost, service, risk and sustainability criteria.
2. Calculate and interpret landed cost, safety stock, reorder points, economic order quantity and inventory turns.
3. Explain the role of packaging, cargo insurance, freight documentation and logistics intermediaries.
4. Evaluate consolidation, postponement, cross-docking and multimodal transport.
5. Design and compare alternative supply-network configurations.
6. Establish performance measures and governance mechanisms for logistics service providers.
7. Critically assess resilience, digitalisation, sustainability and ethical issues in global logistics.

1. The global logistics system

A global supply network links suppliers, ports, carriers, customs authorities, warehouses, distribution centres, customers, financial institutions and information systems. It is a network rather than a simple linear chain because materials may flow through multiple suppliers, alternative ports, regional hubs and reverse-logistics channels.

The principal flows are:

• Material flow: raw materials, components, finished goods, returns and waste.
• Information flow: forecasts, purchase orders, shipment status, specifications, customs data and proof of delivery.
• Financial flow: invoices, freight charges, duties, insurance premiums, taxes and payments.
• Risk and compliance flow: certificates, licences, sanctions screening, security declarations and audit evidence.

The design objective is to balance the logistics triangle:

Optimising one dimension in isolation often damages another. Air freight may lower cycle time but increase cost and emissions. A large central warehouse may reduce facility cost but increase delivery distance and disruption concentration. A sourcing manager must therefore use total-cost and total-risk analysis.

2. Transport-mode and route selection

2.1 Principal transport modes

Ocean transport is economical for high-volume, non-urgent and relatively low-value cargo. It offers low unit cost but long and variable transit times, port dependence, container availability risk and exposure to congestion and weather.

Air transport provides speed, high frequency and global reach. It is suitable for urgent, high-value, compact or obsolescence-sensitive products. Its disadvantages include high freight cost, capacity constraints, security requirements and comparatively high emissions.

Road transport offers flexible door-to-door movement and is essential for first-mile and last-mile operations. It is affected by fuel prices, driver availability, border delays, road quality and regulatory restrictions.

Rail transport can provide a lower-emission alternative for land corridors and is often more reliable than road for long distances. Its limitations include fixed infrastructure, terminal handling and geographic dependence.

Inland waterways are cost-effective for heavy cargo on navigable routes but generally slower and less flexible.

Multimodal transport combines modes under coordinated planning. Intermodal transport normally means that the cargo remains in the same loading unit, such as a container, while modes change. This reduces handling and can improve security.

2.2 Selection criteria

Mode selection should consider:

• required delivery date and acceptable variability;
• cargo value, density, dimensions, perishability and hazardous characteristics;
• shipment size, frequency and load-factor potential;
• origin and destination infrastructure;
• customs and security requirements;
• damage, theft and temperature risk;
• fuel-price and capacity volatility;
• carbon emissions and customer requirements;
• total landed cost rather than quoted freight alone.

A useful decision matrix assigns weights to criteria and scores each option. For example, a buyer may assign 35% to cost, 30% to reliability, 20% to transit time and 15% to emissions. The weighted score is:

where S_j is the score of transport option j, w_i is the weight of criterion i, and r_ij is the normalised rating of option j on criterion i. Weights should sum to 1. The method supports comparison but does not eliminate managerial judgement; poor data or arbitrary scores can produce misleading results.

Route design should examine direct and indirect routes, port pairs, border crossings, trans-shipment points, seasonal weather, congestion, security exposure and contingency alternatives. The fastest nominal route is not necessarily the fastest reliable route.

2.3 Freight and landed cost

The landed cost of an imported item can be represented as:

where C_L is landed cost; C_P purchase cost; C_F freight; C_H handling and terminal charges; C_I insurance; C_D customs duty; C_T non-recoverable taxes; C_A administrative and documentation cost; and C_R expected risk or disruption cost. Units must be consistent, usually monetary units per shipment or per unit. Recoverable taxes should not be included as cost unless accounting policy treats them as non-recoverable.

Expected risk cost can be estimated as:

where p_k is the probability of risk event k and L_k is its financial loss. This is an expected-value approximation; it does not capture risk appetite, catastrophic consequences or correlation between events.

Illustrative calculation. A shipment has purchase cost $48,000, freight $3,200, handling $700, insurance $180, duty $2,400 and non-recoverable tax $1,100. Administrative cost is $320 and expected disruption cost is $600. Therefore:

C_L = 48,000 + 3,200 + 700 + 180 + 2,400 + 1,100 + 320 + 600 = $56,500

If the shipment contains 1,000 units, landed cost is $56.50 per unit. The calculation is meaningful only when all cost components use the same currency, shipment scope and tax treatment.

3. Trade terms, documentation and intermediaries

International commercial terms allocate responsibilities for delivery, risk transfer, transport arrangements and selected costs between buyer and seller. They do not by themselves determine ownership, payment, product conformity or all customs obligations. Organisations must use the current contractual rules and specify the named place precisely.

Important documents may include the commercial invoice, packing list, transport document, certificate of origin, export licence, import entry, insurance certificate, inspection certificate, dangerous-goods declaration and proof of delivery. Errors in product classification, valuation, origin or quantity can cause delays, penalties and incorrect duty payments.

Intermediaries include customs brokers, freight forwarders, non-vessel-operating common carriers, consolidators, third-party logistics providers and fourth-party logistics coordinators. Their roles differ:

Contracts should define service scope, liability, claims procedures, confidentiality, subcontracting, data access, compliance duties, insurance, audit rights, business continuity and termination. The buyer must distinguish the intermediary’s contractual liability from the carrier’s liability and from cargo insurance coverage.

4. Packaging, cargo security and insurance

Packaging is a logistics engineering decision, not merely a marketing activity. It must protect the product from compression, vibration, shock, moisture, temperature variation, corrosion, contamination and unauthorised access. It should also support efficient cube utilisation, safe handling, barcode scanning, palletisation and recycling.

Packaging design involves several levels:

• Primary packaging: directly contains or protects the product.
• Secondary packaging: groups primary packages.
• Tertiary packaging: supports transport and handling, such as pallets, crates and containers.

The designer should test the complete distribution environment rather than the factory environment alone. Packaging may need to withstand repeated handling, stacking, port exposure and long dwell times. Over-packaging increases material and disposal cost; under-packaging increases damage, returns and safety risk.

Cargo insurance should be selected after identifying the risk retained by the buyer or seller under the contract. Key issues include insured value, covered perils, exclusions, deductible, claims evidence, survey requirements, concealed damage, general average and time limits for notification. Insurance transfers selected financial consequences; it does not prevent delay, lost sales or reputational harm. A sound programme combines prevention, contractual allocation, physical security and insurance.

5. Inventory in international sourcing

Inventory performs several functions: cycle stock supports replenishment cycles, safety stock protects against uncertainty, pipeline stock is in transit, anticipation stock prepares for predictable peaks, and decoupling stock separates stages of a process. International sourcing often increases pipeline and safety stock because replenishment is slower and more variable.

5.1 Economic order quantity

Under the classical deterministic assumptions of constant demand, fixed ordering cost, constant unit price, instantaneous replenishment and no stockouts, the economic order quantity is:

where Q^* is order quantity in units, D annual demand in units, S ordering cost per order, and H annual holding cost per unit. If unit purchase price is C and the annual holding-rate is i, then H = iC, provided the holding rate is applied consistently.

Example. Annual demand is 24,000 units, ordering cost is $75 per order and annual holding cost is $3 per unit:

Q^* = √((2 × 24,000 × 75)/3) = √1,200,000 ≈ 1,095 units.

The result is a theoretical economic balance, not an automatic order quantity. Container capacity, supplier minimum order quantity, quantity discounts, shelf life and cash constraints may require adjustment.

5.2 Reorder point and safety stock

For continuous review with uncertain demand but approximately constant lead time:

where μ_L is expected demand during lead time and SS is safety stock. If demand per period has standard deviation σ_d, lead time is L periods, and demand periods are independent, then:

where z is the standard-normal service factor. This formula assumes independent, identically distributed demand and known, constant lead time. It is not appropriate without modification when lead time is variable, demand is intermittent, or demand and lead time are correlated.

When both demand and lead time vary independently, a commonly used approximation is:

and:

Here μ_d is mean demand per period, σ_L is lead-time standard deviation, and L is mean lead time. The first expression assumes demand and lead-time independence and measures standard deviation of demand during lead time. If demand and lead time are correlated, a covariance term is required.

Example. Mean daily demand is 100 units, daily standard deviation is 20 units, mean lead time is 9 days, and lead-time standard deviation is 2 days. For a 95% cycle-service target, use z ≈ 1.645.

σ_DL = √(9 × 20² + 100² × 2²) = √43600 ≈ 208.8 units.

Thus safety stock is 1.645 × 208.8 ≈ 344 units, and expected lead-time demand is 100 × 9 = 900 units. Therefore:

ROP ≈ 900 + 344 = 1,244 units.

The service factor must be matched to the chosen service definition. A cycle service level is the probability of no stockout during a replenishment cycle; it is not the same as fill rate, which measures the proportion of demand supplied immediately.

5.3 Inventory performance

Average inventory may include on-hand, safety, cycle, pipeline and sometimes consignment inventory. Inventory turnover is:

Days inventory outstanding, using a 365-day convention, is:

These measures are useful for comparison but can be distorted by seasonality, inflation, valuation method, stockouts and product mix. A high turnover ratio is not necessarily desirable if service levels are poor.

6. Lead-time management and consolidation

Total replenishment lead time can be decomposed into order processing, production, pre-carriage, export clearance, terminal dwell, main carriage, import clearance, inland delivery and receiving. Measuring each component helps identify waiting rather than movement time.

Lead-time reduction methods include supplier scheduling, electronic documentation, pre-clearance, standardised packaging, postponement, regional inventory, supplier development and reliable sailing or flight plans. Variability is often more damaging than average duration because it increases safety stock and emergency freight.

Consolidation combines shipments to improve vehicle or container utilisation. It may reduce freight cost per unit and emissions but can increase waiting time and handling. Consolidation is attractive when demand is compatible in destination, timing, product requirements and ownership.

Cross-docking transfers goods from inbound to outbound vehicles with minimal storage. It requires accurate advance information, synchronised schedules, stable demand and operational discipline.

Postponement delays final configuration, labelling or packaging until demand is clearer. It reduces the number of finished-goods variants held globally and can improve responsiveness, but it may require flexible production and regional finishing capability.

7. Supply-network design

Network design determines the number, location, role and capacity of suppliers, plants, ports, warehouses and fulfilment centres, together with allocation rules and transport links. Strategic decisions normally have long lives and high switching costs, so scenario analysis is essential.

7.1 Centralisation and decentralisation

A central distribution centre can pool demand variability, reduce duplicated safety stock and improve inventory visibility. Regional facilities can shorten customer lead time, reduce border exposure and provide market proximity. The trade-off depends on demand correlation, product value, delivery promise, transport rates, facility costs and disruption risk.

Risk pooling is strongest when regional demands are not perfectly correlated. However, centralisation also creates concentration risk: a single facility, port or supplier may become a critical failure point.

7.2 Facility-location model

A simplified capacitated facility-location model can be written as:

subject to:

Here F_j is fixed cost of opening facility j, y_j indicates whether it opens, x_ij is quantity shipped from facility j to customer i, c_ij is unit transport cost, h_ij represents other variable costs, d_i is customer demand and K_j is facility capacity. The model assumes known demand, costs and capacities and generally requires mixed-integer optimisation. Real decisions should add service constraints, taxes, duties, inventory, emissions, disruption scenarios and labour or regulatory restrictions.

7.3 Resilience and robustness

Efficiency seeks low expected cost; resilience seeks the ability to absorb, adapt to and recover from disruption. Design practices include dual sourcing, geographic diversification, qualified substitutes, strategic buffers, flexible contracts, visibility, alternate ports and documented recovery plans. Redundancy has a cost, so it should be targeted at critical products, nodes and lanes rather than applied uniformly.

A robust network can be evaluated through scenarios such as port closure, supplier failure, sanctions, cyberattack, extreme weather, demand surge and quality recall. Useful questions include: What is the time to recover? Which customers are prioritised? What substitute materials are approved? Which inventory is strategically located? Who has decision authority?

8. Logistics service-provider performance

A logistics provider should be managed through a balanced scorecard rather than freight price alone.

A KPI must have a precise definition, data source, owner, frequency, target and escalation rule. For example, on-time delivery should specify the agreed delivery window, treatment of customer-caused delay and whether partial delivery qualifies. Incentives should avoid encouraging undesirable behaviour, such as hiding exceptions or splitting shipments to improve a superficial metric.

Provider governance includes regular operational reviews, quarterly business reviews, root-cause analysis, corrective-action plans, audits and periodic benchmarking. Digital dashboards are useful only when underlying master data, event definitions and accountability are sound.

9. Digital technology, analytics and artificial intelligence

Transport-management systems can support rate management, tendering, route planning, carrier selection, shipment tracking and freight audit. Warehouse-management systems coordinate receiving, storage, picking, packing and dispatch. Control towers integrate events across providers and highlight exceptions.

Artificial intelligence can assist demand forecasting, ETA prediction, document extraction, anomaly detection, dynamic routing and inventory segmentation. Its value depends on data quality, process integration and human oversight. Forecast models should be tested against appropriate baselines and monitored for drift. Automated recommendations should not override regulatory controls, safety requirements or responsible procurement judgement.

Important governance principles include data ownership, cybersecurity, access control, explainability, human approval for high-impact decisions and protection of commercially sensitive information. A predictive system that uses biased historical data may systematically allocate scarce capacity away from smaller markets or suppliers. Technology is an enabler, not a substitute for sound network design.

10. Sustainability, ethics and regulatory responsibility

Global logistics creates emissions, packaging waste, congestion and labour risks. Organisations should measure transport emissions using a consistent methodology, improve load factors, reduce empty movement, select lower-carbon modes where service permits, use renewable or lower-carbon energy and design reusable or recyclable packaging. Carbon offsets should not replace direct reduction.

Ethical sourcing extends beyond the buying organisation. Forced labour, unsafe warehouses, excessive working hours, discrimination and corruption can occur at suppliers, carriers, subcontractors and informal intermediaries. Due diligence should include risk assessment, contractual expectations, worker-protection requirements, credible audits, grievance channels and remediation rather than merely terminating high-risk suppliers.

Regulatory responsibilities may include customs classification, valuation and origin, sanctions and export controls, dangerous-goods rules, product safety, data protection, environmental reporting and competition law. Compliance is a shared management responsibility; outsourcing customs or logistics activities does not eliminate the importer’s accountability.

Applied case: Northstar Medical Devices (illustrative)

Background

Northstar Medical Devices, a fictional manufacturer, imports sterile pump components from an overseas supplier. Annual demand is 36,000 units. The firm currently uses ocean freight through one port and holds inventory at a single national distribution centre. Average total lead time is 42 days, with a standard deviation of 10 days. Port congestion has caused stockouts and emergency air shipments.

Problem

The ocean option has a low freight rate but poor reliability. Air freight is reliable but would increase transport cost substantially. Management must decide whether to retain one route, use a dual-route policy, regionalise inventory or change the supplier allocation.

Analysis

The team first calculates total landed cost rather than comparing freight quotations. It identifies purchase cost, ocean and air freight, port charges, inventory carrying cost during pipeline time, quality inspection, insurance and the expected cost of stockouts. It then separates routine demand from critical hospital orders.

A scenario analysis compares:

1. Single ocean route: lowest routine freight cost, highest disruption exposure.
2. Ocean plus pre-approved air emergency lane: moderate expected cost and faster recovery.
3. Dual ports with two qualified carriers: higher coordination cost but lower concentration risk.
4. Regional postponement centre: additional facility cost but lower finished-goods variety and faster customer response.

The firm also discovers that safety stock was based on average lead time only and ignored lead-time variability. It revises the reorder-point policy, establishes a critical-item classification and negotiates carrier visibility and priority capacity.

Approach and outcome

Northstar retains ocean freight for normal replenishment, qualifies a second port, reserves limited emergency air capacity and establishes a regional postponement operation for selected configurations. Packaging is redesigned to improve cube utilisation and reduce damage. The logistics provider contract adds milestone visibility, port-diversion procedures, claims deadlines and emissions reporting.

The outcome is not the elimination of all risk. Rather, the firm accepts a modest increase in planned logistics cost in exchange for lower expected stockout exposure, improved recovery capability and clearer management of critical medical demand.

Learning points

• Freight price is only one component of total landed cost.
• Lead-time variability can justify safety stock and route diversification.
• Resilience should be targeted at critical products and bottlenecks.
• Postponement may improve service without duplicating all finished-goods inventory.
• Contracts, data quality and operating routines determine whether a network strategy works in practice.

Challenges and limitations

Global logistics models rely on forecasts, rates, probabilities and capacity assumptions that may change rapidly. Expected-cost analysis can understate low-probability catastrophic events. Multi-criteria scoring can conceal subjective preferences. Inventory formulas based on normal demand may fail under intermittent demand, promotions, shortages or correlated disruption. Facility-location models may produce mathematically efficient but operationally impractical designs if labour, customs, political and behavioural constraints are omitted.

Managers should therefore combine quantitative models with sensitivity analysis, scenario planning, expert review and periodic recalibration. Model outputs are decision support, not decision authority.

Best-practice implementation sequence

1. Map physical, information, financial and risk flows.
2. Define customer service requirements by product segment.
3. Measure actual lead-time components and variability.
4. Calculate total landed cost and working-capital implications.
5. Segment products by value, criticality, demand pattern and risk.
6. Compare transport modes, routes and consolidation options.
7. Evaluate network alternatives under normal and disruption scenarios.
8. Specify packaging, insurance, documentation and compliance controls.
9. Contract and govern providers using unambiguous KPIs.
10. Pilot changes, monitor outcomes and update the design continuously.

Chapter Summary

Global logistics and supply-network design integrate transportation, inventory, facilities, documentation, packaging, insurance, intermediaries and risk. Mode and route choices should be based on service requirements and total landed cost rather than freight price alone. Inventory decisions require clear assumptions about demand, lead time and service level; EOQ, reorder-point and safety-stock formulas are useful within their conditions but must be adapted to real-world variability.

Consolidation, cross-docking and postponement can reduce cost or improve responsiveness when supported by accurate information and coordinated operations. Network design balances centralisation, proximity, capacity, cost and resilience. Facility-location models provide analytical structure, while scenario analysis addresses disruption and uncertainty. Effective 3PL and 4PL governance requires precise KPIs, auditability, visibility and continuous improvement. Finally, sustainable and ethical logistics requires attention to emissions, packaging, labour conditions, customs, data governance and regulatory accountability.

Key Terms

Total landed cost: Complete cost of acquiring and delivering an imported item, including purchase, freight, duties, handling, insurance and relevant risk costs.

Pipeline inventory: Stock that has been ordered or dispatched but has not yet reached its intended location.

Safety stock: Inventory held to protect against demand or supply uncertainty.

Reorder point: Inventory position at which a replenishment order is triggered.

Economic order quantity: Order quantity that balances ordering and holding costs under specified assumptions.

Cycle stock: Inventory arising from ordering or production in batches.

Lead-time variability: Dispersion in the time required to replenish inventory.

Consolidation: Combining shipments to improve utilisation and reduce unit transport cost.

Cross-docking: Moving inbound goods to outbound dispatch with minimal storage.

Postponement: Delaying final configuration or differentiation until demand is clearer.

Intermodal transport: Movement using multiple modes while retaining the same loading unit.

Freight forwarder: Intermediary that arranges transport and associated logistics documentation.

Third-party logistics provider: External organisation performing logistics activities for a client.

Fourth-party logistics provider: Integrator managing multiple logistics providers and the broader logistics architecture.

Cargo insurance: Financial protection against specified insured losses affecting goods in transit.

Risk pooling: Reduction of aggregate uncertainty by combining demand or supply sources.

Network resilience: Ability of a supply network to absorb, adapt to and recover from disruption.

Cycle service level: Probability of completing a replenishment cycle without a stockout.

Fill rate: Proportion of demand supplied immediately from available inventory.

Inventory turnover: Annual cost of goods sold divided by average inventory at cost.

Review Questions

Short Answer Questions

1. Distinguish between material, information and financial flows in global logistics.
2. State four factors that influence transport-mode selection.
3. What is the difference between safety stock and pipeline inventory?
4. Explain the purpose of cargo insurance.
5. Distinguish between a 3PL and a 4PL.

Long Answer Questions

1. Explain how a sourcing manager should evaluate transport modes and routes using total landed cost, service and risk.
2. Derive and discuss the assumptions behind the EOQ and reorder-point models.
3. Compare centralised and decentralised distribution networks.
4. Discuss how packaging, documentation, insurance and intermediaries influence international logistics performance.
5. Explain a governance framework for managing logistics service providers.

Analytical / Application Questions

1. A company has unstable demand and highly variable international lead time. Design an inventory-policy approach, identify the data required and explain which formula assumptions may fail.
2. Compare a single-port network with a dual-port network for a critical imported component. Include cost, resilience, inventory and governance considerations.
3. Develop a weighted scorecard for selecting between ocean, rail and air transport for a high-value product, and justify the criteria weights.

Multiple Choice Questions with Answers

Question 1

Which cost is normally included in total landed cost?

A. Only the supplier’s purchase price

B. Purchase price, freight, duties, handling and relevant insurance

C. Only warehouse labour cost

D. Only the carrier’s quoted line-haul rate

Correct Answer: B. Purchase price, freight, duties, handling and relevant insurance

Question 2

Under the classical EOQ assumptions, increasing annual demand while all other variables remain constant will generally:

A. Decrease EOQ proportionally

B. Leave EOQ unchanged

C. Increase EOQ by the square root of the demand increase

D. Make EOQ equal to annual demand

Correct Answer: C. Increase EOQ by the square root of the demand increase

Question 3

A reorder point is principally intended to indicate:

A. The supplier’s selling price

B. When replenishment should be initiated

C. The maximum warehouse capacity

D. The annual transport budget

Correct Answer: B. When replenishment should be initiated

Question 4

Which transport mode is generally most suitable for urgent, high-value and compact cargo?

A. Air

B. Inland waterway

C. Ocean

D. Pipeline

Correct Answer: A. Air

Question 5

What is a principal benefit of postponement?

A. It eliminates all transport costs

B. It delays differentiation until demand information improves

C. It guarantees zero inventory

D. It removes the need for forecasting

Correct Answer: B. It delays differentiation until demand information improves

Question 6

If annual cost of goods sold is $2,000,000 and average inventory at cost is $500,000, inventory turnover is:

A. 2 turns

B. 4 turns

C. $1,500,000 turns

D. 25 turns

Correct Answer: B. 4 turns

Question 7

Which party primarily integrates and manages multiple logistics providers in a coordinated architecture?

A. 4PL

B. Customs authority

C. Packaging supplier

D. Final customer

Correct Answer: A. 4PL

Question 8

When both demand and lead time are variable, safety stock should generally consider:

A. Only purchase price

B. Only warehouse size

C. Demand variability and lead-time variability

D. Only the number of suppliers

Correct Answer: C. Demand variability and lead-time variability

Question 9

Which KPI most directly measures whether shipments arrive within the agreed delivery window and quantity?

A. On-time, in-full delivery

B. Inventory valuation method

C. Supplier advertising spend

D. Number of product colours

Correct Answer: A. On-time, in-full delivery

Question 10

A major limitation of a single-facility global network is:

A. It always has the highest number of facilities

B. It cannot use ocean transport

C. It may create concentration and single-point-of-failure risk

D. It eliminates inventory pooling

Correct Answer: C. It may create concentration and single-point-of-failure risk