13. Digital Supply Chains, Automation and Artificial Intelligence

Introduction and Context

Logistics and supply chain management have moved from transaction processing to digitally enabled coordination of networks. A contemporary supply chain may include manufacturers, suppliers, contract logistics providers, carriers, marketplaces, financial institutions, regulators and customers operating across several jurisdictions. The managerial challenge is not merely to move materials efficiently, but to create reliable visibility, respond rapidly to disruption, and make decisions using data generated across the entire network.

Digital supply chains use integrated information systems, connected devices, automation, advanced analytics and artificial intelligence to improve the flow of products, information and decisions. Their purpose is not technology adoption for its own sake. The purpose is to improve service, resilience, cost, asset utilisation, working capital, sustainability and decision quality. A highly automated warehouse with inaccurate inventory records, for example, may perform worse than a less automated facility with disciplined master data and sound processes.

This chapter examines the principal technologies supporting digital supply chains: enterprise resource planning (ERP), warehouse management systems (WMS), transportation management systems (TMS), the Internet of Things (IoT), radio-frequency identification (RFID), cloud platforms, digital twins, blockchain, robotics and artificial intelligence. It also evaluates data quality, cybersecurity, governance and responsible AI, because digital capability creates new dependencies and risks as well as new opportunities.

Learning Objectives

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

1. Explain the architecture and strategic role of major digital supply-chain systems.

2. Distinguish the applications and limitations of ERP, WMS, TMS, IoT, RFID, cloud platforms, digital twins, blockchain, robotics and AI.

3. Evaluate how digital technologies improve forecasting, planning, execution and end-to-end visibility.

4. Design an appropriate approach to data governance, cybersecurity and technology implementation.

5. Assess automation and AI initiatives using operational, financial, ethical and organisational criteria.

6. Apply these concepts to managerial decisions and supply-chain improvement cases.

From Functional Systems to Digital Supply Networks

Traditional information systems often reflected organisational boundaries. Purchasing, production, warehousing, transport and sales maintained separate records and exchanged information through periodic reports. This arrangement created delays, duplicated data entry and conflicting versions of operational reality. Digital supply chains seek a more connected flow of information, although complete integration is rarely achieved through a single system.

An ERP system normally provides the transactional backbone. It records orders, purchasing, inventory movements, production transactions, invoicing, financial postings and sometimes human-resource activities. Its strength is process integration and a common record of business transactions. Its weakness is that it may not provide the specialised, real-time control required for warehouse execution, transport optimisation or machine-level monitoring.

A WMS manages warehouse activities such as receiving, put-away, location control, replenishment, picking, packing, cycle counting and dispatch. It can assign work according to priority, location, skill and equipment. A TMS supports carrier selection, shipment planning, load consolidation, route design, freight audit, tendering and transport visibility. In a mature architecture, ERP, WMS and TMS exchange data through interfaces or application programming interfaces rather than operating as isolated applications.

The digital supply chain can therefore be understood as a layered architecture:

• The transaction layer records orders, inventory, purchase orders, production and financial events.
• The execution layer controls warehouse, transport, manufacturing and fulfilment activities.
• The connectivity layer captures data from sensors, vehicles, devices, suppliers and customers.
• The data and analytics layer stores, cleanses and analyses operational information.
• The decision and automation layer produces forecasts, recommendations, alerts or autonomous actions.
• The governance layer establishes security, access rights, quality standards, accountability and compliance.

This architecture highlights an important principle: digital transformation is a socio-technical change. Systems, processes, people, incentives and governance must be redesigned together.

Core Technologies and Their Supply-Chain Roles

ERP as the Transactional Backbone

ERP supports integrated planning and control by linking demand, procurement, production, inventory and finance. Material requirements planning uses demand, bills of material, inventory records and lead times to calculate requirements. In a broader planning process, sales and operations planning or integrated business planning aligns demand assumptions with capacity, supply and financial objectives.

ERP improves traceability and reduces duplicate entry, but implementation is difficult. Organisations must standardise processes, cleanse master data, define ownership and manage changes in roles. Excessive customisation can make upgrades expensive and preserve inefficient legacy practices. ERP should provide a stable core while specialised systems handle executional complexity.

WMS and the Digitally Controlled Warehouse

WMS converts warehouse strategy into executable work. It maintains stock by location and status, supports directed put-away and picking, and can coordinate labour and equipment. Advanced systems use wave planning, batch picking, task interleaving and slotting analysis. Integration with barcode scanners, conveyors, automated storage and retrieval systems, voice terminals and robots creates a more responsive fulfilment operation.

The value of a WMS depends on inventory accuracy and disciplined scanning. If goods are received without correct identification or moved without system confirmation, the WMS becomes an inaccurate representation of the warehouse. Thus, physical process discipline remains essential even in highly automated facilities.

TMS and Transport Decision-Making

TMS applications coordinate shipment execution across modes and carriers. They may compare rates, consolidate orders, select carriers, generate delivery schedules, monitor milestones and reconcile freight invoices. Optimisation engines can consider cost, capacity, service commitments, vehicle restrictions, delivery windows and carbon emissions.

TMS data can also support strategic decisions, such as carrier portfolio design, lane performance analysis and the evaluation of centralised versus regional distribution. However, optimisation is only as sound as its constraints and input data. A mathematically efficient route that ignores driver hours, loading limitations or customer receiving rules is not operationally feasible.

IoT, Sensors and RFID

The Internet of Things refers to connected physical objects that sense, communicate and sometimes act. In logistics, sensors may capture temperature, humidity, shock, location, vibration, door opening, fuel use or equipment condition. These data support condition monitoring, cold-chain compliance, predictive maintenance and exception management.

RFID identifies objects through radio signals and can read multiple tagged items without direct line of sight. It is useful for pallet, carton, returnable-container and asset tracking. Compared with manual barcode scanning, RFID may reduce handling and improve automatic capture, although tag cost, reader configuration, interference, standards and process design affect its business case.

IoT projects should begin with a decision that needs better information. Installing sensors without a response process may generate large volumes of data but little operational value. A temperature alert is valuable only when responsibility, escalation time and corrective action are clearly defined.

Cloud Platforms and Digital Integration

Cloud supply-chain platforms provide computing, storage and software through shared infrastructure or managed services. They can accelerate deployment, support collaboration and scale capacity according to demand. A supplier portal, for example, can provide shared access to forecasts, purchase orders, shipment status and quality documentation.

Cloud adoption does not eliminate integration or governance challenges. Organisations must evaluate data residency, service availability, exit arrangements, vendor concentration, identity management, interface standards and contractual responsibility for incidents. Multi-enterprise platforms may improve visibility, but participants may disagree about data ownership, commercial confidentiality and the cost of participation.

Digital Twins

A digital twin is a dynamic digital representation of a physical asset, process, facility or network that is updated using operational data. In supply chains, a twin may represent a warehouse, a distribution network, a production line or a set of transport flows. Managers can use it to test alternative inventory policies, facility layouts, capacity allocations or disruption scenarios before making physical changes.

A digital twin is more than a static simulation model. Its distinctive feature is the continuing relationship between the physical system and the digital representation. Nevertheless, the twin remains an approximation. If lead times, capacity, behavioural responses or demand uncertainty are poorly modelled, the apparent precision of the simulation can mislead decision-makers.

Blockchain and Shared Records

Blockchain is a distributed ledger technology in which transactions are recorded in a linked and tamper-evident sequence across participating nodes. It may support provenance, chain-of-custody records, smart-contract execution and multi-party verification where parties do not fully trust a central administrator.

Blockchain does not automatically prove that the original physical observation was truthful. A fraudulent or incorrect entry can still be recorded permanently. Its use is most defensible when several independent organisations need a shared record, reconciliation is costly, and the transactions have clear rules and identifiers. A conventional database may be cheaper and more controllable where one organisation already governs the process.

Robotics and Physical Automation

Robotics includes autonomous mobile robots, robotic arms, automated guided vehicles, drones and robotic process automation. In warehouses, robots can transport shelving, move pallets, sort parcels or assist picking. In administrative processes, robotic process automation can transfer data between systems or validate routine transactions.

Automation should be evaluated by total cost, throughput, flexibility, safety, maintenance, integration and labour redesign rather than by machine capability alone. Fixed automation may be attractive for stable high-volume flows, while flexible robots may be preferable when product variety and demand uncertainty are high. Human–robot collaboration requires clear safety zones, training, exception procedures and accountability.

Artificial Intelligence in Forecasting, Planning and Visibility

Artificial intelligence describes computational methods that perform tasks associated with learning, prediction, pattern recognition, language processing or decision support. Machine learning models identify relationships in data and use them to estimate future outcomes. In supply chains, common applications include demand forecasting, anomaly detection, ETA prediction, inventory optimisation, supplier-risk monitoring, warehouse slotting and customer-service assistance.

Forecasting and Demand Sensing

AI forecasting can incorporate sales history, promotions, price changes, weather, holidays, web activity, product relationships and external signals. It may detect nonlinear patterns that traditional methods do not capture. Demand sensing attempts to update short-term forecasts using recent orders, point-of-sale data or market signals.

AI does not remove uncertainty. Forecast accuracy may deteriorate when a product is newly launched, historical data reflect stock-outs, promotions are irregular, or market conditions change. A model trained on biased or incomplete history can reproduce that bias. Forecasts should therefore be evaluated by product segment and decision purpose, not by a single aggregate accuracy number. Human planners remain important for interpreting structural changes, commercial events and information not present in the data.

Planning and Optimisation

AI can recommend safety stocks, reorder points, production sequences, transport plans and allocation decisions. Prescriptive analytics combines forecasts with constraints and objectives to recommend actions. Reinforcement learning may be used in environments where an agent learns through repeated decisions, although governance and explainability are particularly important when actions affect customers, workers or critical operations.

A practical planning design separates prediction from policy. A model may predict demand, while a planning engine applies service-level targets, capacity limits, minimum order quantities and managerial priorities. This separation makes assumptions easier to examine and change.

Visibility and Exception Management

End-to-end visibility is not the same as displaying more data. It means providing timely, relevant and trustworthy information about a flow and enabling an appropriate response. Control towers aggregate orders, inventory, transport events, supplier information and risk signals. AI can identify unusual dwell times, probable late deliveries, inventory discrepancies or emerging supplier risks.

Effective exception management follows a sequence: detect, classify, prioritise, assign, act and learn. If every deviation generates an alert, users experience alert fatigue. Thresholds should reflect business impact, urgency and controllability. The system should record whether an alert was useful and whether the corrective action succeeded.

Data Quality, Governance and Interoperability

Data is the foundation of digital supply chains, but organisations often underestimate the effort required to make it usable. Master data includes product identifiers, units of measure, locations, suppliers, customers, bills of material, packaging dimensions and lead times. Transaction data records events such as receipts, issues, shipments and returns. Event data may come from sensors, scanners or external platforms.

Important dimensions of data quality include accuracy, completeness, consistency, timeliness, validity, uniqueness and traceability. A product dimension recorded in centimetres in one system and inches in another can distort freight calculations and warehouse capacity planning. A supplier lead time that is never updated can create systematic stock-outs even when the forecasting model is sound.

Data governance assigns ownership and establishes standards for creation, validation, access, retention, correction and deletion. A useful governance arrangement includes data owners who are accountable for meaning and quality, data stewards who maintain standards, and technology teams who manage platforms and controls. Data lineage should show where information originated, how it was transformed and where it was used.

Interoperability is equally important. Common identifiers, message standards, APIs and event schemas allow systems to exchange information. Integration should be designed around business events and decision requirements rather than merely connecting applications. Data should be captured once where possible and reused consistently.

Cybersecurity and Digital Resilience

Connected supply chains expand the attack surface. Threats include phishing, ransomware, credential theft, insecure devices, malicious software updates, denial-of-service attacks, insider misuse and third-party compromise. A disruption to a warehouse control system, transport platform or supplier portal can stop physical operations even when the goods and equipment remain intact.

A risk-based security programme commonly includes asset identification, network segmentation, strong authentication, least-privilege access, encryption, secure configuration, vulnerability management, monitoring, tested backups and incident-response plans. Operational technology requires special care because immediate patching may interrupt safety-critical operations. Segmentation can prevent a compromised office device from reaching warehouse controllers.

Third-party risk is central to supply-chain cybersecurity. Contracts should define security obligations, notification periods, access controls, audit rights, recovery expectations and responsibilities for data breaches. Business continuity planning should identify manual workarounds, alternative communication channels and recovery priorities. Resilience is not achieved by preventing every incident; it is achieved by limiting impact and restoring critical capability.

Responsible AI and Digital Governance

AI decisions can affect inventory availability, supplier evaluation, employee performance, delivery promises and customer treatment. Responsible AI therefore requires more than technical accuracy. Core principles include fairness, transparency, explainability appropriate to the decision, privacy, security, human oversight, accountability and contestability.

A responsible AI lifecycle includes defining the legitimate purpose, assessing risks, examining data provenance, testing performance across relevant groups and conditions, validating the model before deployment, monitoring drift and documenting changes. High-impact decisions should have an identifiable accountable manager and a route for review or appeal. Sensitive personal data should be minimised and protected according to applicable law and organisational policy.

Managers should distinguish correlation from causation. If a model predicts supplier delays from a variable associated with a particular region or firm size, using that prediction to impose penalties may create unfair outcomes. Explainability does not require revealing every mathematical detail to every user; it requires providing meaningful reasons, limitations and action implications to those affected.

Implementation: From Business Problem to Scaled Capability

A sound digital initiative begins with a business problem and a measurable outcome. Examples include reducing order-cycle time, increasing inventory accuracy, lowering empty transport kilometres or improving cold-chain compliance. The organisation should establish a baseline, identify process owners, map current data flows and assess readiness before selecting technology.

A useful implementation sequence is:

1. Define the decision or process to be improved.

2. Establish baseline performance and economic value.

3. Assess process maturity, data quality, integration and cyber risk.

4. Select a limited pilot with measurable success criteria.

5. Redesign roles, controls and exception procedures.

6. Test technically and operationally, including failure scenarios.

7. Train users and communicate the purpose of change.

8. Deploy in stages, monitor benefits and correct defects.

9. Scale only after evidence supports wider adoption.

Investment appraisal should include software, devices, integration, training, process redesign, maintenance, cybersecurity, downtime and change-management costs. Benefits may include revenue protection, service improvement, risk reduction and employee safety, not only labour savings. Post-implementation reviews should compare realised benefits with the original business case.

Illustrative Case: Northstar Components’ Connected Distribution Network

Background

Northstar Components is an illustrative mid-sized manufacturer supplying industrial spare parts through two distribution centres. It used an ERP system for orders and purchasing, but warehouse records were updated in batches. Transport was arranged through email and spreadsheets. Management frequently expedited urgent orders, while customers complained about uncertain delivery dates.

Situation and Problem

A review identified three related problems. Inventory accuracy was poor because receipts and internal movements were not consistently scanned. Transport planners lacked a consolidated view of shipments and carrier capacity. Forecasts were generated monthly using historical sales and did not adequately reflect service parts with intermittent demand. Senior managers proposed a fully autonomous warehouse and a blockchain-based supplier network, but the operations team lacked basic data discipline.

Analysis

The project team first mapped the order-to-delivery process and measured baseline performance. It found that the largest immediate loss arose from incorrect inventory status and delayed shipment confirmation, not from a lack of advanced algorithms. Product dimensions and units of measure were inconsistent across systems. Several suppliers also used different item identifiers.

The team classified initiatives according to value and readiness. A WMS integrated with ERP was judged essential for location control and scanning. A TMS was selected for carrier tendering, shipment consolidation and milestone visibility. RFID was recommended only for high-value returnable containers, where manual reconciliation was expensive. AI forecasting was proposed for selected product families after data cleansing, not as an immediate replacement for planner judgement.

Approach and Solution

Northstar appointed process owners for product, supplier and location data. It standardised units of measure, created cross-reference tables for supplier identifiers and introduced cycle-count rules based on item criticality. The WMS pilot covered one zone and required scan confirmation at receiving, put-away, picking and dispatch.

The TMS pilot integrated carrier status messages and established exception categories for late pickup, missed milestone and capacity shortfall. A control-tower dashboard displayed only exceptions exceeding defined service or financial thresholds. A forecasting model was tested on a selected product group, with planners reviewing unusual outputs and documenting overrides.

Cybersecurity controls included role-based access, multifactor authentication for remote users, network separation between office and warehouse systems, and a recovery exercise involving manual dispatch procedures. Employees were trained not only to use the systems but also to understand why transaction accuracy mattered.

Outcome

The illustrative case does not assume a particular numerical result. Its principal outcome was a shift from technology-led purchasing to capability-led improvement. Northstar gained a trusted inventory record, more consistent shipment information and a controlled basis for evaluating AI. The project team also discovered that several proposed technologies were premature until data standards and operating procedures had improved.

Learning Points

The case demonstrates five lessons. First, integration creates value only when processes and data are reliable. Second, visibility should lead to prioritised action rather than uncontrolled alerts. Third, pilots should test operational behaviour and failure recovery, not merely software functionality. Fourth, AI should be introduced where data and decision ownership are sufficiently mature. Finally, cybersecurity and workforce adoption must be designed into the initiative from the beginning.

Challenges, Limitations and Common Mistakes

Digital supply-chain programmes commonly fail through fragmented ownership. A procurement department may optimise supplier cost, a warehouse may maximise local throughput and transport may minimise freight expense, while total customer service deteriorates. Governance must therefore align local metrics with network outcomes.

Another mistake is confusing digitisation with transformation. Replacing paper forms with electronic forms may improve speed but does not necessarily redesign the process. Similarly, a dashboard with many indicators may create the appearance of visibility without improving decisions.

Other limitations include vendor lock-in, integration complexity, uncertain returns, employee resistance, model drift, poor external data, unreliable connectivity and the difficulty of scaling a successful pilot. Automation may also reduce flexibility if demand, product variety or process conditions change. Ethical concerns arise when surveillance technologies monitor workers excessively or when algorithmic recommendations are treated as neutral despite biased data.

Managers should resist three forms of overconfidence: believing that more data automatically produces better decisions, believing that AI forecasts are objective, and believing that automation removes the need for human expertise. In complex supply networks, expert judgement remains essential for interpreting uncertainty, negotiating relationships and responding to unprecedented events.

Emerging Directions and Best Practices

Emerging digital supply chains are likely to combine edge computing, generative AI, autonomous agents, computer vision, low-carbon optimisation and increasingly interoperable data spaces. Edge computing processes data close to the device, which can reduce latency in equipment monitoring. Generative AI may assist planners by summarising disruptions, explaining forecast changes or drafting responses, but its outputs require verification because fluent language does not guarantee factual accuracy.

Best practice is to maintain a balanced portfolio. Use stable transactional systems for control, specialised applications for execution, analytics for insight and automation where rules and exceptions are understood. Establish data ownership before scaling connectivity. Treat cybersecurity as an operational requirement rather than an IT-only concern. Measure service, cost, resilience, sustainability, safety and workforce outcomes together. Maintain human oversight for high-impact decisions and regularly review whether technology is producing the intended business and social results.

Chapter Summary

Digital supply chains connect transactions, physical execution, sensing, analytics and decision-making. ERP provides the transactional backbone; WMS and TMS manage warehouse and transport execution; IoT and RFID capture physical events; cloud platforms support scalable collaboration; digital twins enable scenario analysis; blockchain can provide shared, tamper-evident records in suitable multi-party settings; robotics automates physical and administrative work; and AI supports forecasting, optimisation, anomaly detection and visibility.

Technology does not compensate for weak processes or poor data. Data quality, master-data governance, interoperability and accountable ownership are foundations of digital performance. Cybersecurity, third-party controls, resilience planning and responsible AI are essential because connected systems create operational, ethical and regulatory exposure. Successful implementation starts with a defined business problem, measurable outcomes, phased pilots, workforce participation and continuous evaluation. The most effective digital supply chain is not the one with the greatest number of technologies, but the one that improves decisions and execution reliably across the network.

Key Terms

Artificial Intelligence: Computational capability that performs tasks involving learning, prediction, reasoning, language or pattern recognition.

Blockchain: A distributed, linked and tamper-evident ledger maintained across participating nodes.

Cloud Platform: Internet-delivered computing, storage or software services that support scalable access and collaboration.

Control Tower: A visibility and decision-support environment that aggregates supply-chain data and prioritises exceptions.

Data Governance: Policies, roles and controls for data ownership, quality, access, use, retention and correction.

Digital Twin: A dynamic digital representation of a physical asset, process or network updated with operational data.

ERP: An integrated enterprise system supporting transactions such as purchasing, production, inventory and finance.

Exception Management: The process of detecting, prioritising, assigning and resolving deviations from expected performance.

Internet of Things: A network of connected physical objects that sense, communicate or act.

Master Data: Relatively stable reference information such as products, locations, suppliers, units and customers.

Model Drift: Decline in model performance when operating conditions or data relationships change.

Prescriptive Analytics: Analytics that recommends actions by combining predictions, objectives and constraints.

RFID: Radio-frequency identification technology that captures the identity of tagged objects through radio signals.

Responsible AI: Design and use of AI that addresses fairness, transparency, privacy, security, accountability and human oversight.

Supply-Chain Visibility: Timely access to reliable information about flows, status, risks and events across a network.

TMS: A transportation management system supporting shipment planning, carrier management, routing and freight control.

WMS: A warehouse management system controlling inventory locations and warehouse execution activities.

Zero-Trust Security: A security approach that continuously verifies users, devices and permissions rather than trusting network location.

Review Questions

Short Answer Questions

1. What is the role of ERP in a digital supply-chain architecture?

2. Distinguish between RFID and IoT.

3. Why is data lineage important in supply-chain analytics?

4. What is meant by exception management?

5. State two principles of responsible AI.

Long Answer Questions

1. Explain how ERP, WMS and TMS can be integrated to support end-to-end execution.

2. Discuss the contribution and limitations of AI in demand forecasting and inventory planning.

3. Explain the principal cybersecurity risks of connected supply chains and the controls used to address them.

4. Evaluate the business conditions under which blockchain is preferable to a conventional database.

5. Develop an implementation framework for introducing warehouse automation in a mid-sized organisation.

Analytical / Application Questions

1. A company has frequent stock-outs despite purchasing a sophisticated AI forecasting platform. Diagnose the likely process and data causes and recommend a sequence of corrective actions.

2. A logistics provider wants to install sensors on every shipment. Design a decision framework for determining which shipments should be instrumented and how alerts should be managed.

3. A supplier-risk model consistently assigns higher risk to small regional suppliers. Explain how you would investigate possible bias, validate the model and govern its use in procurement decisions.

22. Multiple Choice Questions with Answers

Question 1

Which system primarily serves as the transactional backbone linking purchasing, inventory, production and finance?

A. ERP

B. RFID

C. Digital twin

D. TMS

Correct Answer: A. ERP

Question 2

A warehouse introduces scanners, but staff move goods without confirming transactions. What is the most likely consequence?

A. The WMS will automatically correct all records

B. The WMS will become an inaccurate representation of physical inventory

C. RFID will replace the need for process discipline

D. Forecast accuracy will improve immediately

Correct Answer: B. The WMS will become an inaccurate representation of physical inventory

Question 3

Which situation provides the strongest justification for using a digital twin?

A. Recording employee attendance

B. Testing alternative distribution-network designs before physical changes

C. Replacing all supplier contracts with smart contracts

D. Printing product labels

Correct Answer: B. Testing alternative distribution-network designs before physical changes

Question 4

Why can a blockchain record still contain false information?

A. Blockchain cannot store transactions permanently

B. A tamper-evident ledger does not guarantee that the original physical input was truthful

C. Blockchain automatically changes data after verification

D. Only one participant can access a blockchain

Correct Answer: B. A tamper-evident ledger does not guarantee that the original physical input was truthful

Question 5

Which practice best supports responsible AI in supplier evaluation?

A. Allowing the model to make unreviewable decisions

B. Using historical data without checking its bias or provenance

C. Providing accountable human oversight and monitoring outcomes across relevant groups

D. Maximising the number of variables regardless of privacy concerns

Correct Answer: C. Providing accountable human oversight and monitoring outcomes across relevant groups

Question 6

A temperature sensor generates hundreds of alerts, most of which require no action. Which improvement is most appropriate?

A. Remove all sensors

B. Create impact-based thresholds and defined escalation procedures

C. Send every alert to every manager

D. Ignore alerts until a product is rejected

Correct Answer: B. Create impact-based thresholds and defined escalation procedures

Question 7

Which data-quality problem is most likely to distort warehouse capacity and freight calculations?

A. Different units of measure across systems

B. Using multifactor authentication

C. Having a documented recovery plan

D. Using role-based access

Correct Answer: A. Different units of measure across systems

Question 8

What is the main purpose of a TMS optimisation engine?

A. To design product bills of material

B. To select and plan shipments subject to cost, capacity and service constraints

C. To identify tagged objects through radio signals

D. To train warehouse employees

Correct Answer: B. To select and plan shipments subject to cost, capacity and service constraints

Question 9

Which implementation approach is most consistent with the chapter’s recommendations?

A. Deploy every available technology simultaneously

B. Begin with a defined business problem, baseline performance and a measurable pilot

C. Postpone data governance until after scaling

D. Evaluate success only through labour reduction

Correct Answer: B. Begin with a defined business problem, baseline performance and a measurable pilot

Question 10

Why should cybersecurity be treated as an operational supply-chain issue?

A. Cyber incidents affect only financial reporting

B. A compromised digital system can interrupt physical warehouse, transport or production operations

C. Cloud platforms eliminate all cyber risk

D. Only customers are responsible for system security

Correct Answer: B. A compromised digital system can interrupt physical warehouse, transport or production operations