Inventory, Materials and Resource Planning

Introduction and Context

Inventory is the stock of materials, components, work-in-process, finished goods, maintenance supplies and information-dependent resources held to support operations. For an entrepreneurial firm, inventory is both an operational capability and a financial commitment. It protects customer service against uncertainty, enables economies of scale and decouples successive activities; however, it also ties up working capital, occupies space, becomes obsolete and can conceal process problems.

Inventory planning therefore seeks an appropriate balance between availability and investment. The central question is not “How much inventory can the firm hold?” but “What level and location of inventory provide the required service at the lowest total cost and risk?” This chapter integrates independent-demand inventory control with dependent-demand planning through materials requirements planning (MRP), enterprise resource planning (ERP), lot sizing and multi-echelon coordination.

Learning Objectives

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

1. Classify inventory and explain its operational and financial roles.
2. Apply economic order quantity (EOQ), reorder-point, safety-stock, service-level and newsvendor models.
3. Explain ABC analysis, cycle counting and inventory-record governance.
4. Construct and interpret bills of material and MRP records.
5. Compare lot-sizing rules and identify limitations of deterministic planning.
6. Explain multi-echelon inventory and the role of ERP and analytics.
7. Design an inventory policy that recognizes uncertainty, cash constraints, sustainability and supply risk.

1. Inventory as an Operations Decision

Inventory performs several functions:

• Cycle stock results from ordering or producing in batches.
• Safety stock protects against demand, supply or processing uncertainty.
• Anticipation stock is accumulated for predictable peaks, promotions or disruption risk.
• Pipeline inventory is in transit or moving between stages.
• Decoupling stock separates operations so that one interruption does not immediately stop another.
• Speculative stock is acquired because of expected price increases, scarcity or exchange-rate movements.
• Obsolete, excess and dead stock cannot be sold or used economically and require special governance.

Inventory may also be classified by physical form: raw materials, purchased components, work-in-process, finished goods, maintenance-repair-operating (MRO) supplies and returned or quarantined goods. The same item can have different policies at different locations. A component may be cycle stock at a plant, pipeline stock in transit and safety stock at a distribution centre.

Inventory decisions involve three broad costs:

1. Ordering or setup cost, including procurement administration, transport arrangements and production changeovers.
2. Holding cost, including capital, storage, insurance, handling, shrinkage, damage and obsolescence.
3. Shortage cost, including lost margin, expediting, lost goodwill, production stoppage and contractual penalties.

A sound policy also considers cash-flow volatility, supplier reliability, minimum order quantities, shelf life, hazardous-material rules and the strategic importance of the item.

2. EOQ: Balancing Ordering and Holding Cost

The basic EOQ model determines the order quantity that minimizes annual ordering plus holding cost when demand is relatively stable.

Here, Q^* is the economic order quantity in units per order, D is annual demand in units per year, S is ordering or setup cost per order, and H is annual holding cost per unit. The model assumes constant known demand, instantaneous replenishment, no shortages, fixed unit price, independent replenishment cycles and a constant holding cost. If the unit price is constant, purchase cost does not affect the optimizing quantity.

Annual relevant cost is:

The first term is annual ordering cost and the second is average cycle-stock holding cost. Differentiating with respect to Q and setting the derivative to zero gives the EOQ expression. At EOQ, ordering cost equals holding cost, although total cost also includes purchasing cost when relevant.

Worked example. An enterprise uses 12,000 units annually. Ordering cost is $75 per order and annual holding cost is $3 per unit.

Q^* = √((2 × 12,000 × 75)/3) = √600,000 ≈ 775 units.

Orders per year are 12,000/775 ≈ 15.5. If demand is evenly distributed over 300 operating days, the cycle interval is approximately 19.4 operating days. EOQ is a benchmark, not an instruction to ignore supplier minimums, capacity, discounts or cash constraints.

Extensions of EOQ

With a finite production rate, replenishment occurs while demand continues. If p is the production rate and d is the demand rate, with p > d, the maximum cycle inventory is reduced and the economic production quantity is:

Quantity discounts require evaluating each feasible price bracket, including the purchase cost DC, ordering cost and holding cost. A candidate quantity is valid only if it lies within its bracket; otherwise the bracket boundary must be tested. Discounts should not be accepted merely because the unit price is lower: incremental holding and cash costs may exceed the saving.

3. Reorder Points, Lead Time and Safety Stock

A continuous-review system issues an order when inventory position reaches a reorder point. Inventory position is normally:

inventory position = on-hand inventory + on-order inventory − backorders.

For constant demand and lead time:

where d is demand per time unit and L is lead time in the same time unit. This is lead-time demand, not a universal “one month’s stock” rule.

When demand varies but lead time is constant, a common policy is:

Here, μ_L is mean demand during lead time, σ_L is its standard deviation and z is the standard-normal multiplier associated with the desired cycle-service probability. Safety stock is zσ_L.

If independent daily demand has mean μ_d and standard deviation σ_d, and lead time is L days:

If both demand and lead time vary independently, with mean lead time μ_L, standard deviation σ_L, mean demand μ_d and standard deviation σ_d, an often-used approximation is:

The first-order approximation omits the interaction term σ_d²σ_L²; a more exact variance under independence is:

The approximation is reasonable when coefficients of variation are not large. If demand and lead time are correlated, covariance must be considered and simulation or empirical lead-time-demand data may be preferable.

Worked example. Daily demand has mean 40 and standard deviation 8 units. Lead time is constant at 5 days. For a 95% cycle service level, take z ≈ 1.645. Then μ_L = 5 × 40 = 200, σ_L = 8√5 ≈ 17.9, safety stock is approximately 1.645 × 17.9 ≈ 29.5 units, and the reorder point is approximately 230 units. The result depends on the interpretation of service level: cycle service level is the probability of no stockout during a replenishment cycle, not the fraction of demand filled.

Service Measures

• Cycle service level (CSL): probability of completing a replenishment cycle without a stockout.
• Fill rate: percentage of demand units supplied immediately.
• Ready rate: proportion of time an item is available.
• Order-line service: proportion of customer order lines supplied complete and on time.

These measures are not interchangeable. A high CSL may coexist with a lower fill rate when occasional shortages are large. Managers must specify the measure, review period and treatment of backorders.

4. Single-Period Decisions: The Newsvendor Model

Perishable, seasonal or rapidly obsolete products often have one meaningful selling period. Ordering too much creates an overage loss; ordering too little creates an underage loss. Let selling price be p, purchase cost c, salvage value v and stockout or lost-margin consequence be represented through the underage cost. In the simple no-shortage-penalty case:

The critical fractile is:

Thus, Q^* is the demand quantile corresponding to that probability. The formula assumes a single order, known demand distribution, no replenishment during the period and specified salvage and shortage economics. If lost goodwill, contractual penalties or substitution effects exist, they must be included in C_u.

For a normally distributed demand with mean μ and standard deviation σ, Q^* = μ + zσ, where z corresponds to the critical fractile. For example, if C_u = 6 and C_o = 4, the fractile is 0.60; the order quantity is the 60th percentile, not automatically the mean.

5. ABC Analysis, Cycle Counting and Control

ABC analysis ranks items by annual dollar usage:

A typical, but not universal, classification is A items as the small proportion representing most value, B items as intermediate, and C items as many low-value items. ABC is a prioritization method, not a statement that C items are unimportant. A low-cost safety-critical bearing may require tighter control than its financial value suggests.

A robust classification may add criticality, lead time, demand variability, obsolescence risk and supply scarcity. This produces a multidimensional segmentation such as “AX” for high-value, stable-demand items and “CZ” for low-value, erratic-demand items.

Cycle counting counts selected items continuously rather than waiting for an annual physical count. A policy might count A items weekly, B items monthly and C items quarterly or semi-annually, but frequencies should reflect error risk and legal requirements. Differences must be investigated at the source: receiving, picking, production consumption, returns, unit-of-measure conversion, scrap and master-data errors. Adjusting records without root-cause correction merely hides process failure.

Controls include barcode or RFID identification, authorized transactions, location discipline, segregation of quarantined stock, serial or lot traceability, expiry-date management and reconciliation between physical and ERP records.

6. Dependent Demand, BOM and MRP

Independent demand is determined externally by customers or markets. Dependent demand is calculated from the requirement for another item. If one finished product requires three units of a component, production of 500 finished units creates a gross component requirement of 1,500 units, before considering inventory and scheduled receipts.

A bill of materials (BOM) is a structured product definition. It identifies parent-child relationships, quantity per assembly, engineering revision, effectivity dates, scrap allowances and sometimes routing or process information. A single-level BOM lists immediate components; a multi-level or indented BOM shows all levels.

MRP translates a master production schedule (MPS) into time-phased planned orders. Its principal inputs are:

1. MPS for end items;
2. BOM and component usage quantities;
3. inventory records, including on-hand balance, scheduled receipts, open orders, lead times, lot-sizing rules and safety stock.

For each item and period, MRP calculates gross requirements, projected available balance, net requirements, planned order receipts and planned order releases. A simplified record is:

Net requirement arises when projected availability would fall below the prescribed safety stock. Planned order release is the planned receipt offset by the item’s lead time. If a component is required in period 6 and lead time is two periods, its release is planned in period 4, subject to calendar and supplier constraints.

Illustrative MRP calculation. Beginning inventory is 80 units, safety stock is 20, there are no scheduled receipts, and gross requirements in periods 1–3 are 30, 50 and 60. With lot-for-lot ordering:

• Period 1 projected balance: 80 − 30 = 50; no receipt.
• Period 2 projected balance: 50 − 50 = 0; net requirement to restore safety stock = 20, so planned receipt = 20.
• Period 3 projected balance: 0 + 20 − 60 = −40; to finish at safety stock 20, planned receipt = 60.

Actual MRP tables must specify whether safety stock is included in projected availability and how negative balances are treated. The example illustrates logic rather than a universal software convention.

MRP is powerful for lumpy, dependent demand, but it is not a demand forecast by itself. Poor forecasts, inaccurate BOMs, unrealistic lead times and nervous rescheduling produce unreliable plans. Planners should use time fences, exception messages, frozen zones and controlled rescheduling.

7. Lot Sizing and ERP Planning

Common lot-sizing rules include:

ERP systems integrate purchasing, production, inventory, sales, finance and sometimes supplier collaboration. Their value depends on governance: one item master, controlled BOM revisions, accurate lead times, disciplined transaction timing, visible exceptions and role-based authorization. ERP does not eliminate judgment; it makes assumptions and errors propagate faster.

Advanced planning and scheduling may consider finite capacity, alternative resources, campaign sequencing and material constraints. APS should be reconciled with the financial plan and shop-floor feasibility rather than treated as an unquestionable optimizer.

8. Multi-Echelon Inventory

A supply chain contains multiple stocking points: suppliers, plants, regional distribution centres, local warehouses and retailers. Optimizing each location independently can inflate total inventory because every echelon protects itself against uncertainty. Multi-echelon inventory management coordinates where safety stock should be held and how demand information and lead-time variability propagate.

A one-echelon policy controls one location in isolation. An echelon inventory position includes stock at a location and all downstream stock or commitments, depending on the defined convention. In a serial system, upstream inventory can serve downstream demand, so the economically appropriate safety stock may be upstream when postponement, commonality and transport reliability permit it.

Multi-echelon design considers customer service, replenishment lead times, transportation frequency, capacity, product substitution and risk pooling. Centralizing inventory often reduces aggregate safety stock through demand pooling, but may increase delivery time and transport cost. The correct decision is a total-cost and service decision, not simply a warehouse-count decision.

9. Inventory Under Uncertainty and Risk

Uncertainty arises from demand variation, forecast error, supplier failures, transport delays, yield loss, quality holds, currency movements and sudden market changes. Practical responses include:

• differentiated service levels by customer and item;
• supplier qualification and dual sourcing;
• postponement and modular product design;
• shorter and more reliable lead times;
• demand sensing and collaborative planning;
• substitution rules and allocation policies;
• strategic reserves for critical materials;
• scenario analysis and stress testing.

Safety stock should be based on observed lead-time demand distributions where possible, not only average demand. Forecast accuracy should be assessed with measures suited to the data, such as bias, mean absolute deviation or weighted error measures; percentage measures can be misleading when demand is near zero.

AI and machine learning can improve demand sensing, anomaly detection, lead-time prediction and inventory segmentation. They do not remove the need for causal understanding, data-quality controls, explainability, human override and monitoring for drift. An algorithm trained on stock-constrained sales may interpret lost sales as low demand and systematically underforecast. Governance should define data ownership, model validation, override authority and audit trails.

10. Applied Case: Illustrative Case of NovaBrew Equipment

Background

NovaBrew, a fictional entrepreneurial manufacturer of compact commercial coffee machines, sells through distributors. It holds imported pumps, locally sourced housings and finished machines. Stockouts delay customer installations, while obsolete electronic boards become costly after model revisions.

Problem

The firm used the same three-month stock rule for all items. Its ERP showed frequent shortages despite apparently high inventory. A physical audit found inaccurate component balances, unrecorded scrap and several obsolete boards. Purchasing optimized unit price rather than total cost.

Analysis

The operations team segmented items by annual dollar usage, criticality and lead-time variability. Pumps were high-value and critical; screws were low-value but readily available; control boards were high-obsolescence and revision-sensitive. The team calculated lead-time demand using actual demand and supplier lead-time history, then selected service levels according to customer and item consequences. It also checked BOM revisions against engineering change notices.

For pumps, the EOQ benchmark was adjusted for supplier minimum order quantity and cash limits. A reorder-point policy with safety stock was used because demand was independent and replenishment was continuous. For control boards, a time-phased MRP policy tied to the production schedule was more appropriate. Cycle counts were assigned weekly to pumps and boards, monthly to housings and quarterly to low-risk fasteners.

Approach and Outcome

NovaBrew introduced barcode receiving, quarantine locations, mandatory scrap transactions and a frozen MRP planning window. It negotiated shorter pump lead times and postponed final configuration of machines. The company measured fill rate, inventory accuracy, aged stock, forecast bias and expedited freight rather than relying on inventory value alone.

Within the illustrative first six months, inventory accuracy improved, emergency purchases declined and obsolete-board exposure became visible earlier. The improvement did not come from holding more of everything; it came from matching control intensity, replenishment logic and data quality to item economics and risk.

Learning Points

The case demonstrates that EOQ, MRP and ABC are complementary, not competing, tools. Independent-demand items require statistical replenishment logic; dependent-demand items require schedule and BOM logic. Both require reliable records, explicit service objectives and cross-functional accountability.

11. Limitations, Ethics and Best Practices

Models simplify reality. EOQ can be misleading with volatile demand, capacity constraints or quantity-dependent prices. Normal safety-stock calculations can understate fat-tailed disruption risk. Newsvendor assumptions fail when products can be replenished or customers substitute. MRP can generate infeasible plans when capacity and supplier constraints are ignored. Multi-echelon models can be expensive and data intensive.

Ethically, planners should avoid manipulating service measures, hiding shortages through record adjustments or forcing suppliers to absorb unpriced risk. Inventory decisions affect worker safety, product availability, environmental waste and small-supplier liquidity. Expired, hazardous and electronic materials require compliant storage, traceability and disposal. Resilience should not be confused with indiscriminate stockpiling.

Best practice is to:

1. Define service metrics and customer consequences explicitly.
2. Segment items by value, variability, criticality and risk.
3. Maintain accurate item, BOM, lead-time and transaction data.
4. Use EOQ and safety-stock equations as decision aids, not automatic answers.
5. Separate independent-demand control from dependent-demand MRP logic.
6. Reconcile physical, operational and financial records.
7. Review parameters after structural changes, not only at year-end.
8. Measure total cost, working capital, availability, obsolescence and resilience together.
9. Govern analytics and AI with validation, explainability and human accountability.

Chapter Summary

Inventory supports service, production continuity and economies of scale, but creates capital, storage, obsolescence and shortage trade-offs. Inventory can be classified by function, physical form and risk. EOQ balances ordering and holding costs under restrictive assumptions. Reorder points combine expected lead-time demand with safety stock, while service-level definitions must distinguish cycle service from fill rate. The newsvendor model addresses single-period uncertainty through the critical fractile.

ABC analysis prioritizes control, and cycle counting protects the accuracy on which all planning depends. BOMs express product structure; MRP converts the MPS into time-phased component requirements using inventory, lead-time and lot-size data. ERP integrates these decisions but cannot compensate for poor master data. Multi-echelon planning coordinates inventory across locations and exploits risk pooling. Effective practice combines quantitative models with segmentation, supplier risk management, governance, sustainability and managerial judgment.

Key Terms

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

Cycle stock: Inventory created by replenishing in batches.

Safety stock: Inventory held against uncertainty.

Pipeline inventory: Stock currently in transit.

EOQ: Order quantity minimizing ordering and holding cost under stated assumptions.

Ordering cost: Cost incurred each time an order or setup is made.

Holding cost: Cost of carrying one unit for a specified period.

Reorder point: Inventory position at which replenishment is triggered.

Inventory position: On-hand plus on-order minus backorders.

Cycle service level: Probability of no stockout during a replenishment cycle.

Fill rate: Fraction of demand units supplied immediately.

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

ABC analysis: Ranking items by annual dollar usage or related importance.

Cycle counting: Continuous selective physical counting of inventory.

BOM: Structured listing of parent-child product components.

MRP: Time-phased planning system for dependent demand.

Lot-for-lot: Lot-sizing rule ordering exactly the net requirement.

ERP: Integrated system linking operational and financial data.

Multi-echelon inventory: Coordinated inventory across several supply-chain stages.

Postponement: Delaying final differentiation until demand is clearer.

Review Questions

Short Answer Questions

1. Distinguish cycle stock from safety stock.
2. What assumptions underlie the basic EOQ model?
3. Explain the difference between cycle service level and fill rate.
4. What information is contained in a bill of materials?
5. Why is inventory accuracy essential for MRP?

Long Answer Questions

1. Explain how EOQ, reorder point and safety stock work together in a continuous-review system.
2. Compare ABC analysis and cycle counting, including their limitations.
3. Describe the inputs, calculations and outputs of an MRP system.
4. Discuss how multi-echelon inventory can reduce total safety stock while preserving service.
5. Critically evaluate the role of ERP and AI in contemporary inventory planning.

Analytical / Application Questions

1. Annual demand is 18,000 units, ordering cost is $60 per order and annual holding cost is $4 per unit. Calculate EOQ and explain two practical reasons why the adopted quantity might differ.
2. Daily demand has mean 50 units and standard deviation 10 units. Lead time is constant at four days. Using a 95% cycle service level with z = 1.645, calculate lead-time demand, safety stock and reorder point.
3. A retailer must order a seasonal product once. Selling price is $40, purchase cost is $25 and salvage value is $10. Calculate the critical fractile and explain what additional information is needed to convert it into an order quantity.

Multiple Choice Questions with Answers

Question 1

Which inventory type primarily protects against uncertainty in demand or replenishment?

A. Cycle stock

B. Safety stock

C. Pipeline inventory

D. Speculative stock

Correct Answer: B. Safety stock

Question 2

Annual demand is 10,000 units, ordering cost is $50 and annual holding cost is $2 per unit. What is the approximate EOQ?

A. 224 units

B. 500 units

C. 707 units

D. 1,000 units

Correct Answer: C. 707 units

Question 3

In a continuous-review system, inventory position normally equals:

A. On-hand inventory plus demand forecast

B. On-hand inventory plus on-order inventory minus backorders

C. Safety stock plus annual demand

D. Finished goods minus raw materials

Correct Answer: B. On-hand inventory plus on-order inventory minus backorders

Question 4

If daily demand has standard deviation σ and lead time is L constant days, the standard deviation of lead-time demand is:

A. σL

B. σ/L

C. σ√L

D. σ + L

Correct Answer: C. σ√L

Question 5

A 95% cycle service level is best interpreted as:

A. 95% of annual demand is forecast accurately

B. 95% of replenishment cycles avoid a stockout

C. 95% of inventory value is safety stock

D. 95% of orders receive a quantity discount

Correct Answer: B. 95% of replenishment cycles avoid a stockout

Question 6

In the simple newsvendor model, the critical fractile is:

A. Overage cost divided by underage cost

B. Underage cost divided by the sum of underage and overage costs

C. Selling price divided by purchase cost

D. Mean demand divided by standard deviation

Correct Answer: B. Underage cost divided by the sum of underage and overage costs

Question 7

The primary purpose of ABC analysis is to:

A. Determine the engineering structure of a product

B. Prioritize inventory control effort

C. Calculate supplier lead time

D. Replace physical counting entirely

Correct Answer: B. Prioritize inventory control effort

Question 8

Which document defines parent-child relationships and component quantities?

A. Bill of materials

B. Cash-flow statement

C. Routing variance report

D. Service-level dashboard

Correct Answer: A. Bill of materials

Question 9

MRP is most directly designed to plan:

A. Independent customer demand only

B. Dependent component demand over time

C. Marketing expenditure

D. Employee performance

Correct Answer: B. Dependent component demand over time

Question 10

A major advantage of multi-echelon inventory planning is that it:

A. Eliminates all uncertainty

B. Guarantees zero inventory

C. Coordinates stock across locations and can exploit risk pooling

D. Makes item-master accuracy unnecessary

Correct Answer: C. Coordinates stock across locations and can exploit risk pooling