Order Picking and Fulfillment Design

Introduction

Order picking is the warehouse activity in which products are selected from storage to satisfy customer orders. It is often the most labor-intensive and operationally variable warehouse process. In many facilities, picking accounts for approximately one-half or more of operating labor, although the proportion varies by product profile, automation level, wage structure, and order complexity. Picking performance directly affects order cycle time, inventory accuracy, customer service, transportation consolidation, and operating cost.

Picking design is not simply a choice between manual labor and automation. It is a systems-design problem involving order characteristics, inventory policies, warehouse layout, information systems, labor planning, replenishment, release control, ergonomics, and quality assurance. A strategy that is effective for wholesale pallets may be unsuitable for e-commerce orders containing many small lines. Similarly, a highly automated solution may reduce travel but increase capital exposure and reduce flexibility.

This chapter examines discrete, batch, zone, wave, cluster, goods-to-person, and robotic picking. It also develops principles for pick-path design, forward-pick sizing, replenishment, order release, and error prevention across wholesale, retail, and e-commerce environments.

Learning Objectives

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

1. Explain the operational logic, advantages, and limitations of major picking methods.
2. Select a picking strategy for wholesale, retail, and e-commerce contexts.
3. Design efficient pick paths and forward-pick areas.
4. Establish replenishment triggers and order-release policies.
5. Evaluate goods-to-person and robotic picking technologies.
6. Design error-prevention and quality-control systems.
7. Assess picking designs using cost, service, capacity, ergonomic, and resilience criteria.

1. The Fulfillment System and Its Design Variables

A fulfillment system converts demand information into a verified shipment. Its principal stages are order capture, inventory allocation, release, picking, consolidation, packing, verification, dispatch, and exception management. These stages are interdependent. For example, releasing too many orders may overload packing even if pick productivity increases. Conversely, excessive batching may reduce travel but delay urgent orders.

Important design variables include:

• Order profile: orders per day, lines per order, units per line, cube, weight, and urgency.
• SKU profile: demand frequency, product dimensions, unit value, handling requirements, and seasonality.
• Storage structure: pallet reserve, case pick, each pick, shelving, flow rack, automated storage, or mixed systems.
• Service promise: same-day, next-day, scheduled delivery, store replenishment, or wholesale appointment windows.
• Labor conditions: skill availability, wage rates, shift length, turnover, and ergonomic constraints.
• Information quality: inventory accuracy, location accuracy, barcode standards, and system integration.
• Variability: promotions, peak seasons, late orders, stockouts, and carrier cutoffs.

A useful conceptual distinction is between travel time, search and decision time, handling time, and waiting time. Technology often reduces one component without eliminating the others. A robot may remove walking but not the time required to identify, verify, and place items into an order container.

2. Picking Methods and Their Appropriate Uses

2.1 Discrete Picking

In discrete, or single-order, picking, one picker completes one order before beginning another. The worker receives an order or task list, visits the required locations, collects the items, and transfers the completed order to packing or dispatch.

Discrete picking is easy to understand, supports strong order identity, and is appropriate when orders are large, urgent, highly customized, or operationally complex. It is common in wholesale case picking, industrial distribution, spare-parts operations, and low-volume environments.

Its principal weakness is duplicated travel. If ten orders require the same high-demand SKU, a picker may visit that location ten times. Discrete picking can nevertheless be efficient when orders have many lines, order integrity is critical, or consolidation would be difficult.

2.2 Batch Picking

Batch picking combines several orders into one picking tour. The picker collects the total quantity required for all orders and later sorts the items into individual orders. If a batch contains orders i = 1, 2, ..., n, the required quantity of SKU j is:

where q_ij is the quantity of SKU j in order i and Q_j is the total quantity picked for the batch. The formula assumes that all orders are compatible with the same picking area and that the system can maintain accurate order-to-item association.

Batch picking reduces repeated travel and is particularly effective for many small orders with overlapping SKU requirements. It creates, however, a secondary sortation or allocation task. The design question is therefore not whether travel is reduced, but whether the travel savings exceed the additional sorting, verification, and error-control effort.

Batch size should reflect order similarity, cart capacity, picker capacity, cutoff times, and downstream sortation capacity. Oversized batches can create congestion, sorting backlogs, and late orders.

2.3 Zone Picking

In zone picking, the warehouse is divided into zones and each picker works within a designated zone. Orders may move from zone to zone, or picked work may be consolidated after zone completion.

Zone picking reduces walking distance and allows workers to develop location familiarity. It is suitable for large facilities, extensive SKU ranges, and areas with distinct storage or handling requirements. It can also support specialization, such as chilled, hazardous, high-value, or bulky-product zones.

Two principal forms are:

• Progressive zone picking: the order container moves sequentially through zones.
• Parallel zone picking: different zones pick simultaneously and the order is later consolidated.

Progressive systems may suffer from waiting at zone boundaries. Parallel systems may require sophisticated synchronization and consolidation. Zone balancing is essential: a slow or overloaded zone becomes a system bottleneck.

2.4 Wave Picking

Wave picking releases work during defined time windows. A wave may be created by carrier cutoff, customer service level, route, temperature requirement, or warehouse area. Orders in a wave are then processed using discrete, batch, zone, or hybrid methods.

Waves create operational rhythm and facilitate labor scheduling, dock coordination, and transportation planning. They are especially useful when shipment deadlines are meaningful. Their limitation is temporal rigidity. A wave may release work before inventory is available, or it may create bursts of congestion followed by idle periods.

Wave planning should therefore consider downstream capacity. A wave is successful only when picking, packing, staging, and loading can process its volume within the available time window.

2.5 Cluster Picking

Cluster picking uses a cart, trolley, or workstation with multiple totes or compartments. A picker visits a location once and distributes units directly into the correct order container. It combines the travel benefits of batching with immediate order separation.

Cluster picking is common in e-commerce and retail replenishment where orders contain relatively few units and many orders share popular SKUs. It requires accurate location identification, a reliable compartment assignment, and adequate cart maneuverability. The number of compartments should not exceed what the picker can manage accurately and ergonomically.

2.6 Goods-to-Person Picking

In goods-to-person systems, storage equipment or mobile robots deliver inventory to a stationary picker. Examples include shuttle systems, carousel systems, vertical lift modules, and autonomous mobile robots that transport shelves or pods.

The main benefits are reduced walking, improved labor productivity, better space utilization in some configurations, and potentially improved ergonomics. The constraints include capital cost, software integration, system availability, replenishment dependency, and the need to design workstations carefully. Goods-to-person is most attractive where order volume, SKU density, labor cost, or service requirements justify the investment.

2.7 Robotic Picking

Robotic picking may refer to robotic movement of inventory, robotic item grasping, or both. Robotic mobile units commonly transport shelves, totes, or pallets. Robotic arms may pick individual items using vision, machine learning, and force control.

Robotic item picking is more difficult than robotic transport because products vary in shape, packaging, reflectivity, fragility, and graspability. Human intervention remains necessary for exceptions and poorly presented inventory. A realistic evaluation should include uptime, maintenance, integration, changeover, safety, exception handling, and total cost of ownership rather than advertised pick rates alone.

2.8 Comparative Evaluation

Methods are not mutually exclusive. A facility may use wave release, zone organization, batch formation, cluster carts, and goods-to-person equipment simultaneously.

3. Pick-Path Design

Pick-path design determines how workers or equipment travel through the storage system. The objective is not always to minimize distance alone. A high-quality path also reduces congestion, crossing traffic, unsafe movements, search time, and unnecessary handling.

Common routing policies include:

• Return routing: enter an aisle, pick required items, and return to the same cross-aisle.
• S-shape routing: traverse every aisle containing a pick, entering from one end and exiting from the other.
• Largest-gap routing: enter aisles and turn back at a calculated gap to avoid unnecessary traversal.
• Combined routing: use different entry and exit decisions for different aisles.

The best method depends on pick density, aisle geometry, cross-aisles, one-way restrictions, congestion, and order distribution. A routing algorithm should account for actual travel time rather than geometric distance alone when congestion or lifting constraints are significant.

A practical layout normally places high-frequency SKUs near the start or most accessible portions of a route, while also considering product affinity. Product affinity means that items frequently ordered together should be located near one another, provided that doing so does not create congestion or violate compatibility rules.

A basic activity-based slotting principle is to place an SKU according to its demand frequency, handling burden, and cube. A fast-moving, small, high-frequency item usually belongs in an easily reached forward location. A slow-moving, bulky item may be better placed in a reserve or peripheral location if travel and handling remain acceptable.

4. Forward-Pick Areas and Slotting

The forward-pick area is the accessible inventory location from which routine orders are picked. Reserve storage replenishes the forward area. The forward area should be sized to balance picker travel, replenishment frequency, space, and inventory exposure.

For a SKU, forward capacity must cover expected demand between replenishment opportunities while allowing for variability. A simple cycle-stock estimate is:

where C_f is required forward capacity in units, d is average demand per unit of time, T is the replenishment interval in the same time unit, and s is a safety allowance for demand and execution variability. The equation assumes relatively stable demand and a replenishment policy based on a fixed review interval. It is not a substitute for a probabilistic inventory model when demand or lead time is highly variable.

If demand during the replenishment interval is uncertain, a common safety-stock expression is:

where SS is safety stock, z is the standard-normal service factor, and σ_DL is the standard deviation of demand during replenishment lead time. The formula is appropriate only when the demand distribution and independence assumptions are sufficiently reasonable. If demand is measured per period and lead time is L periods, with independent identical demand having standard deviation σ_d, then:

The units of σ_DL are units of product, ensuring that safety stock is also measured in units.

Slotting should consider:

1. Demand velocity and variability.
2. Product dimensions, weight, and ergonomics.
3. Order affinity and family grouping.
4. Replenishment effort and reserve location.
5. Security, temperature, hazardous-material, and compatibility requirements.
6. Scanner visibility and label orientation.
7. Seasonal and promotional changes.

A high-velocity item should not be placed in a location that creates unsafe congestion. Similarly, placing all frequently ordered products at the front may increase interference between pickers. Slotting is therefore a network problem involving travel, capacity, safety, and congestion.

5. Replenishment Rules

Picking fails when forward locations are empty even though inventory exists elsewhere. Replenishment must be integrated with picking rather than treated as an independent warehouse activity.

Common policies include:

• Minimum–maximum: replenish when inventory reaches a minimum; restore it to a maximum.
• Top-off: replenish forward locations during low activity or before a wave.
• Demand-triggered: create a task when allocated demand exceeds available forward stock.
• Scheduled replenishment: replenish at fixed times, often before major waves.
• Emergency replenishment: respond to an immediate stockout or imminent order failure.

A minimum–maximum rule can be represented as:

where R is replenishment quantity, S is the target forward level, I_f is usable inventory currently in the forward location, and O_f is inventory already assigned to open forward replenishment tasks. The maximum with zero prevents a negative task. This rule assumes inventory records are accurate and that the target level is expressed in the same units as the inventory quantities.

Replenishment priority should reflect order urgency, expected stockout time, picker demand, and travel efficiency. A replenishment task that arrives after a picker has already bypassed the empty location does not protect service. Real-time systems can reserve stock, create replenishment tasks, and sequence them ahead of picking demand.

6. Order Release and Control

Order release determines when work becomes available to the warehouse. Releasing every order immediately can overload work-in-process, create congestion, and starve priority orders. Releasing too cautiously can produce idle labor and missed carrier cutoffs.

A release policy should define:

• eligible orders and their priorities;
• inventory-availability rules;
• wave or continuous-release timing;
• batch formation criteria;
• capacity limits for picking, packing, and staging;
• treatment of incomplete or exception orders;
• cutoff and escalation rules.

Useful priority dimensions include promised ship time, customer class, route departure, order age, product temperature, and value. A purely first-in-first-out policy may be fair but can be operationally inferior when orders have different deadlines. Conversely, excessive priority overrides can undermine fairness and produce unstable work.

A practical control principle is to release only the work that downstream processes can absorb. This is a form of work-in-process control. It reduces congestion and exposes bottlenecks earlier. Supervisors should monitor queue length, aging, completion rate, and the percentage of orders at risk of missing promise time.

7. Error Prevention and Quality Assurance

Picking errors include wrong SKU, wrong quantity, wrong lot or serial number, wrong location, damaged product, and missed item. Errors may originate in master data, receiving, slotting, inventory records, picker decisions, or system design.

Effective quality systems use multiple controls:

• barcode or RFID identification;
• scan confirmation at location and item level;
• visual and electronic location verification;
• weight checks at packing;
• pick-to-light or put-to-light indicators;
• image capture and audit trails;
• serial, lot, and expiry validation;
• exception workflows rather than informal overrides;
• cycle counting based on risk and discrepancy history.

The hierarchy of error prevention is important. Designing a process that makes the error difficult is stronger than merely detecting the error later. For example, a system that requires scanning both the location and the item can prevent a wrong-location pick. A final audit may detect the error, but only after labor and time have been spent.

Quality metrics should distinguish accuracy from completeness. If 9,900 of 10,000 order lines are correct, line accuracy is 99.0 percent. However, if one error causes each affected order to fail, order accuracy may be lower. Metrics must specify the denominator and unit of analysis.

Human factors are central. Labels should be legible, locations consistently named, screens uncluttered, and confirmation steps positioned at the point of decision. Incentive systems should never reward speed so strongly that workers bypass safety or verification controls.

8. Measuring Picking Performance

Important measures include:

• lines picked per labor hour;
• units picked per labor hour;
• orders completed per hour;
• travel distance per order or line;
• pick cycle time;
• inventory availability at pick face;
• order-line accuracy;
• order completeness;
• replenishment response time;
• labor utilization;
• equipment availability;
• cost per order or line;
• percentage of orders shipped on time.

Productivity must be interpreted with quality and service. A higher pick rate that causes more errors, injuries, or late packing may reduce total system performance. A balanced scorecard is preferable to a single productivity target.

Labor productivity can be expressed as:

where P is productivity in units per labor hour, N_u is the number of accepted units picked, and H is labor hours. The measure should define whether indirect time, breaks, training, replenishment, and rework are included. Comparisons are valid only when the measurement boundaries are consistent.

For economic evaluation, total relevant cost may include labor, equipment, software, space, maintenance, energy, quality failures, and transition costs. A technology with lower direct labor but higher capital and maintenance cost is not automatically superior.

9. Application to Different Operating Contexts

Wholesale Distribution

Wholesale operations often have larger orders, case or pallet handling, customer-specific requirements, and scheduled deliveries. Discrete or zone-based case picking may be appropriate. Wave release aligned with route departures can coordinate picking and loading. Pallet stability, lot control, and dock sequencing are often more important than minimizing each-unit travel.

Retail Store Replenishment

Retail distribution commonly processes repeated store orders, promotional surges, and route-specific loads. Batch and zone picking can exploit order similarity. Wave planning should reflect store delivery windows and vehicle loading sequence. Cases may be picked in a sequence that supports efficient store receiving, not merely warehouse convenience.

E-Commerce

E-commerce usually involves many orders, few units per order, high SKU breadth, and tight parcel cutoffs. Cluster picking, batch picking with put walls, goods-to-person systems, and hybrid automation are common. The design must accommodate returns, substitutions, gift orders, personalization, and rapid changes in demand. Order identity and scan-based verification are particularly important.

10. Illustrative Case: Northstar Home Distribution Center

Background

Northstar Home is a fictional omnichannel distributor of household products. Its 18,000-SKU facility serves retail stores and direct-to-consumer customers. Store orders are case-oriented and dispatched on scheduled routes; consumer orders average 2.6 lines and are shipped by parcel carriers.

Problem

The existing system used discrete picking for all orders. Pickers traveled long distances, consumer orders missed the parcel cutoff, and replenishment frequently interrupted picking. Line accuracy was 98.7 percent, while the target was 99.5 percent.

Analysis

Order data showed that 20 percent of SKUs generated 78 percent of consumer order lines. Consumer orders had substantial overlap, whereas store orders had larger quantities and different cutoff requirements. The operation also found that several high-volume forward locations held insufficient stock for a full wave, causing emergency replenishment.

Approach

Northstar introduced a segmented design:

1. Consumer orders were released in short waves every 30 minutes.
2. High-overlap consumer orders were batch-picked using carts with twelve clearly numbered totes.
3. A put wall separated orders after the batch, with scan confirmation at each placement.
4. Store case orders remained in zone-based wave picking organized by route departure.
5. High-velocity consumer SKUs were moved to ergonomic forward locations.
6. Forward minimum–maximum levels were recalculated using demand during the replenishment interval plus a safety allowance.
7. Replenishment tasks were prioritized before the next consumer wave.
8. Packing stations used item scans and parcel-weight tolerance checks.

Outcome

In the illustrative scenario, walking distance per consumer order declined, parcel cutoff performance improved, and line accuracy rose after the combination of batch discipline and scan verification. The most important improvement did not come from batching alone; it came from coordinating release, replenishment, put-wall capacity, and packing verification. Northstar retained discrete handling for unusual or urgent orders, demonstrating that hybrid designs are often more robust than a single universal method.

Learning Points

• Order profiles should determine process segmentation.
• Batch picking requires a controlled sortation or allocation method.
• Forward-pick capacity and replenishment policy must be designed together.
• A productivity improvement is meaningful only if service and accuracy are preserved.
• Hybrid systems can reduce the risk of forcing incompatible work into one process.

11. Technology, AI, and Contemporary Developments

Warehouse management systems increasingly use real-time inventory allocation, dynamic batching, congestion-aware routing, and labor dashboards. Machine-learning models may forecast order arrivals, identify likely stockouts, predict equipment failures, and recommend slotting changes. Computer vision can support item identification, dimension capture, and robotic grasping.

These tools should be treated as decision support rather than unquestionable authority. Models depend on accurate master data and may reproduce historical biases, such as consistently assigning difficult tasks to particular workers. Governance should include explainability appropriate to the decision, human override, data security, cybersecurity, and auditability.

Autonomous mobile robots and flexible robotic workstations are expanding because they can be deployed incrementally in some facilities. Nevertheless, automation resilience requires manual fallback procedures, spare capacity, maintenance skills, and recovery plans for network or system outages. A system that performs well only under ideal conditions may be less valuable than a slightly slower system with strong recovery capability.

12. Challenges, Trade-offs, and Best Practices

The central trade-off in picking design is between travel reduction, handling complexity, flexibility, capital intensity, and control effort. Batch picking reduces travel but increases sorting. Zone picking reduces walking but creates interfaces. Goods-to-person reduces walking but increases dependence on equipment and software. Robotic picking may improve consistency but remains sensitive to product variation.

Best practices include:

1. Segment orders by operational characteristics rather than customer label alone.
2. Use accurate, current SKU, location, dimension, and inventory data.
3. Slot using demand, affinity, ergonomics, replenishment, and congestion together.
4. Design replenishment before finalizing forward-pick capacity.
5. Release work according to downstream capacity and service deadlines.
6. Use prevention-oriented scanning and visual controls.
7. Measure quality, service, safety, and cost together with productivity.
8. Pilot changes using representative peak and non-peak conditions.
9. Maintain human fallback and exception-handling capability.
10. Review the design as demand, product mix, and labor conditions change.

Ethically, managers must consider worker safety, surveillance, workload intensity, accessibility, and fair performance evaluation. Regulatory requirements may apply to occupational safety, hazardous materials, food or pharmaceutical traceability, data protection, and automated equipment safeguarding. Compliance should be embedded in process design rather than added after implementation.

Chapter Summary

Order picking is a system-design problem linking demand, storage, information, labor, equipment, replenishment, release control, packing, and transportation. Discrete picking provides simplicity and order identity; batch picking reduces repeated travel; zone picking supports specialization; wave picking coordinates time and capacity; cluster picking combines batching with direct order separation; and goods-to-person and robotic systems reduce walking or automate repetitive work.

Effective design begins with order and SKU segmentation. Pick paths should address travel, congestion, affinity, ergonomics, and safety. Forward-pick areas require sufficient capacity for demand between replenishment opportunities, while replenishment rules must prevent stockouts without excessive handling. Order release should control work-in-process and protect service deadlines. Error prevention is strongest when systems make incorrect actions difficult through scanning, clear labeling, verification, and exception control.

No picking method is universally best. Wholesale, retail, and e-commerce operations require different combinations of methods. Hybrid designs, supported by reliable data and balanced performance metrics, generally provide the flexibility needed for variable fulfillment environments.

Key Terms

Order picking: Selecting products from storage to satisfy customer orders.

Discrete picking: Completing one order before beginning another.

Batch picking: Picking quantities for several orders during one tour.

Zone picking: Assigning pickers or equipment to defined warehouse areas.

Wave picking: Releasing groups of orders during planned time windows.

Cluster picking: Picking multiple orders into separate containers carried together.

Goods-to-person: Delivering inventory to a stationary picker or workstation.

Robotic picking: Using robotic equipment to move or select inventory.

Forward-pick area: Accessible stock location used for routine order picking.

Reserve storage: Inventory held away from the primary picking face.

Slotting: Assigning SKUs to storage locations.

Pick path: The travel route used to collect required items.

Product affinity: The tendency of products to be ordered together.

Replenishment: Moving inventory from reserve to a forward-pick location.

Order release: The decision and timing by which orders become warehouse work.

Put wall: A compartmented sortation structure used to separate batch-picked orders.

Line accuracy: The proportion of picked order lines completed correctly.

Exception management: Handling orders or tasks that cannot follow the normal process.

Work-in-process: Orders or tasks released but not yet completed.

Service level: The extent to which promised fulfillment performance is achieved.

Review Questions

Short Answer Questions

1. What is the principal operational difference between discrete and batch picking?
2. Why can zone picking create a bottleneck even when individual zones are productive?
3. What is the purpose of a forward-pick area?
4. Name two controls that prevent picking errors rather than merely detecting them.
5. Why should productivity be evaluated together with service and accuracy?

Long Answer Questions

1. Compare discrete, batch, zone, wave, and cluster picking in terms of travel, order identity, labor requirements, and typical applications.
2. Explain how forward-pick sizing and replenishment policy should be designed as an integrated system.
3. Discuss the strategic advantages and limitations of goods-to-person and robotic picking.
4. Develop an order-release policy for an e-commerce operation with parcel carrier cutoffs and highly variable demand.
5. Explain why a hybrid picking system may be more effective than a single method in an omnichannel distribution center.

Analytical / Application Questions

1. A facility serves both store replenishment orders and small consumer orders. Design a segmented picking architecture and justify the release, picking, replenishment, and verification policies.
2. A warehouse reports higher units per labor hour after introducing batching, but order accuracy and packing congestion have worsened. Diagnose the likely causes and propose corrective actions.
3. Using the forward-capacity relationship C_f = dT + s, explain how an operation should respond when demand variability increases but available forward space cannot be expanded.

Multiple Choice Questions with Answers

Question 1

Which picking method completes one customer order before beginning another?

A. Discrete picking

B. Batch picking

C. Zone picking

D. Cluster picking

Correct Answer: A. Discrete picking

Question 2

What is the main operational benefit of batch picking?

A. It eliminates the need for order verification

B. It reduces repeated travel to common SKU locations

C. It guarantees equal workload across zones

D. It removes the need for replenishment

Correct Answer: B. It reduces repeated travel to common SKU locations

Question 3

In a progressive zone-picking system, what generally happens to the order?

A. It remains permanently at one workstation

B. It moves sequentially through designated zones

C. It is always picked by a robot

D. It is combined with every order in the warehouse

Correct Answer: B. It moves sequentially through designated zones

Question 4

Which situation is most suitable for cluster picking?

A. A few urgent pallet orders with no SKU overlap

B. Many small orders with overlapping SKU demand

C. A warehouse handling only hazardous bulk liquids

D. A facility with no order-level identification system

Correct Answer: B. Many small orders with overlapping SKU demand

Question 5

In the expression C_f = dT + s, what does T represent?

A. The number of customer orders

B. The replenishment interval

C. The product unit cost

D. The number of storage locations

Correct Answer: B. The replenishment interval

Question 6

What is a principal advantage of goods-to-person picking?

A. It eliminates all inventory inaccuracies

B. It reduces picker walking by bringing inventory to a workstation

C. It requires no software integration

D. It is always less expensive than manual picking

Correct Answer: B. It reduces picker walking by bringing inventory to a workstation

Question 7

Which control is most directly designed to prevent a wrong-location pick?

A. End-of-shift productivity reporting

B. Scanning the storage location before picking

C. Increasing batch size

D. Delaying replenishment

Correct Answer: B. Scanning the storage location before picking

Question 8

Why can excessive batch sizes be harmful?

A. They always increase walking distance

B. They can overload sorting, consolidation, and downstream processes

C. They prevent the use of barcodes

D. They eliminate product affinity

Correct Answer: B. They can overload sorting, consolidation, and downstream processes

Question 9

A warehouse picks 12,000 accepted units in 600 labor hours. Using P = N_u/H, what is productivity?

A. 12 units per labor hour

B. 20 units per labor hour

C. 600 units per labor hour

D. 7,200 units per labor hour

Correct Answer: B. 20 units per labor hour

Question 10

Which principle best describes a prevention-oriented quality system?

A. Reward speed even when verification is skipped

B. Make incorrect actions difficult through controls at the point of decision

C. Inspect only a random sample after dispatch

D. Allow informal overrides without records

Correct Answer: B. Make incorrect actions difficult through controls at the point of decision