Dependent-Demand Inventory and Production Planning

Introduction

Inventory control differs fundamentally according to the nature of demand. Independent demand arises outside the firm—for example, demand for finished bicycles, refrigerators, or replacement parts. Dependent demand is derived from the production requirements of another item. If one bicycle requires two wheels, demand for wheels depends on the planned production of bicycles. This relationship makes dependent-demand inventory control a planning problem rather than merely a stock-replenishment problem.

Material Requirements Planning (MRP) provides the traditional logic for converting a production schedule into time-phased requirements for components, subassemblies, raw materials, and work-in-process (WIP). Modern Enterprise Resource Planning (ERP) systems extend this logic by integrating purchasing, production, inventory, finance, quality, maintenance, and supply-chain information.

The effectiveness of MRP depends less on computational power than on the quality of its inputs. An inaccurate bill of material (BOM), unrealistic lead time, incorrect inventory balance, or obsolete planning parameter can generate systematic shortages, excess inventory, expediting, and schedule instability. Consequently, dependent-demand control requires both technical models and managerial discipline.

Learning Objectives

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

1. Distinguish independent demand from dependent demand.
2. Construct and interpret a multilevel bill of material.
3. Explain the logic of MRP records and time-phased netting.
4. Apply lot-sizing rules and evaluate their effects on inventory and capacity.
5. Incorporate lead time, safety lead time, safety stock, and release timing.
6. Recognize the relationship between MRP, capacity planning, and shop-floor execution.
7. Identify the ERP master data required for reliable planning.
8. Diagnose common causes of MRP nervousness and recommend governance controls.

1. Dependent Demand as a Planning Relationship

Dependent demand is calculated from a known or forecast requirement for a higher-level item. The demand for a component is therefore not normally forecast independently. Forecasting a component separately can create double counting and inappropriate safety stock, particularly when the component is used in several products.

The basic relationship is:

where:

• D_c = gross demand for component c during a planning period;
• D_p = planned production quantity of parent item p;
• q_cp = quantity of component c required per unit of parent p;
• the summation covers all relevant parent items.

This equation assumes that the BOM is accurate, usage quantities are known, and the parent production plan is feasible. Scrap, yield loss, substitutes, and engineering changes require additional adjustments.

Independent-demand methods such as the economic order quantity (EOQ) model remain useful for finished goods and some service parts. For dependent-demand components, however, the timing and quantity of requirements are driven primarily by the master production schedule (MPS), BOM, inventory records, and lead times.

Important terminology

2. Bills of Material

A bill of material is a structured product definition showing the components, quantities, and relationships required to produce an item. It is both an engineering document and a planning master-data object.

A single-level BOM lists the immediate components of a parent. A multilevel BOM represents the complete product structure. For example:

Figure: A multilevel BOM showing parent–component relationships.

In this example, one unit of P requires one A and two C. Each A requires three B and two D. Thus, one P requires three B and two D indirectly. BOM accuracy depends on correct item identification, quantity per, units of measure, effectivity dates, scrap factors, phantom assemblies, substitutes, and revision status.

BOM structures and planning implications

A manufacturing BOM describes how an item is built. An engineering BOM emphasizes design relationships and may not correspond directly to production stages. A planning BOM may group options or represent a product family for aggregate planning. A phantom assembly is logically present in the BOM but is not normally stocked; its components are exploded directly when the parent is planned. A configurable BOM supports products assembled from customer-selected options, but requires valid configuration rules.

Common BOM errors include omitted components, duplicated requirements, incorrect quantity-per values, obsolete revisions, and failure to record yield loss. Governance therefore requires controlled engineering-change procedures, revision approval, cross-functional review, and periodic audit against actual shop-floor consumption.

3. Master Production Scheduling and MRP Logic

The MPS states when and how many end items are to be produced. It translates aggregate plans, sales orders, forecasts, available capacity, and inventory policy into a time-phased schedule. MRP then explodes the MPS through the BOM and determines when component supply must be received or released.

For each item and period, the basic MRP record contains:

1. gross requirements;
2. scheduled receipts from existing purchase or production orders;
3. projected available balance;
4. net requirements;
5. planned order receipts; and
6. planned order releases.

A simplified projected-available calculation is:

where PAB_t is projected available balance at the end of period t, SR_t is the scheduled receipt, POR_t is the planned order receipt, and GR_t is the gross requirement. Beginning inventory is represented by PAB₀. The equation assumes that all receipts are usable in the stated period and that requirements occur at the period level.

If projected available inventory would fall below the permitted level, the system creates a net requirement. Under a simple lot-for-lot rule:

where NR_t is net requirement and SS_t is the desired safety stock. If a fixed lot size Q applies, the planned receipt may be:

The ceiling function rounds upward to the next whole lot. This rule is appropriate only when fixed-lot purchasing or production is required. It can create excess inventory, so the economic and capacity consequences must be evaluated.

The planned release is obtained by offsetting the receipt by the item's lead time. If lead time is two periods:

This simple representation assumes deterministic lead time. In practice, queue time, inspection, transport, setup, and administrative approval may make actual lead time variable.

Worked example: single-level MRP

Suppose an assembly requires 100 units in week 4. The component has beginning inventory of 20 units, no scheduled receipts, a safety stock requirement of 10 units, and a fixed lot size of 50 units. The component lead time is one week.

Gross requirement in week 4 = 100 units.

Net requirement:

The fixed-lot planned receipt is:

The planned order release is therefore 100 units in week 3. Ending projected inventory in week 4 is:

The ending balance exceeds the safety stock by 10 units because the lot size forced a 100-unit receipt rather than a 90-unit receipt. This is a normal consequence of fixed-lot logic, not necessarily an error.

4. Lot-Sizing Rules

Lot sizing determines the quantity produced or purchased when a net requirement occurs. No single rule is universally superior; the appropriate choice depends on setup costs, holding costs, supplier constraints, shelf life, capacity, price breaks, and service requirements.

Lot-for-lot

Lot-for-lot sets the planned receipt equal to the net requirement. It minimizes planned leftover inventory and is attractive for expensive, perishable, or rapidly changing items. However, it may create frequent setups, small supplier orders, and unstable workloads.

Fixed order quantity

A fixed quantity is ordered whenever a shortage occurs. Examples include supplier minimum order quantities, standard container quantities, and economic batch sizes. Fixed quantities simplify operations but may create inventory peaks.

Period-order quantity

Period-order quantity groups requirements over a selected number of periods. For example, a two-week rule combines requirements in the current and next week. It reduces order frequency but increases average inventory and may obscure capacity peaks.

Least total cost and least unit cost

These heuristic methods compare candidate order groupings by balancing setup and holding costs. They can be useful when demand is uneven, but their results depend strongly on cost estimates and the planning horizon.

Part-period balancing

Part-period methods compare the setup cost with cumulative holding cost. A part-period is one unit held for one period. If setup cost is S and holding cost per unit per period is h, the approximate economic number of part-periods is:

where EPP is the economic part-period target. This is a heuristic, not a universal optimum. It assumes that setup and holding costs are reasonably stable and that requirements are known over the horizon.

Lot-sizing evaluation

Managers should compare alternatives using total relevant cost, service performance, capacity effects, and operational complexity. A low material cost does not imply a good lot size if it generates excessive WIP, overtime, obsolescence, or congestion. Conversely, a mathematically small lot may be impractical when setup times are long or suppliers impose minimum quantities.

5. Lead Time, Safety Stock, and Safety Lead Time

Lead time is the elapsed time between releasing an order and receiving usable material. MRP lead time may include purchasing, manufacturing, queue, inspection, transport, and put-away time. Using only processing time produces systematically late orders.

Safety stock protects against uncertainty in demand, supply, or process yield. Safety lead time protects by releasing or scheduling an order earlier than the nominal requirement date. The two policies are not identical:

• safety stock absorbs quantity variation;
• safety lead time absorbs timing variation;
• using both without analysis can overprotect the system and inflate inventory.

For a simple continuous-review model with variable demand during lead time, a common safety-stock expression is:

where SS is safety stock in units, z is the desired service-factor quantile, and σ_LT is the standard deviation of demand during lead time. If demand per period has standard deviation σ_d and lead time is a constant L periods, then:

These formulas assume approximately independent demand across periods, stable statistical behavior, and a service definition consistent with the selected z. If both demand and lead time vary, a more complete model is required; blindly applying the constant-lead-time formula understates uncertainty.

Safety lead time is often implemented as an offset in the planning calendar. For example, if a component is required in week 8 and a two-week safety lead time is applied, the planned release may be moved from week 6 to week 4. This may improve delivery reliability, but it also increases WIP and can conceal poor supplier or process performance. A sound policy uses measured lead-time variability and reviews the cause of lateness rather than permanently advancing every order.

6. Capacity Considerations and MRP-II

Traditional MRP answers “what” and “when” but does not automatically prove that the plan is feasible. Capacity Requirements Planning (CRP) converts planned and released orders into capacity requirements by work centre, machine group, labor category, or supplier resource.

For operation o, a basic load calculation is:

where Q_i is the quantity of item i processed, t_io is processing time per unit at operation o, and s_io is setup time for each applicable batch. Units must be consistent, such as machine-hours.

Capacity utilization may be expressed as:

Available capacity should reflect working time, planned downtime, breaks, maintenance, absenteeism, and an appropriate efficiency factor. A utilization value above 100% indicates overload under the stated assumptions; a value near 100% may still be risky because variability leaves little recovery capacity.

MRP-II extends material planning by integrating sales and operations planning, MPS, MRP, CRP, purchasing, and feedback from execution. Closed-loop planning compares planned dates and quantities with actual performance. If a work centre is overloaded, managers may revise the MPS, adjust lot sizes, use alternate resources, subcontract, authorize overtime, change routing, or negotiate delivery dates. Material availability alone cannot justify a production promise.

7. MRP Nervousness and Planning Stability

MRP nervousness is excessive change in planned orders caused by small changes in demand, inventory, lead time, or dates. It can produce repeated rescheduling, supplier confusion, premature releases, and loss of planner confidence.

Causes include inaccurate forecasts, unstable MPS, short planning fences, poor inventory records, unreliable lead times, order cancellations, and inappropriate lot-sizing rules. Control methods include:

• a frozen or firm zone near the execution date;
• time fences that limit automatic changes;
• exception messages rather than indiscriminate rescheduling;
• order modifiers and minimum-change quantities;
• accurate reporting of completions, scrap, and receipts;
• supplier communication of revision status;
• root-cause analysis of repeated changes.

A time fence should not merely suppress information. It should identify which changes require managerial approval and which can be accepted automatically. Excessive firmness can preserve an infeasible plan; excessive flexibility can destabilize execution.

8. ERP Master Data and Governance

ERP planning quality is determined by the integrity of master and transactional data. Essential master-data fields include:

The distinction between on-hand inventory and available inventory is critical. Material may be physically present but unavailable because it is allocated, quarantined, expired, damaged, or awaiting quality release. Likewise, a scheduled receipt should not be treated as reliable if its supplier date is unrealistic.

Data governance requires ownership, approval workflows, audit trails, version control, periodic parameter review, and reconciliation between system balances and physical counts. Cycle counting should be risk-based: high-value, high-usage, safety-critical, and shortage-prone items generally warrant greater attention.

9. Practical Implementation Framework

A disciplined implementation can proceed through the following sequence:

1. Define the planning scope. Identify plants, products, inventory types, planning horizons, and service objectives.
2. Stabilize the product definition. Validate BOMs, units, revisions, effectivity, scrap, and substitutes.
3. Establish a credible MPS. Reconcile demand, supply, inventory, and aggregate capacity.
4. Clean item and supplier data. Correct lead times, order multiples, make/buy codes, and status fields.
5. Run MRP and review exceptions. Focus on shortages, excesses, past-due orders, reschedules, and capacity conflicts.
6. Validate with CRP. Check work-centre, labor, tooling, and supplier constraints.
7. Release selectively. Convert plans into purchase and production orders only after feasibility and priority review.
8. Measure execution. Track schedule adherence, inventory accuracy, supplier reliability, shortage frequency, WIP, and parameter stability.
9. Improve causes, not symptoms. Investigate recurring shortages, expediting, scrap, late receipts, and order changes.

Useful performance measures include inventory record accuracy, schedule adherence, planned-order stability, shortage hours, supplier on-time delivery, WIP days, inventory turns, and percentage of orders completed without expedite action. No single measure is sufficient. High inventory turns accompanied by chronic shortages may indicate understocking rather than excellence.

10. Illustrative Case: Apex Medical Device Assembly

Background

Apex Medical Devices assembles a portable monitoring unit. Each finished unit requires one sensor module, one control board, one housing, and four fasteners. The sensor module itself requires one calibrated sensor and one connector. Demand is driven by a monthly hospital-equipment schedule, while components have different procurement and production lead times.

Problem

Apex experienced frequent shortages of connectors and excessive inventory of control boards. Planners had independently forecast several components, engineering revisions were not consistently reflected in the BOM, and the ERP system used supplier quoted lead times rather than measured receipt lead times. Expedites increased, and WIP accumulated before the final test area.

Analysis

A cross-functional team audited the BOM and discovered that the connector quantity had been entered as two instead of one in one product revision, while an obsolete control-board revision remained active for a subset of orders. Inventory records included quarantined sensors as available stock. CRP analysis showed that final test capacity was the bottleneck, so advancing all component releases merely increased WIP without increasing shipments.

The team rebuilt the MPS from confirmed customer requirements, froze the near-term schedule, and recalculated component demand through the approved multilevel BOM. Connector planning used a supplier minimum order quantity, while sensors used lot-for-lot planning because of high value and obsolescence risk. Safety stock was established only for items with measured demand or supply variability. Safety lead time was applied temporarily to connectors while supplier performance was improved.

Solution and outcome

ERP master data were placed under engineering and operations ownership. BOM revisions required effectivity dates and approval. Quarantined and allocated inventory were excluded from available balance. Supplier lead times were based on actual receipt history and reviewed monthly. The MPS and CRP review became a weekly cross-functional process, with a daily exception review for critical shortages.

After implementation, connector shortages declined, obsolete control-board purchases stopped, and WIP before final test was reduced. The improvement did not result from increasing every inventory target. It resulted from aligning product structure, demand translation, capacity, data governance, and execution discipline.

Learning points

1. MRP cannot correct an incorrect BOM.
2. Material acceleration does not solve a capacity bottleneck.
3. Safety stock and safety lead time should be based on the type of uncertainty.
4. ERP data ownership must be explicit.
5. Performance measures should include stability and service, not inventory alone.

11. Limitations and Contemporary Developments

MRP commonly assumes that the BOM is deterministic, requirements are known, lead times are sufficiently stable, and capacity can be adjusted or verified separately. These assumptions become weak in engineer-to-order environments, highly variable service operations, remanufacturing, products with uncertain yields, and supply networks affected by disruption.

Advanced planning and scheduling systems add finite-capacity scheduling, optimization, constraint-based sequencing, and scenario analysis. Cloud ERP platforms improve collaboration and event visibility, while IoT data can provide real-time production progress and equipment status. Machine-learning methods may improve lead-time prediction, demand sensing, anomaly detection, and parameter recommendations. However, predictive accuracy does not eliminate the need for causal understanding, clean master data, human oversight, and clear accountability.

AI-enabled planning also raises governance issues. Planners should be able to inspect the data and logic behind recommendations, distinguish predictions from decisions, and monitor bias toward suppliers, products, or customers. Automated rescheduling should have approval thresholds for safety-critical, regulated, or financially material items. Cybersecurity, access control, auditability, and segregation of duties are essential because ERP changes can affect procurement, production, financial commitments, and product compliance.

For regulated products, traceability of revisions, lots, serial numbers, approvals, and deviations may be mandatory. Ethical planning also requires avoiding systematic concealment of shortages, unsafe substitution, excessive supplier pressure, or unrealistic labor expectations. A technically efficient plan is not acceptable if it compromises product safety, worker welfare, legal compliance, or truthful customer communication.

Best-Practice Checklist

• Maintain one approved source of product structure.
• Use effectivity dates and revision control for every engineering change.
• Separate gross, allocated, blocked, inspection, and available inventory.
• Measure actual lead times and variability rather than relying only on standard values.
• Select lot-sizing rules according to cost, capacity, service, and obsolescence.
• Validate MRP plans through CRP and supplier-capacity review.
• Protect near-term execution with rational time fences.
• Use exception management instead of reviewing every planned order equally.
• Reconcile physical inventory and ERP records continuously.
• Treat AI recommendations as governed decision support, not unquestionable instructions.

Chapter Summary

Dependent-demand inventory is derived from the production requirements of higher-level items. A reliable planning system begins with an accurate multilevel BOM and a feasible MPS. MRP explodes the MPS, nets gross requirements against available supply, applies lot-sizing rules, and offsets receipts by lead time to create planned releases. Lot-for-lot, fixed quantity, period-order, and cost-balancing methods produce different trade-offs among setups, holding cost, capacity, and responsiveness.

Safety stock protects against quantity uncertainty, whereas safety lead time protects against timing uncertainty. Both must be based on measured variation and should not be applied indiscriminately. MRP plans must be checked against finite or constrained capacity through CRP and closed-loop MRP-II processes. ERP effectiveness depends on accurate item, BOM, routing, supplier, inventory, and planning-parameter data supported by ownership and auditability.

The central managerial lesson is that component availability, production capacity, data quality, and execution discipline are interdependent. Modern analytics and AI can improve prediction and visibility, but they cannot replace sound product governance, validated assumptions, ethical controls, and informed managerial judgment.

Key Terms

Dependent demand: Demand derived from the planned or actual requirements of another item.

Independent demand: Demand arising from external customers or markets.

Bill of material: Structured record of components and quantities required to make an item.

Multilevel BOM: BOM showing relationships across several product levels.

Master production schedule: Time-phased plan for producing end items.

Material Requirements Planning: Systematic calculation of component requirements and order timing.

Gross requirement: Total requirement before available supply is deducted.

Net requirement: Requirement remaining after inventory and scheduled receipts are considered.

Planned order receipt: Quantity planned to become available in a period.

Planned order release: Order start date after lead-time offsetting.

Lot sizing: Determination of the quantity produced or purchased for an order.

Safety stock: Inventory held to protect against uncertainty.

Safety lead time: Advance scheduling allowance used to protect against timing variation.

Capacity Requirements Planning: Process of comparing planned workload with available capacity.

MRP nervousness: Excessive changes in planned orders caused by instability or poor parameters.

Pegging: Tracing a component requirement to its originating parent demand.

Phantom assembly: Non-stocked logical assembly whose components are planned directly.

ERP master data: Core item, BOM, routing, supplier, inventory, and planning records used by an ERP system.

Time fence: Planning boundary restricting or controlling automatic schedule changes.

Review Questions

Short Answer Questions

1. Why is component demand generally not forecast independently in an MRP environment?
2. What is the difference between a planned order receipt and a planned order release?
3. State two differences between safety stock and safety lead time.
4. Why must MRP results be checked through capacity requirements planning?
5. What types of information are contained in an item master and a BOM?

Long Answer Questions

1. Explain the complete flow from the master production schedule to component order release in a multilevel MRP system.
2. Compare lot-for-lot, fixed-order-quantity, period-order-quantity, and part-period balancing methods.
3. Discuss the causes and control mechanisms of MRP nervousness.
4. Explain how inaccurate BOMs, lead times, and inventory records affect production performance.
5. Evaluate the role of ERP, analytics, and AI in dependent-demand planning, including governance risks.

Analytical / Application Questions

1. A parent requires 120 units of a component in week 5. Beginning inventory is 25 units, scheduled receipts are 30 units in week 4, safety stock is 15 units, and the fixed lot size is 50 units. Calculate the net requirement, planned receipt, and release week if lead time is two weeks.
2. A work centre has 160 available hours in a week. Three planned batches require 45, 55, and 70 hours, including setup. Calculate utilization and recommend two possible managerial responses.
3. A firm has repeated shortages despite high total inventory. Design a diagnostic investigation covering BOM accuracy, inventory status, lead times, capacity, lot sizing, and ERP governance.

Multiple Choice Questions with Answers

Question 1

Which statement best describes dependent demand?

A. Demand generated directly by end customers for finished goods

B. Demand derived from the planned requirements of a higher-level item

C. Demand that is always random and normally distributed

D. Demand that can be controlled only through safety stock

Correct Answer: B. Demand derived from the planned requirements of a higher-level item

Question 2

An item has beginning inventory of 20 units, a gross requirement of 100 units, no scheduled receipt, and safety stock of 10 units. Under lot-for-lot planning, what is the planned receipt?

A. 70 units

B. 80 units

C. 90 units

D. 110 units

Correct Answer: C. 90 units

Question 3

A component has a net requirement of 90 units and a fixed lot size of 50 units. What planned receipt does the fixed-lot rule generate?

A. 50 units

B. 90 units

C. 100 units

D. 140 units

Correct Answer: C. 100 units

Question 4

If a planned receipt is required in week 8 and the item lead time is two weeks, when should the planned order be released?

A. Week 5

B. Week 6

C. Week 7

D. Week 10

Correct Answer: B. Week 6

Question 5

What is the primary purpose of a multilevel BOM?

A. To record only supplier prices

B. To show relationships and quantities across product levels

C. To calculate financial depreciation

D. To replace capacity planning

Correct Answer: B. To show relationships and quantities across product levels

Question 6

Which policy primarily protects against timing uncertainty?

A. Safety lead time

B. Economic order quantity only

C. Inventory write-off

D. Phantom assembly

Correct Answer: A. Safety lead time

Question 7

A work centre has 160 available hours and a planned load of 170 hours. What is its utilization?

A. 94.1%

B. 100.0%

C. 106.25%

D. 330.0%

Correct Answer: C. 106.25%

Question 8

Which condition is most likely to cause MRP nervousness?

A. A stable MPS and accurate inventory records

B. Frequent changes to demand, dates, and lead times

C. A validated BOM with controlled revisions

D. A frozen near-term planning zone

Correct Answer: B. Frequent changes to demand, dates, and lead times

Question 9

What does pegging accomplish in an MRP system?

A. It converts all demand into independent demand

B. It traces a component requirement to its parent demand

C. It eliminates the need for inventory records

D. It guarantees supplier capacity

Correct Answer: B. It traces a component requirement to its parent demand

Question 10

Which is an appropriate ERP master-data governance practice?

A. Allowing uncontrolled BOM changes by any user

B. Treating quarantined inventory as freely available

C. Using revision control, effectivity dates, and approval workflows

D. Using quoted lead times permanently without performance review

Correct Answer: C. Using revision control, effectivity dates, and approval workflows