Inventory Management

Introduction and Context

Inventory is the stock of materials, components, work in process, finished goods, spare parts, and supplies held to support production and customer service. It is an operational necessity, but it is also a form of capital tied up in the supply chain. Excess inventory increases financing, storage, insurance, obsolescence, and handling costs; insufficient inventory causes stockouts, production stoppages, lost sales, expediting, and reputational damage.

Inventory management is therefore the coordinated design and control of replenishment, storage, allocation, and review policies so that an organization achieves an economically appropriate balance between availability, cost, responsiveness, and risk. In contemporary supply chains, this balance must be considered across suppliers, plants, warehouses, distributors, and customers rather than at a single location.

The central managerial challenge is not simply to minimize inventory. It is to determine where inventory should be held, in what form, in what quantity, and under which replenishment rules.

Learning Objectives

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

1. Classify inventory according to its role, form, and stage in the supply chain.
2. Explain the cost and service trade-offs involved in inventory decisions.
3. Apply ABC analysis and design differentiated control policies.
4. Calculate economic order quantity, reorder points, safety stock, and service levels.
5. Apply the newsvendor model to single-period decisions.
6. Compare continuous-review and periodic-review systems.
7. Explain multi-echelon inventory logic and the effects of risk pooling.
8. Evaluate technological, ethical, and managerial implications of inventory policies.

1. The Nature and Functions of Inventory

Inventory performs several important functions:

• Cycle-stock function: inventory held because replenishment occurs in batches rather than continuously.
• Safety-stock function: inventory held to protect against demand and supply uncertainty.
• Anticipation function: inventory built before seasonal demand, promotions, shutdowns, or expected price increases.
• Pipeline function: inventory in transit or otherwise committed but not yet available for use.
• Decoupling function: inventory separating operations so that disruption in one process does not immediately stop another.
• Speculation function: inventory purchased in expectation of scarcity, inflation, exchange-rate changes, or supplier price increases.

Inventory may be classified by physical stage as raw materials, purchased components, work in process (WIP), finished goods, maintenance-repair-operating supplies (MRO), and returned or obsolete stock. It may also be classified by ownership: owned inventory, consignment inventory, vendor-managed inventory, or inventory held by a third-party logistics provider.

A useful managerial distinction is between dependent demand and independent demand. Demand for finished products and service parts is generally independent and must be forecast or observed. Demand for components is dependent on the production plan, bill of materials, and inventory records; it is often planned through material requirements planning (MRP).

2. Inventory Costs and Performance Measures

The principal cost categories are as follows.

Annual inventory cost can be represented conceptually as:

The exact components included depend on the decision. For example, purchase cost is constant in the basic EOQ model when unit price does not vary with quantity, so it does not affect the optimal order quantity, although it remains relevant to total cost and cash planning.

Important performance measures include inventory turnover, days of supply, fill rate, cycle-service level, stockout frequency, inventory accuracy, and obsolete-inventory percentage.

Turnover must be interpreted carefully. A high turnover may indicate efficient inventory use, but it may also indicate inadequate stock and frequent shortages. Similarly, a high fill rate does not necessarily mean that every customer order is delivered completely if the measurement definition is inconsistent.

3. ABC Analysis and Selective Control

ABC analysis ranks inventory by annual usage value, usually calculated as:

Items are then ranked from highest to lowest usage value and divided into policy groups. A typical, though not universal, arrangement is:

The percentages are heuristics, not laws. ABC analysis is valuable because managerial attention is scarce. A-items may justify demand analysis, supplier collaboration, cycle counting, and strict authorization. C-items may be managed with two-bin systems, broad reorder parameters, or periodic replenishment.

ABC analysis has limitations. It considers value but not necessarily criticality, lead time, substitutability, supply risk, regulatory importance, or demand variability. A low-cost safety component can stop an entire production line. Consequently, organizations often combine ABC with a second dimension, such as criticality or supply risk, producing an ABC–criticality or ABC–XYZ matrix. XYZ classifications commonly distinguish demand variability: X items have relatively stable demand, Y items show moderate variation or seasonality, and Z items are highly irregular.

4. The Economic Order Quantity Model

The basic EOQ model determines the replenishment quantity that minimizes relevant annual ordering and holding costs.

where:

• Q* = optimal order quantity, in units per order;
• D = annual demand, in units per year;
• S = ordering or setup cost, in currency per order;
• H = annual holding cost, in currency per unit per year.

If holding cost is expressed as a fraction i of unit price C, then H = iC. The formula assumes constant and known demand, constant lead time, instantaneous replenishment, no shortages, no quantity discounts, and a constant unit price. It is most appropriate for stable, repetitive replenishment decisions.

The relevant annual cost function is:

The first term is annual ordering cost and the second is annual cycle-stock holding cost. With instantaneous replenishment, average cycle stock is Q ÷ 2. Differentiating the cost function with respect to Q and setting the result to zero gives the EOQ. At the optimum, annual ordering cost equals annual cycle-stock holding cost.

Worked Example: EOQ

A distributor has annual demand of 12,000 units. The ordering cost is $75 per order and annual holding cost is $4 per unit.

The number of orders per year is approximately:

The order interval, using 365 days, is approximately 20.4 days. This result is a starting point rather than an unquestionable prescription. Capacity constraints, supplier minimums, quantity discounts, transportation economics, and demand uncertainty may require adjustment.

Extensions of EOQ

If replenishment occurs gradually during an internal production run, the economic production quantity is used:

where d is the demand rate and p is the production rate, with p > d. The factor reflects the fact that inventory accumulates more slowly than under instantaneous delivery.

For quantity discounts, each feasible price bracket must be evaluated. The organization should calculate the EOQ using the holding cost applicable to each price, adjust it to a feasible quantity in that bracket, calculate total cost including purchase cost, and select the lowest feasible total cost. It is incorrect to choose a discount solely because the unit price is lower.

5. Reorder Point, Lead-Time Demand, and Safety Stock

The reorder point (ROP) specifies when a replenishment order should be placed. Under deterministic demand and lead time:

where d is average demand per time unit and L is lead time in the same time unit. Dimensional consistency is essential: daily demand must be paired with lead time in days.

When demand or lead time is uncertain, safety stock is added:

If demand per period is normally distributed with standard deviation σ_d and lead time is constant at L periods, the standard deviation of lead-time demand is:

For a target cycle-service level associated with standard-normal factor z, safety stock is:

Thus:

These equations assume independent, identically distributed period demand and a normal approximation. They are not automatically valid for intermittent demand, strong seasonality, correlated demand, random lead time, or very low-volume items.

If both demand and lead time are random and independent, with mean demand d, mean lead time L, demand standard deviation σ_d, and lead-time standard deviation σ_L, a commonly used approximation is:

Then SS = zσ_LT. If both variables are uncertain and correlated, an additional covariance term is required; ignoring correlation can understate or overstate risk.

Worked Example: Reorder Point

Daily demand has mean 100 units and standard deviation 20 units. Lead time is constant at 5 days. Management wants a 95% cycle-service level, for which z ≈ 1.645.

Therefore:

The system should trigger replenishment when inventory position reaches approximately 574 units. Inventory position includes on-hand inventory plus on-order inventory minus valid backorders; using only on-hand stock can cause premature or delayed orders.

6. Service Levels and the Cost of Availability

A cycle-service level is the probability of not experiencing a stockout during a replenishment cycle. A fill rate is the proportion of demand units supplied immediately from stock. These measures are related but not identical. A system can have a high cycle-service level yet a lower fill rate when a stockout, once it occurs, is severe.

For a normal lead-time-demand distribution, a target cycle-service level can be converted to a standard-normal value z. For example, 90%, 95%, and 99% correspond approximately to z = 1.282, 1.645, and 2.326. The incremental safety stock required for higher service levels grows because the normal distribution has a progressively thinner tail.

Service targets should be differentiated. Critical medical supplies, aircraft parts, and production bottleneck components may justify high availability. Low-value, easily substituted items may not. The target should reflect the economic consequences of shortage, not an arbitrary corporate percentage.

7. The Newsvendor Model for Single-Period Decisions

The newsvendor model applies when inventory is ordered once for a selling season or short life cycle and leftover stock has a different value from a stockout. Examples include fashion products, newspapers, seasonal food, event merchandise, and capacity for a one-time service.

Let:

• p = selling price per unit;
• c = purchase or production cost per unit;
• v = recovery or salvage value per leftover unit;
• g = p − c = underage cost, the margin lost when one unit is not available;
• l = c − v = overage cost, the loss from ordering one unit too many.

The critical fractile is:

where Cu is underage cost and Co is overage cost. In the standard model, Cu = p − c and Co = c − v, so:

The optimal quantity is the demand quantile corresponding to this cumulative probability. The model assumes one ordering opportunity, known demand distribution, and specified treatment of leftovers and shortages.

Example

Suppose a product sells for $50, costs $30, and has a salvage value of $10. Then:

The organization should order the median of the demand distribution. If salvage value decreases, overage becomes more expensive and the target quantile falls. If lost sales are very costly, the target quantile rises.

8. Continuous-Review and Periodic-Review Systems

In a continuous-review, or (Q,r), system, inventory position is monitored continuously. A fixed quantity Q is ordered when inventory position reaches reorder point r. This system responds quickly to changes but requires reliable transaction data and frequent monitoring.

In a periodic-review, or (P,S), system, stock is reviewed every P time units and replenished up to an order-up-to level S. Because no order can be placed between reviews, protection must cover the review period plus lead time. If demand per period has standard deviation σ_d, a simplified order-up-to level is:

The order quantity at a review is:

The equation assumes independent period demand, constant lead time, and a normal approximation. Periodic review is attractive when many items are ordered from one supplier on a common schedule, but it generally requires more safety stock than continuous review because of the additional exposure during the review interval.

9. Multi-Echelon Inventory and Risk Pooling

A multi-echelon supply chain contains several inventory locations, such as suppliers, plants, regional distribution centres, local warehouses, and retailers. An echelon inventory position includes stock at a location and, conceptually, inventory downstream that it is responsible for supporting. Decisions made independently at each node can create excessive total inventory, duplicated safety stock, and poor customer service.

Multi-echelon management asks three connected questions:

1. At which echelon should decoupling inventory be held?
2. How should safety stock be allocated across locations?
3. How should replenishment decisions account for upstream and downstream lead times?

Risk pooling occurs when uncertain demand is aggregated. If independent demand streams with standard deviations σ₁, σ₂, …, σ_n are pooled, aggregate standard deviation is:

For identical independent locations, each with standard deviation σ, the pooled standard deviation is σ√n, whereas separate safety stocks sum to nσ. Thus centralization can reduce total safety stock relative to decentralized stocking. The benefit is weaker when demands are positively correlated, when response times from the central location are long, or when customers require local availability.

Multi-echelon optimization must also consider transportation cost, postponement, product customization, capacity, disruption exposure, taxation, and service differentiation. Centralization is not automatically superior; resilience may require strategically duplicated inventory or regional buffers.

10. Inventory Systems, Technology, and AI

Accurate inventory policy depends on accurate data. Enterprise resource planning (ERP), warehouse management systems, barcode scanning, radio-frequency identification, electronic data interchange, and Internet-of-Things sensors can improve transaction visibility and traceability. However, technology does not correct poor master data, inaccurate bills of material, unrecorded scrap, or weak process discipline.

Analytics can support:

• probabilistic demand forecasting;
• lead-time estimation and supplier reliability analysis;
• anomaly detection for shrinkage and inventory-record errors;
• dynamic safety-stock recommendations;
• segmentation by value, volatility, criticality, and margin;
• simulation of disruptions and policy changes.

AI-based recommendations should be governed through human review, explainability, data-quality controls, monitoring for forecast bias, and clear accountability. A model trained on historical stockouts may interpret censored demand as low demand, because sales were not observed when products were unavailable. Managers must distinguish actual demand from observed sales.

11. Applied Case: Meridian Medical Devices

The following is an illustrative case.

Background

Meridian Medical Devices assembles infusion pumps and maintains hospitals’ critical spare parts. It holds approximately 8,000 stock-keeping units (SKUs). Finance reports rising inventory value, while hospitals report occasional delays for replacement components.

Problem

The existing policy applies the same monthly review and target service level to every item. High-value electronic modules are overstocked, while several low-cost seals and connectors cause emergency shipments because they are classified only by purchase price rather than operational criticality.

Analysis

The operations team first calculates annual usage value and performs ABC analysis. It then adds a criticality dimension based on patient-safety impact, equipment downtime, substitutability, and regulatory approval. Demand history is cleaned to separate true demand from periods of stockout. For stable A-items, a continuous-review policy is selected. For low-value C-items used across many service centres, a two-bin periodic replenishment system is considered.

For an electronic module with annual demand of 2,400 units, ordering cost of $90, and annual holding cost of $12, the basic EOQ is:

The team does not adopt 190 automatically. It checks supplier minimum order quantities, shelf life, cash constraints, and the effect of a supplier’s 30-day lead time. For critical seals, service-level targets are raised despite their low unit value, while an approved substitute is qualified to reduce supply risk.

Approach and Solution

Meridian introduces:

• ABC–criticality segmentation;
• continuous review for critical and high-value components;
• periodic review for stable, low-value supplies;
• lead-time and demand variability in reorder-point calculations;
• cycle counting focused on A and critical items;
• supplier scorecards covering quality, lead time, flexibility, and disruption risk;
• ageing alerts and engineering review before components become obsolete.

Outcome

The organization reduces non-critical excess stock while improving availability of critical service parts. The improvement does not result from a universal reduction in inventory; instead, stock is relocated and differentiated. Some buffers are reduced, while selected critical items receive higher protection.

Learning Points

The case demonstrates that inventory management is a portfolio problem. Unit cost alone is an inadequate basis for control. Service objectives, demand uncertainty, supply risk, regulatory constraints, and the cost of downtime must be integrated into policy design.

12. Limitations, Governance, and Best Practice

Inventory models simplify reality. Forecast distributions may be wrong; lead times may be correlated with demand; supplier capacity may be constrained; and customer behaviour may change after a stockout. EOQ can produce impractical quantities when fixed costs, minimum lots, or shelf life dominate. Normal approximations can be inappropriate for intermittent or highly skewed demand. Models should therefore be validated through sensitivity analysis and simulation where consequences are material.

Ethical and governance issues include expiry management for medicines and food, safe storage, transparent treatment of customer priorities, avoidance of product dumping, responsible disposal, and protection against counterfeit or diverted products. Inventory allocation during shortages may involve morally significant decisions; criteria should be explicit, lawful, and consistent rather than based solely on purchasing power.

Best practices include:

1. Define service measures precisely and distinguish fill rate from cycle-service level.
2. Segment items by value, variability, criticality, and supply risk.
3. Use inventory position, not merely on-hand balance, for replenishment decisions.
4. Match model complexity to decision importance and data quality.
5. Measure forecast bias and distinguish lost demand from zero demand.
6. Review safety stock when lead times, product designs, or service promises change.
7. Integrate procurement, production, sales, finance, and customer-service decisions.
8. Use scenario analysis for disruption, promotion, seasonality, and supplier failure.
9. Audit inventory records and investigate discrepancies rather than simply increasing buffers.
10. Treat obsolescence, resilience, and sustainability as explicit costs and risks.

Chapter Summary

Inventory enables continuity, responsiveness, economies of scale, and protection against uncertainty, but it consumes capital and creates operational risk. Effective management begins by classifying inventory according to stage, function, demand type, value, criticality, and supply risk. ABC analysis directs attention toward high-usage-value items, but it should be supplemented by criticality and variability analysis.

The EOQ model balances ordering and holding costs under stable assumptions. Reorder points combine expected lead-time demand with safety stock, while service-level decisions determine the amount of protection required. Continuous-review systems react more frequently; periodic-review systems consolidate reviews but require protection over the review interval. The newsvendor model addresses single-period decisions in which overage and underage costs are asymmetric. Multi-echelon management extends the analysis across the supply chain and uses risk pooling, postponement, and strategic positioning to balance cost and availability.

No formula substitutes for managerial judgment. Data quality, supplier reliability, customer consequences, ethical priorities, and resilience must shape the final policy. The best inventory system is differentiated, transparent, continuously reviewed, and aligned with the organization’s competitive strategy.

Key Terms

Inventory: Stock held to support production, distribution, or service.

Cycle stock: Inventory resulting from replenishment in batches.

Safety stock: Buffer held against demand or supply uncertainty.

Pipeline inventory: Stock in transit or otherwise not yet available for use.

Dependent demand: Demand derived from the requirements of another item or production plan.

Independent demand: Demand determined by external customers or markets.

Holding cost: Annual cost of owning and storing inventory.

Ordering cost: Cost incurred when placing an order or initiating a setup.

ABC analysis: Ranking inventory by annual usage value.

Criticality: Consequence of an item’s unavailability.

EOQ: Order quantity that minimizes basic ordering plus holding cost.

Reorder point: Inventory-position level that triggers replenishment.

Lead-time demand: Demand expected during replenishment lead time.

Safety stock: Additional inventory above expected demand protection.

Cycle-service level: Probability of avoiding a stockout during a replenishment cycle.

Fill rate: Proportion of demand supplied immediately from stock.

Newsvendor model: Single-period model balancing overage and underage costs.

Periodic review: Inventory policy reviewed at fixed time intervals.

Risk pooling: Reduction of aggregate uncertainty through demand aggregation.

Multi-echelon inventory: Inventory distributed across multiple supply-chain levels.

Review Questions

Short Answer Questions

1. What are the principal functions performed by inventory?
2. How does ABC analysis direct managerial attention?
3. Distinguish cycle-service level from fill rate.
4. State the assumptions of the basic EOQ model.
5. Why can centralized inventory reduce safety stock?

Long Answer Questions

1. Explain the cost and service trade-offs involved in inventory management, using appropriate examples.
2. Derive the basic EOQ relationship and discuss situations in which the result should be modified.
3. Compare continuous-review and periodic-review systems, including their information and safety-stock implications.
4. Explain how demand variability, lead-time variability, and service-level targets influence reorder points.
5. Discuss why ABC analysis should be combined with criticality and supply-risk segmentation.

Analytical / Application Questions

1. Daily demand for an item averages 80 units with a standard deviation of 15 units. Lead time is constant at 4 days. Calculate safety stock and reorder point for a 95% cycle-service level, using z = 1.645.
2. A seasonal item sells for $40, costs $24, and has a salvage value of $8. Calculate the critical fractile and explain how the optimal order quantity would change if salvage value fell to $2.
3. A firm has four regional warehouses with independent demand standard deviation of 100 units each. Compare the uncertainty measure for separate stocking with the pooled uncertainty measure, and discuss operational conditions that could reduce the benefit of centralization.

Multiple Choice Questions with Answers

Question 1

Which inventory type is primarily held to protect against uncertainty in demand or supply?

A. Cycle stock

B. Safety stock

C. Pipeline inventory

D. Anticipation stock

Correct Answer: B. Safety stock

Question 2

Annual demand is 12,000 units, ordering cost is $75, and annual holding cost is $4 per unit. What is the approximate EOQ?

A. 300 units

B. 500 units

C. 671 units

D. 1,200 units

Correct Answer: C. 671 units

Question 3

Which item is most likely to receive the tightest control under ABC analysis?

A. A low-value item with negligible usage

B. An item with the highest annual usage value

C. An item ordered only once per year

D. An item with the largest physical size

Correct Answer: B. An item with the highest annual usage value

Question 4

If average daily demand is 100 units and lead time is 5 days, what is expected lead-time demand?

A. 20 units

B. 105 units

C. 500 units

D. 5,000 units

Correct Answer: C. 500 units

Question 5

For constant lead time and independent daily demand, how does the standard deviation of lead-time demand change when lead time increases from L to 4L?

A. It remains unchanged

B. It doubles

C. It quadruples

D. It increases eightfold

Correct Answer: B. It doubles

Question 6

In the newsvendor model, the critical fractile increases when:

A. The salvage value decreases

B. The underage cost increases relative to overage cost

C. The selling price decreases while all else remains constant

D. The overage cost increases relative to underage cost

Correct Answer: B. The underage cost increases relative to overage cost

Question 7

A periodic-review order-up-to level must generally protect against demand during:

A. Only the review interval

B. Only supplier lead time

C. The review interval plus lead time

D. The previous replenishment cycle only

Correct Answer: C. The review interval plus lead time

Question 8

Which statement best describes inventory position?

A. It is always equal to physical on-hand stock

B. It includes on-hand and on-order inventory, adjusted for valid backorders

C. It excludes inventory in transit

D. It measures only obsolete inventory

Correct Answer: B. It includes on-hand and on-order inventory, adjusted for valid backorders

Question 9

For four independent, identical demand streams with standard deviation 100 units each, pooled standard deviation is:

A. 100 units

B. 200 units

C. 400 units

D. 1,600 units

Correct Answer: B. 200 units

Question 10

Which is a limitation of applying a normal safety-stock formula to intermittent demand?

A. It always produces zero safety stock

B. Intermittent demand may not be well represented by a normal distribution

C. It eliminates the need for lead-time data

D. It applies only to finished goods

Correct Answer: B. Intermittent demand may not be well represented by a normal distribution