Demand Planning and Omnichannel Inventory Management

Introduction

Omnichannel retailing integrates stores, websites, mobile applications, marketplaces, social-commerce interfaces, and fulfillment partners into a customer proposition that appears seamless to the buyer. Inventory, however, is rarely seamless. Stock may be physically dispersed across stores, distribution centers, supplier locations, in-transit shipments, returns facilities, and customer orders. Demand is similarly fragmented across channels and is affected by promotions, delivery promises, substitution, stockouts, returns, and changing customer behavior.

Demand planning and inventory management therefore have a strategic role. Their purpose is not simply to minimize inventory. The objective is to provide an economically and ethically appropriate level of product availability while controlling working capital, markdowns, waste, service failures, and operational complexity. In omnichannel systems, a unit of inventory can serve several channels, and a store can function simultaneously as a selling location, pickup point, return center, and fulfillment node.

This chapter develops an integrated approach to forecasting, demand sensing, safety stock, reorder points, single-period decisions, inventory segmentation, multi-echelon control, allocation, substitution, pooling, visibility, and the bullwhip effect.

Learning Objectives

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

1. Explain how omnichannel demand differs from traditional store demand.
2. Apply forecasting, demand-sensing, safety-stock, and reorder-point methods.
3. Use the newsvendor model for short-life or seasonal products.
4. Segment inventory using ABC and XYZ analysis.
5. Evaluate allocation, replenishment, substitution, pooling, and multi-echelon decisions.
6. Diagnose causes of poor availability, excess inventory, and the bullwhip effect.
7. Assess technology, governance, ethical, and managerial implications of inventory decisions.

1. The Nature of Omnichannel Demand

Demand is the quantity customers would purchase under specified price, availability, service, and communication conditions. Observed sales are not identical to demand. If a product is unavailable, the recorded sales may be zero even though customers wanted it. If a retailer substitutes a similar product, the sale may be recorded against the substitute rather than the originally requested item. Promotions may advance purchases, while returns and cancellations may distort the timing of demand.

Important demand distortions include:

• Lost sales: customers leave without purchasing because the product is unavailable.
• Backorders: demand is recorded but fulfilled later.
• Substitution: a customer accepts an alternative item, size, brand, or fulfillment method.
• Channel switching: a shopper moves from store to website or marketplace.
• Order splitting: one customer order is supplied from multiple locations.
• Returns: demand is reversed or inventory is reintroduced after inspection.
• Phantom inventory: the information system reports stock that is damaged, misplaced, reserved, or otherwise unavailable.

A useful planning distinction is between demand creation and demand fulfillment. Marketing actions, assortment, price, and communication influence demand creation. Inventory positioning, capacity, transportation, and order promising determine whether that demand can be fulfilled.

An omnichannel demand plan should therefore be built at several levels: total market demand, category demand, item demand, channel demand, location demand, and fulfillment-method demand. Excessive detail can create noise and unstable forecasts; insufficient detail can hide local constraints. The appropriate level depends on product characteristics, lead time, replenishment frequency, and the cost of forecast error.

2. Forecasting and Demand Sensing

2.1 Forecasting principles

A forecast is a structured estimate of future demand, not a target or a promise. Forecasts should be statistically reproducible, explicitly adjusted for known events, and evaluated against actual outcomes. A forecast process normally includes:

1. cleansing sales and availability data;
2. identifying trend, seasonality, promotions, and lifecycle effects;
3. generating a baseline statistical forecast;
4. incorporating causal information and managerial judgment;
5. reconciling forecasts across product, channel, and organizational levels;
6. measuring error and revising parameters.

For a simple moving average over k periods:

Here D_t is observed demand in period t, F_t+1 is the forecast for the next period, and k is the number of periods. The method assumes that recent observations are reasonably representative and that no strong trend or seasonal pattern has been omitted.

An exponential smoothing model gives greater weight to recent observations:

where 0 ≤ α ≤ 1 is the smoothing constant. A high α responds rapidly to changes but may overreact to random variation; a low α is more stable but slower to detect change.

Forecast errors should be monitored using measures appropriate to the decision. Mean absolute deviation is:

where n is the number of observations. Mean absolute percentage error is:

MAPE is unsuitable when actual demand is zero or close to zero. Weighted absolute percentage error, bias, or scale-dependent measures are often preferable for intermittent demand. Forecast bias may be assessed as:

A persistent positive bias indicates systematic underforecasting; a persistent negative bias indicates overforecasting.

2.2 Demand sensing

Demand sensing uses short-term signals to update a baseline forecast. Signals may include current orders, browsing and search behavior, stockout records, weather, local events, promotion response, delivery bookings, competitor prices, and social-media activity. It is most valuable when demand changes faster than the conventional planning cycle.

Demand sensing is not a substitute for causal understanding. A surge in online searches may reflect curiosity rather than purchases; a sudden fall in store sales may reflect a stockout rather than weak demand. Data provenance, latency, bias, and representativeness must be documented. Machine-learning models can improve nonlinear prediction, but they can also reproduce historical allocation bias or become unstable when product, price, or channel conditions change.

2.3 Forecast hierarchy and reconciliation

Forecasts often exist at total-company, category, channel, region, store, and SKU levels. Independently generated forecasts may not add up. Hierarchical reconciliation ensures that detailed forecasts are consistent with aggregate forecasts. Management should decide whether the hierarchy is based on geography, channel, product, or a combination. In omnichannel operations, a channel forecast should not be treated as an immutable entitlement if inventory can be flexibly pooled.

3. Service Levels, Safety Stock, and Reorder Points

3.1 Cycle stock and safety stock

Cycle stock covers expected demand between replenishment opportunities. Safety stock protects against uncertainty in demand, lead time, or both. The economic cost of safety stock includes capital, storage, obsolescence, insurance, handling, and markdown exposure; the benefit is improved availability and customer service.

For demand with standard deviation σ_D during a fixed lead time L, safety stock is commonly estimated as:

where z is the standard-normal quantile associated with the desired cycle-service level, and time units for σ_D and L must be compatible. This formula assumes independent, identically distributed demand errors and fixed lead time.

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

where μ_D is mean demand per period, σ_L is the standard deviation of lead time, and L is mean lead time. This approximation assumes demand and lead-time variation are suitably represented and are approximately independent. If they are correlated, a covariance term should be considered rather than applying the formula mechanically.

The reorder point is:

The first term is expected lead-time demand, and the second is safety stock. The reorder point is a trigger, not necessarily the order quantity.

Worked Example 1: Reorder point

A retailer sells 80 units per day. Daily demand standard deviation is 20 units, lead time is 5 days, and the desired cycle-service level is 95 percent. For a one-sided normal approximation, z ≈ 1.645.

Lead-time demand is:

80 × 5 = 400 units.

Safety stock is:

1.645 × 20 × √5 ≈ 73.6 units.

Thus:

ROP ≈ 400 + 73.6 = 473.6,

or approximately 474 units after rounding upward. This result is appropriate only if the demand distribution and service-level interpretation are acceptable. It does not account for case-pack constraints, supply disruption, or forecast bias.

3.2 Service measures

Cycle service level is the probability of avoiding a stockout during a replenishment cycle. Fill rate is the proportion of demand units fulfilled immediately. They are not equivalent: a cycle may experience one stockout that affects many units, producing a high cycle-service result but a low fill rate. Managers should select the service metric that reflects customer experience and economic consequences.

4. Single-Period Inventory and the Newsvendor Model

For seasonal, perishable, fashionable, or rapidly obsolete products, a replenishment decision may be made once before demand is known. The newsvendor model balances the cost of ordering too much against the cost of ordering too little.

Let:

• C_u = underage cost per unit;
• C_o = overage cost per unit;
• p = selling price;
• c = purchase cost;
• v = recovery or salvage value;
• g = lost contribution or shortage penalty, if applicable.

A common formulation is C_u = p − c and C_o = c − v, excluding special shortage penalties. The critical fractile is:

where Q* is the optimal order quantity and D is random demand. Equivalently, Q* is the demand quantile corresponding to the critical ratio. The model assumes a single procurement decision, known cost parameters, and a specified recovery value. It is not appropriate when replenishment is continuous and rapid or when pricing and substitution are ignored.

Worked Example 2: Seasonal fashion

A jacket sells for $120, costs $70, and has an end-of-season recovery value of $40. Therefore, C_u = 120 − 70 = 50 and C_o = 70 − 40 = 30. The critical ratio is:

50/(50 + 30) = 0.625.

The retailer should order the 62.5th percentile of the demand distribution. If demand is normally distributed with mean 1,000 and standard deviation 200, the corresponding standard-normal value is approximately 0.319. Thus:

Q* = 1,000 + 0.319 × 200 ≈ 1,064 jackets.

The result is sensitive to the recovery value and to whether markdowns genuinely recover $40. A model that ignores brand damage, returns, or ethical disposal costs may overstate the attractiveness of over-ordering.

5. Inventory Segmentation: ABC and XYZ

ABC analysis classifies items by annual dollar usage:

A typical policy assigns high-value A items more frequent review and tighter controls, medium-value B items standard controls, and low-value C items simplified controls. Cutoffs such as 70–80 percent of value for A items are conventions, not laws; they should reflect the portfolio.

XYZ analysis classifies demand predictability. X items have stable demand, Y items have moderate variation or seasonality, and Z items are highly erratic or intermittent. Combining the dimensions produces a more useful policy matrix.

ABC does not measure strategic importance. A low-cost safety component may cause severe reputational or operational damage if unavailable. Criticality, margin, shelf life, substitutability, and regulatory status should supplement ABC/XYZ.

6. Allocation, Replenishment, and Substitution

6.1 Allocation under scarcity

When supply is insufficient, allocation should be based on transparent objectives. Possible priorities include contractual commitments, contribution margin, customer lifetime value, promised delivery dates, fairness across locations, or protection of essential goods. A purely margin-based rule can disadvantage remote customers or lower-income communities; a purely equal rule can waste scarce inventory where demand is strongest.

Allocation decisions should distinguish initial allocation from replenishment. Initial allocation distributes a new product before demand evidence exists. Replenishment responds to observed sales and remaining inventory. A store with low stock may have high demand, or it may simply have received too little inventory. Sell-through must therefore be interpreted alongside availability and lost-sales estimates.

6.2 Replenishment policies

Common policies include:

• Continuous review ,Q: order a fixed quantity when inventory position reaches the reorder point.
• Periodic review P,S: review every P periods and order up to level S.
• Min–max: replenish when stock falls below a minimum to a specified maximum.
• Order-up-to policy: set a target based on expected demand over the review and protection period plus safety stock.

Inventory position is generally:

The exact definition must match the system’s reservation logic. Counting allocated stock as available when it is already promised produces phantom availability.

6.3 Substitution

Substitution can protect sales and customer satisfaction, but it may reduce margin, create dissatisfaction, or violate customer preferences. A substitution policy should specify acceptable alternatives by product attributes, price range, brand, size, dietary or safety requirements, and customer consent. Digital recommendation systems should explain material differences and should not silently substitute regulated, medical, allergen-sensitive, or safety-critical products.

7. Inventory Pooling and Multi-Echelon Management

7.1 Risk pooling

Pooling inventory across stores or channels can reduce aggregate safety stock because independent demand fluctuations partially offset one another. If demand standard deviations are σ₁, σ₂, …, σ_n and pairwise correlations are ρ_ij, the standard deviation of aggregate demand is:

If demands are independent, all correlations are zero. If they are positively correlated, pooling benefits are smaller; if perfectly correlated, variability grows in direct proportion to the number of locations rather than being diversified. Pooling also requires transportation capacity, accurate inventory visibility, and acceptable delivery time.

7.2 Physical and virtual pooling

Physical pooling places inventory in a common location, such as a regional distribution center. Virtual pooling keeps stock physically distributed but makes it visible and orderable across channels. Virtual pooling can improve availability quickly, but it may create fulfillment complexity, store labor requirements, transfer costs, and customer promise failures if inventory records are inaccurate.

7.3 Multi-echelon inventory

A multi-echelon system contains several connected stocking levels, such as supplier, national distribution center, regional distribution center, store, and customer. Local optimization can increase total inventory: each echelon may independently add safety stock for the same uncertainty. Echelon planning considers inventory downstream and upstream as a system.

Strategic questions include:

• Which products should be held centrally versus locally?
• Which node can meet the customer promise at lowest total cost?
• Should stores retain safety stock for walk-in demand while using the distribution center for online orders?
• How should returns be inspected and reinjected into the network?

The answer depends on demand density, lead times, product value, delivery promise, transportation cost, and the value of immediate store availability.

8. Availability, Markdown, and the Bullwhip Effect

Availability management begins with a reliable distinction between on-hand, sellable, reserved, in-transit, and promised inventory. Key measures include:

• on-shelf availability;
• online availability;
• fill rate;
• perfect-order rate;
• inventory record accuracy;
• stockout duration;
• sell-through;
• inventory turnover;
• weeks of supply;
• markdown rate;
• return-to-stock cycle time.

Inventory turnover is commonly expressed as:

Weeks of supply is approximately:

These metrics require consistent valuation and time windows. High turnover is not automatically desirable if it is achieved through stockouts; high availability is not desirable if achieved through excessive markdowns and waste.

Markdowns are a form of demand and inventory control. They may clear obsolete stock, stimulate demand, or correct an initial allocation error. However, frequent markdowns can train customers to delay purchases, weaken price integrity, and obscure forecast performance. Markdown decisions should consider remaining selling time, inventory exposure, price elasticity, channel cannibalization, and recovery value.

The bullwhip effect is the amplification of demand variability as one moves upstream. It can result from forecast updating, order batching, price promotions, rationing, shortage gaming, long lead times, and poor information sharing. A simplified measure is:

A ratio greater than one indicates amplification, although measurement must use comparable time periods and definitions. Remedies include sharing point-of-sale data, shortening lead times, reducing batch sizes, stabilizing prices, using collaborative planning, and separating genuine demand changes from replenishment noise.

9. An Integrated Planning Process

An effective omnichannel process links commercial planning and operations rather than treating inventory as a late-stage constraint.

1. Establish the demand signal. Reconcile orders, sales, availability, returns, promotions, and channel switching.
2. Create a baseline forecast. Select a method suited to demand pattern, history, and product lifecycle.
3. Add causal intelligence. Incorporate price, promotion, events, weather, assortment changes, and supply constraints.
4. Segment inventory. Combine value, variability, criticality, margin, shelf life, and substitutability.
5. Set service objectives. Differentiate customer promises and economic consequences rather than applying one service level to every item.
6. Position inventory. Decide where units should be held and whether pooling is feasible.
7. Allocate and replenish. Use inventory position, capacity, lead time, and scarcity rules.
8. Sense and replan. Update short-term decisions without destabilizing the entire plan.
9. Measure outcomes. Evaluate availability, working capital, markdowns, forecast error, and customer experience together.
10. Learn. Investigate root causes rather than merely changing safety-stock parameters.

10. Technology, AI, and Governance

Modern inventory platforms can integrate point-of-sale, enterprise resource planning, warehouse-management, order-management, transport, marketplace, and customer data. Computer vision may detect shelf gaps; radio-frequency identification can improve item-level visibility; digital twins can test network policies; and machine learning can forecast nonlinear demand or recommend transfers.

Technology does not eliminate managerial judgment. Data must be governed through clear ownership, master-data standards, event timestamps, exception rules, and audit trails. Important controls include:

• versioning of forecasts and overrides;
• explanation of automated allocation decisions;
• monitoring for systematic service disparities by geography or customer group;
• protection of personal and behavioral data;
• human approval for high-impact substitutions and rationing;
• cybersecurity for connected inventory systems;
• contingency procedures when systems or data feeds fail.

Ethically, retailers should not use personalized availability or pricing in ways that unfairly exclude vulnerable customers. Scarce essential products require transparent rationing. Disposal and markdown decisions should consider food waste, environmental impact, labor conditions, and product safety. Sustainability is not achieved by merely shifting unsold inventory to returns or liquidation channels.

11. Illustrative Case: Northstar Outfitters

Background

Northstar Outfitters, a fictional regional apparel retailer, operates 40 stores, one distribution center, and an online channel. Customers may buy online for home delivery, collect from stores, or return online purchases in stores. The company experiences both online stockouts and end-of-season markdowns.

Problem

The planning team forecasts online and store demand independently. Store inventory records are accurate only 88 percent of the time. Stores reserve safety stock for walk-in shoppers, while the online system promises ship-from-store availability. During a winter promotion, online orders increased, but several stores could not locate supposedly available units. The distribution center then expedited replenishment, while other stores later liquidated excess stock.

Analysis

The apparent shortage was partly a visibility problem. Reported on-hand inventory included damaged units, customer reservations, and misplaced garments. Independent forecasts also treated channel switching as new demand. Finally, the promotion created order batching and caused upstream suppliers to interpret temporary orders as a permanent demand increase.

Northstar segmented products using ABC/XYZ analysis. High-value, stable basics were candidates for pooled inventory and frequent replenishment. Fashion items with short selling windows were managed with a single-period model and explicit markdown scenarios. A cycle count program focused on high-value omnichannel-enabled items. The order-management system was changed so that allocated and unsellable stock was excluded from available-to-promise inventory.

Approach and solution

Northstar implemented four changes. First, it created a common demand view combining store sales, online orders, returns, stockouts, and promotion calendars. Second, it introduced a regional pooling rule: stores could fulfill online orders only when inventory accuracy and labor capacity met defined thresholds. Third, it used service-level differentiation rather than a uniform safety-stock percentage. Fourth, it replaced blanket promotional ordering with a causal forecast and supplier information-sharing process.

The retailer also introduced substitution rules for basic garments, but required customer approval when size, material, or brand attributes differed. A dashboard reported availability, inventory accuracy, markdown exposure, and fulfillment cost together.

Outcome

In the illustrative scenario, inventory accuracy for selected items rose from 88 percent to 97 percent, online cancellations declined, and emergency transfers fell. Total inventory did not fall immediately because Northstar invested in cycle counting and central availability. Over subsequent seasons, improved pooling and more disciplined initial allocation reduced markdown exposure without lowering service for core products.

Learning points

The case demonstrates that inventory problems are often systemic rather than purely quantitative. Better safety-stock calculations cannot compensate for inaccurate records, fragmented forecasts, unclear reservations, or unsuitable fulfillment rules. Omnichannel performance requires coordinated decisions about data, inventory position, customer promises, labor, transportation, and commercial policy.

Challenges and Limitations

No inventory model fully captures reality. Normal approximations can underestimate tail risk, especially during promotions or disruption. Historical sales may be censored by stockouts and may encode past allocation bias. Forecast accuracy can deteriorate after assortment, price, channel, or fulfillment changes. Pooling can reduce safety stock while increasing transport emissions and store workload. Centralization may improve efficiency but weaken immediate local availability. AI may generate precise-looking recommendations from incomplete or biased data.

Managers should use scenario analysis, sensitivity testing, exception management, and post-event review. They should also distinguish parameter uncertainty from demand uncertainty. A policy that is optimal under average conditions may be unacceptable during a pandemic, port closure, cyberattack, or sudden regulatory change.

Best-Practice Checklist

• Forecast demand rather than recorded sales alone.
• Correct for stockouts, returns, substitutions, and channel switching.
• Use safety stock formulas only when their assumptions are defensible.
• Differentiate service levels by economics and customer promise.
• Combine ABC/XYZ with criticality, margin, shelf life, and substitutability.
• Treat inventory accuracy as a strategic capability.
• Pool inventory only when visibility, capacity, and delivery economics support it.
• Share point-of-sale and promotion information to reduce bullwhip amplification.
• Evaluate availability, working capital, markdowns, waste, and fulfillment cost together.
• Govern automated decisions with transparency, privacy, fairness, and human oversight.

Chapter Summary

Omnichannel inventory management coordinates fragmented demand and distributed stock across stores, digital channels, fulfillment nodes, and returns flows. Forecasts must distinguish demand from observed sales and should incorporate availability, promotions, channel switching, and short-term signals. Moving averages and exponential smoothing provide useful baselines, while error measures and bias monitoring support control.

Safety stock protects against demand and lead-time uncertainty, and reorder points combine expected lead-time demand with a protection buffer. The newsvendor model supports one-time decisions for seasonal and perishable products through the critical fractile. ABC/XYZ analysis helps align control effort with value and predictability, but must be supplemented by criticality and strategic importance.

Allocation, replenishment, substitution, and pooling determine how scarce or dispersed inventory serves customers. Multi-echelon planning prevents local decisions from inflating total system inventory. Availability depends on accurate records and clear distinctions among on-hand, reserved, sellable, and promised units. Markdown policies manage aging inventory but can create behavioral and financial consequences. Information sharing, shorter lead times, smaller batches, and stable pricing help reduce the bullwhip effect.

Technology and AI can improve sensing, visibility, and decision quality, but they require data governance, auditability, cybersecurity, privacy protection, and fairness controls. The central managerial lesson is that service, cost, resilience, sustainability, and customer trust must be optimized as an integrated system.

Key Terms

Demand planning: The process of estimating, interpreting, and shaping future demand for operational and commercial decisions.

Demand sensing: Short-term updating of a forecast using current operational and market signals.

Forecast bias: Systematic tendency to overforecast or underforecast.

Lost sales: Customer demand that is not fulfilled and is not recorded as a completed sale.

Stockout: A condition in which required inventory is unavailable for immediate fulfillment.

Safety stock: Inventory held to protect against demand, lead-time, or supply uncertainty.

Reorder point: Inventory-position threshold that triggers replenishment.

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

Fill rate: Proportion of demand units fulfilled immediately.

Newsvendor model: A single-period inventory model balancing underage and overage costs.

Critical fractile: Target demand quantile determined by underage and overage costs.

ABC analysis: Classification based primarily on annual dollar usage.

XYZ analysis: Classification based on demand variability or predictability.

Inventory position: On-hand plus on-order inventory minus backorders and allocated stock.

Inventory pooling: Combining demand or inventory resources to reduce variability or improve availability.

Multi-echelon inventory: Inventory positioned across connected supply-chain levels.

Substitution: Fulfilling demand with an alternative product or service option.

Markdown: A price reduction used to stimulate demand or clear aging inventory.

Bullwhip effect: Amplification of demand variability upstream in a supply network.

Phantom inventory: System-recorded stock that is unavailable because it is lost, damaged, reserved, or otherwise unsellable.

Review Questions

Short Answer Questions

1. Why are observed sales not always an accurate measure of demand in an omnichannel system?
2. Define safety stock and explain its purpose.
3. What is the difference between cycle service level and fill rate?
4. What dimensions are combined in ABC/XYZ analysis?
5. Identify two causes of the bullwhip effect.

Long Answer Questions

1. Explain how an omnichannel retailer should design an integrated demand-planning process.
2. Compare continuous-review and periodic-review replenishment policies, including their advantages and limitations.
3. Derive and interpret the reorder-point logic for variable demand and fixed lead time.
4. Discuss the advantages and risks of inventory pooling across stores and digital channels.
5. Evaluate the role of AI and data governance in omnichannel inventory management.

Analytical / Application Questions

1. Daily demand is 120 units, daily standard deviation is 30 units, lead time is 4 days, and the target service level is 97.5 percent with z = 1.96. Calculate safety stock and reorder point using the fixed-lead-time formula.
2. A seasonal product sells for $90, costs $50, and has a salvage value of $20. Calculate the critical fractile and explain what demand percentile should determine the order quantity.
3. An online system reports 500 units on hand, 80 units on order, 60 units backordered, and 100 units allocated to customer orders. Calculate inventory position and explain why treating allocated units as available would create a customer-service problem.

Multiple Choice Questions with Answers

Question 1

Which statement best describes the difference between observed sales and demand?

A. Observed sales always exceed demand because of returns.

B. Observed sales may understate demand when stockouts cause lost sales.

C. Observed sales are independent of inventory availability.

D. Observed sales exclude all promotional effects.

Correct Answer: B. Observed sales may understate demand when stockouts cause lost sales.

Question 2

A retailer uses exponential smoothing with a high value of α. What is the most likely effect?

A. The forecast responds more rapidly to recent changes.

B. The forecast ignores the latest observation.

C. The forecast becomes permanently equal to average demand.

D. The forecast cannot be updated after a promotion.

Correct Answer: A. The forecast responds more rapidly to recent changes.

Question 3

Which formula represents the reorder point under fixed lead time?

A. ROP = SS − μ_D L

B. ROP = μ_D/L + SS

C. ROP = μ_D L + SS

D. ROP = μ_D + L − SS

Correct Answer: C. ROP = μ_D L + SS

Question 4

Daily demand is 100 units, lead time is 3 days, daily demand standard deviation is 10 units, and z = 1.645. What is approximate safety stock?

A. 16 units

B. 28 units

C. 49 units

D. 300 units

Correct Answer: B. 28 units

Question 5

In the newsvendor model, an increase in the underage cost, holding all else constant, generally causes the optimal order quantity to:

A. Decrease toward zero.

B. Remain unchanged in every case.

C. Increase because the critical fractile rises.

D. Become equal to the salvage value.

Correct Answer: C. Increase because the critical fractile rises.

Question 6

What is the primary basis of ABC analysis?

A. Annual dollar usage

B. Customer age

C. Supplier location only

D. Product color

Correct Answer: A. Annual dollar usage

Question 7

Which condition generally reduces the benefit of inventory pooling?

A. Independent demand across locations

B. Accurate shared inventory visibility

C. Highly positively correlated demand across locations

D. Flexible fulfillment capacity

Correct Answer: C. Highly positively correlated demand across locations

Question 8

Which item should normally be excluded from available-to-promise inventory?

A. Sellable stock that is physically present

B. Stock already allocated to a customer order

C. Stock that has passed quality inspection

D. Unreserved stock at an enabled fulfillment node

Correct Answer: B. Stock already allocated to a customer order

Question 9

Which action is most likely to reduce the bullwhip effect?

A. Increasing order batch sizes

B. Withholding point-of-sale data

C. Stabilizing prices and sharing demand information

D. Extending replenishment lead times

Correct Answer: C. Stabilizing prices and sharing demand information

Question 10

A product has a selling price of $100, purchase cost of $60, and salvage value of $20. What is its critical fractile under the basic newsvendor formulation?

A. 0.40

B. 0.50

C. 0.60

D. 0.67

Correct Answer: C. 0.60