Hierarchical Planning and Execution Integration

Introduction

Aggregate production planning determines broad production, workforce, inventory, subcontracting, and capacity decisions over a medium-term horizon. Its output, however, is expressed in product families and time buckets, whereas factories execute individual items, orders, operations, and material movements. Hierarchical planning is the structured process through which these different levels are connected.

The central managerial challenge is to preserve consistency while progressively increasing detail. A plan that is feasible at the aggregate level may become infeasible when translated into specific products, machines, materials, and dates. Conversely, excessive detail introduced too early creates instability, computational burden, and frequent replanning. Effective production management therefore requires a disciplined hierarchy, clear decision rights, controlled feedback, and integration with execution systems.

This chapter explains how aggregate plans are disaggregated into master production schedules (MPS), translated into material requirements planning (MRP), checked through capacity requirements planning (CRP), converted into detailed schedules, and connected with procurement, distribution, and shop-floor control.

Learning Objectives

After studying this chapter, the reader should be able to:

1. Explain the logic and levels of hierarchical production planning.
2. Disaggregate an aggregate plan into a feasible MPS.
3. Connect the MPS with MRP, CRP, procurement, distribution, and execution.
4. Calculate projected available balance and available-to-promise quantities.
5. Evaluate planning time fences, order modifiers, capacity constraints, and rescheduling rules.
6. Diagnose causes of disconnection between planning and execution.
7. Design governance and performance measures for an integrated planning system.

1. The Hierarchy of Production Decisions

Production planning is hierarchical because decisions differ in time horizon, level of aggregation, uncertainty, and reversibility.

The hierarchy is not simply a chain in which information flows downward. It is a feedback system. Actual output, supplier performance, scrap, downtime, labour availability, and customer changes flow upward and may require revisions to the MPS or aggregate plan.

A useful principle is aggregation upward and disaggregation downward. Aggregate planning groups items with similar capacity and demand characteristics. Master scheduling separates those families into specific end items. MRP separates end items into components using bills of material (BOMs). Shop-floor control separates component requirements into operations and dispatch priorities.

2. From Aggregate Plan to Master Production Schedule

2.1 The role of the MPS

The MPS states which end items will be produced, in what quantities, and in which periods. It is not necessarily a final shop-floor schedule. Rather, it is a contractual and operational bridge between market demand, aggregate capacity, and detailed material and capacity planning.

An effective MPS must satisfy three conditions:

• Demand responsiveness: it represents the best current view of customer requirements and forecast demand.
• Aggregate consistency: its quantities, when converted to product-family totals, remain consistent with the aggregate plan.
• Feasibility: it respects inventory, lead-time, material, capacity, and policy constraints.

2.2 Disaggregation procedure

A practical disaggregation procedure is as follows:

1. Define the planning calendar. Establish periods, working days, shift calendars, and the MPS horizon.
2. Identify end items. Select items that are sold, stocked, or assembled into customer orders. Avoid scheduling every low-level component directly in the MPS.
3. Translate family quantities. Allocate aggregate family production to end items using forecasts, customer orders, historical mix, contractual commitments, and strategic priorities.
4. Apply opening inventory. Determine how much existing finished-goods inventory can satisfy demand.
5. Respect lot sizes and order modifiers. Planned quantities may need to be multiples of a standard batch, minimum order quantity, or fixed order quantity.
6. Check rough-cut capacity. Compare required capacity at critical resources with available capacity.
7. Review material and supplier constraints. A nominally feasible production quantity is not feasible if a long-lead or constrained component is unavailable.
8. Validate with sales, operations, procurement, and finance. The MPS becomes authoritative only after cross-functional review.

2.3 MPS records

For each end item and period, a typical MPS record contains forecast demand, customer orders, the greater of these two demand signals where appropriate, scheduled receipts, projected available balance, MPS quantities, and available-to-promise (ATP).

A common projected available balance relationship is:

where:

• PAB_t = projected available balance at the end of period t;
• MPS_t = MPS receipt quantity in period t;
• SR_t = scheduled receipts due in period t;
• D_t = demand used for planning in period t.

The opening balance PAB₀ is known inventory or another approved starting quantity. The formula assumes that all quantities use the same unit of measure and that demand, receipts, and inventory are assigned to the same time buckets. Negative PAB indicates a projected shortage, not automatically a recommendation to create an MPS order; the planner must consider lead time, capacity, safety stock, and order policy.

Demand D_t may be defined by the organization as forecast, customer orders, or a rule such as the larger of forecast and orders. The rule must be explicit because different definitions produce different schedules.

ATP protects uncommitted supply for new customer orders. For a simple MPS environment, ATP for the first MPS period can be calculated as:

The visible control character above is undesirable; the correct expression is:

For later MPS receipts, ATP is generally:

These formulas assume that ATP is being calculated for discrete MPS receipts and that customer orders, rather than forecasts, are used for reservation. Organizations may use alternative ATP rules, including cumulative ATP or capable-to-promise (CTP), which additionally checks material and capacity feasibility.

2.4 Worked MPS example

A company begins week 1 with 80 units of finished-goods inventory. A scheduled receipt of 20 units arrives in week 2. The MPS proposes receipts of 100 units in week 2 and 120 units in week 4. Forecast demand is 60, 70, 80, and 90 units in weeks 1–4. Customer orders are 50, 55, 65, and 75 units. The planning rule uses the larger of forecast and customer orders as demand.

For week 1, PAB₁ = 80 − 60 = 20. For week 2, PAB₂ = 20 + 100 + 20 − 70 = 70. Week 3 produces a projected shortage: PAB₃ = 70 − 80 = −10. Week 4 restores positive balance: PAB₄ = −10 + 120 − 90 = 20.

The negative week-3 balance may require moving part of the week-4 MPS receipt forward, accepting a customer delay, using an alternative item, or authorizing overtime. It should not be hidden by simply changing the inventory figure.

If the next MPS receipt after week 2 is in week 4, the week-2 ATP based on customer orders is 100 + 20 − (55 + 65) = 0 units. This means all week-2 supply is already committed under the stated ATP rule.

3. Time Fences, Stability, and Change Management

A planning system must distinguish between periods in which changes are easy and periods in which changes are disruptive. The time fence divides the horizon into zones:

• Frozen zone: changes require high-level authorization because materials, labour, and schedules are already committed.
• Slushy or semi-frozen zone: limited changes are permitted through planner review.
• Liquid zone: forecasts and planned orders may be changed routinely.

Time fences should reflect actual cumulative lead times, supplier flexibility, setup requirements, customer agreements, and the economic cost of change. A fence that is shorter than the replenishment lead time creates false feasibility; a fence that is unnecessarily long reduces responsiveness.

MPS stability can be measured through schedule changes. One useful indicator is:

The measure is meaningful only when the definition of “changed” is specified—for example, quantity change above a threshold, date movement beyond one period, or cancellation. High change rates may indicate poor forecasts, inappropriate lot sizing, unreliable suppliers, insufficient capacity, or weak governance. Low change rates are not automatically good if customer service is poor or obsolete inventory is increasing.

4. MRP: Translating End Items into Materials

MRP converts the MPS into time-phased requirements for components and materials. Its essential inputs are:

1. the MPS or independent demand;
2. the bill of material, including quantities and engineering effectivity;
3. inventory records, including on-hand balances, allocations, scheduled receipts, lead times, and lot-sizing rules.

MRP performs four logical activities:

• Explosion: multiply parent requirements through the BOM.
• Netting: subtract usable inventory and scheduled receipts.
• Lot sizing: convert net requirements into planned order quantities.
• Offsetting: move planned order releases backward by lead time.

For a component with gross requirement GR_t, scheduled receipt SR_t, projected available balance before new planned receipt PAB_t−1, and safety stock SS, a simplified net requirement is:

This relationship assumes that inventory is interchangeable, all quantities use the same unit, and no additional reservations or yield losses need to be represented. In practice, allocated inventory, scrap factors, co-products, substitute materials, and effectivity dates may require more detailed logic.

For a fixed lead time L, a planned order receipt in period t requires a release in period t − L:

The expression assumes a stable and correctly measured lead time. If t − L is earlier than the current period, the requirement is already late or requires an expediting decision.

MRP example

Suppose the MPS requires 100 finished units in week 6. Each finished unit requires 2 units of component A. Thus, gross requirement for A in week 6 is 100 × 2 = 200 units. Inventory before the requirement is 60 units, a scheduled receipt of 40 units is due in week 5, and safety stock is 20 units. Net requirement is:

NR₆ = max(0, 200 + 20 − 60 − 40) = 120 units.

If A has a two-week lead time, the planned order release is 120 units in week 4. If the supplier’s minimum order is 150 units, the release becomes 150 units, subject to the organization’s approved lot-sizing policy. The excess 30 units should be visible as projected inventory rather than treated as consumed demand.

MRP is a planning system, not a guarantee that materials will arrive. Purchase orders, supplier confirmations, inspection status, and receiving transactions must be integrated with the records.

5. Capacity Requirements Planning and Feasibility

MRP produces material dates and quantities; CRP tests whether the associated operations can be performed. CRP uses planned orders, routings, operation times, work-centre calendars, efficiency, and existing load.

For an operation, required load can be expressed as:

A more precise formulation separates setup and run time by order:

where S_j is setup time for job j in hours, R_j is run time per unit in hours per unit, and Q_j is quantity processed. Load must be expressed in hours, while available capacity is also expressed in hours.

If a work centre has m identical machines, H scheduled hours per machine, and an availability factor a between 0 and 1, nominal available capacity is:

The factor may represent planned downtime, absenteeism, or expected availability. Efficiency should not be double-counted if it is already incorporated into standard times. A utilization indicator is:

Utilization above 100% signals overload under the stated assumptions; it does not prove that output is impossible, because overtime, subcontracting, alternate routings, or yield improvements may change available capacity.

CRP responses include changing the MPS, shifting orders, adding shifts, authorizing overtime, using alternate work centres, subcontracting, changing lot sizes, reducing setup time, or revising due dates. Capacity should be evaluated at bottlenecks and at strategically important resources, not only as a plant-wide average.

6. Linking Procurement and Distribution

Procurement integration

MRP planned releases become purchase requisitions, planned production orders, or transfer requirements. Procurement must evaluate supplier capacity, minimum order quantities, price breaks, transport lead times, quality requirements, and delivery reliability. A planning date is not the same as a supplier commitment date.

Supplier integration is stronger when the organization shares forecasts, time-phased requirements, order changes, and engineering revisions while defining which information is binding. Blanket orders, supplier schedules, vendor-managed inventory, and electronic confirmations can reduce transaction delay, but they require accurate master data and clear ownership.

Distribution integration

Finished-goods production should be connected to distribution requirements planning (DRP) and deployment decisions. The central question is not only “When should the plant produce?” but also “Where should available inventory be positioned?”

A plant may meet the aggregate MPS while still failing customers if inventory is deployed to the wrong regional warehouse. Distribution plans should therefore consider transportation lead times, warehouse capacity, service priorities, product shelf life, allocation rules, and available-to-promise commitments. For multi-echelon networks, production, central inventory, regional inventory, and customer demand should be evaluated as one flow rather than as isolated plans.

7. Detailed Scheduling and Shop-Floor Execution

Detailed scheduling transforms planned orders into operation sequences and dispatch lists. It must account for machine eligibility, tooling, labour skills, setup families, material availability, maintenance, due dates, and quality holds.

Common dispatching rules include:

• First come, first served: simple and transparent, but not necessarily due-date effective.
• Earliest due date: prioritizes urgent customer commitments.
• Shortest processing time: often reduces average flow time, but can delay long jobs.
• Critical ratio:

A ratio below 1 suggests that the job is behind its nominal pace. The rule assumes that “work remaining” and “time remaining” are measured consistently. No single rule is universally best; hybrid rules may be required for safety, customer priority, setup reduction, and bottleneck protection.

Execution feedback includes operation completion, scrap, rework, downtime, labour attendance, material shortages, and quality status. These transactions should update enterprise resource planning (ERP) records promptly. Otherwise, MRP will plan against fictitious inventory and the schedule will repeatedly release work that cannot be performed.

A disciplined exception process distinguishes:

• planning exceptions: forecast, inventory, or capacity assumptions require review;
• execution exceptions: actual performance deviates from the released plan;
• master-data exceptions: BOMs, routings, lead times, or calendars are incorrect;
• commercial exceptions: customer priority or promised date changes.

8. Governance and the Integrated Planning Cycle

Integration depends as much on governance as on software. A monthly or weekly integrated business planning cycle commonly includes:

1. demand review;
2. supply and capacity review;
3. inventory and financial reconciliation;
4. pre-MPS analysis;
5. MPS approval;
6. MRP and CRP regeneration;
7. procurement and distribution review;
8. execution monitoring and exception escalation.

Decision rights should be explicit. Sales may own the demand signal, operations may own production feasibility, procurement may own supplier commitments, and the supply-chain leader may arbitrate cross-functional trade-offs. Finance should ensure that volume, inventory, overtime, subcontracting, and service decisions are economically coherent.

Important measures include:

Measures should be interpreted together. Pursuing high utilization at every resource may increase queues and reduce responsiveness. Minimizing inventory without improving reliability may increase shortages. A balanced scorecard is preferable to a single efficiency target.

9. Illustrative Case: Meridian Appliances

Background

Meridian Appliances, an illustrative mid-sized manufacturer, produces three refrigerator models. Its aggregate plan calls for 12,000 units in the next quarter. The family plan was feasible in terms of total labour hours, but customers experienced late deliveries and frequent expediting.

Problem

The company’s MPS was created by copying family quantities into product-level schedules. The schedule ignored model mix, compressor availability, changeover time, and regional distribution requirements. MRP then generated purchase orders from unstable schedules. Shop-floor supervisors frequently changed priorities, while inventory records were updated only at the end of each shift.

Analysis

The company mapped the hierarchy and found four disconnections:

1. sales forecasts were not reconciled with confirmed customer orders;
2. the MPS used monthly buckets while compressor suppliers required weekly releases;
3. CRP omitted setup hours at the painting work centre;
4. finished goods were produced for the total network rather than for regional allocation needs.

A revised MPS separated the three models, used weekly buckets, applied a two-week frozen zone, and reserved constrained compressors against confirmed orders. CRP included setup time and identified painting as the bottleneck. Procurement received a time-phased component schedule, while distribution received a deployment plan by region.

Approach and outcome

The company introduced daily transaction discipline for completions, scrap, and material issues; a weekly MPS review; and an exception board for shortages and bottlenecks. It did not eliminate all changes. Instead, changes inside the frozen zone required authorization and a documented customer or economic justification.

After several planning cycles, the company reported improved schedule adherence, fewer emergency purchase orders, and better regional availability. The exact improvement depends on operating conditions and is not a universal benchmark; the important result was the removal of structural causes of instability.

Learning points

The case demonstrates that aggregate feasibility does not ensure end-item feasibility. Mix, sequence, component constraints, distribution positioning, and data timeliness are essential. Technology can expose these relationships, but governance and disciplined execution are necessary to realize the benefit.

10. Technology, AI, and Contemporary Developments

Modern ERP, advanced planning and scheduling systems, manufacturing execution systems (MES), warehouse systems, supplier portals, and industrial Internet of Things platforms can create a common flow of demand, supply, capacity, and execution data. Digital twins can test alternative MPS and capacity scenarios before implementation.

AI and machine learning may improve demand sensing, lead-time estimation, anomaly detection, predictive maintenance, and schedule recommendations. However, AI-generated recommendations should remain subject to human review, especially where customer commitments, labour agreements, safety, quality, or regulatory requirements are involved. Models can amplify biased historical priorities, react excessively to noise, or recommend infeasible schedules when master data are wrong.

Important controls include data lineage, version management, access authorization, audit trails, explainable exception logic, cybersecurity, and segregation of duties. Automation should reduce clerical delay without removing accountability for material, capacity, and customer decisions.

11. Challenges and Limitations

Hierarchical integration has several inherent limitations:

• Forecast uncertainty increases with horizon and aggregation may conceal mix risk.
• Lead times are often variable rather than fixed.
• BOMs and routings may be inaccurate or change frequently.
• Capacity is not perfectly divisible because of skills, tooling, maintenance, and sequence dependence.
• Lot sizes create inventory and timing distortions.
• Local optimization can damage total system performance.
• Excessive schedule nervousness can destabilize suppliers and production.
• A technically integrated system can still fail when incentives conflict.

Consequently, planning models should be treated as decision-support representations rather than exact descriptions of reality. Scenario analysis, robust buffers, exception management, and periodic parameter review are essential.

Best Practices

1. Maintain one agreed planning calendar and item hierarchy.
2. Reconcile aggregate family plans with end-item MPS quantities.
3. Use explicit time fences and change-authorization rules.
4. Validate BOMs, routings, lead times, calendars, and order modifiers.
5. Include setup, yield, downtime, and constrained-resource effects in capacity review.
6. Integrate supplier confirmations and distribution deployment with MRP.
7. Update execution transactions promptly and investigate data discrepancies.
8. Use exception-based management rather than rescheduling every item continuously.
9. Measure service, stability, inventory, capacity, and execution together.
10. Treat AI and automation as governed decision support, not as substitutes for accountability.

Chapter Summary

Hierarchical planning connects strategic and aggregate decisions with the operational reality of products, components, machines, suppliers, warehouses, and operators. The MPS disaggregates family plans into end-item quantities and dates. Projected available balance and ATP help reconcile supply with demand and customer commitments. MRP explodes the MPS through bills of material, nets inventory and scheduled receipts, applies lot-sizing rules, and offsets releases by lead time. CRP tests whether the resulting work can be performed at specific work centres.

Integration continues through procurement, distribution, detailed scheduling, MES transactions, and feedback from execution. Time fences, master-data accuracy, governance, and cross-functional measures are essential to control schedule nervousness and protect feasibility. Contemporary digital and AI tools can improve visibility and prediction, but they require data quality, human accountability, cybersecurity, and ethical governance. The fundamental managerial lesson is that a feasible aggregate plan is only the starting point: value is created when every planning level remains coherent with the next and responsive feedback returns from execution.

Key Terms

Aggregate production plan: Medium-term plan for product families, resources, inventory, and production rates.

Available-to-promise (ATP): Uncommitted supply that can be promised to customers under an ATP rule.

Bill of material (BOM): Structured description of components required to make a parent item.

Capacity requirements planning (CRP): Process of comparing planned workload with work-centre capacity.

Capable-to-promise (CTP): Promise calculation that checks material and capacity feasibility.

Disaggregation: Conversion of family-level plans into end-item plans.

Distribution requirements planning (DRP): Time-phased planning of inventory deployment through a distribution network.

Enterprise resource planning (ERP): Integrated system supporting cross-functional business transactions and planning.

Frozen zone: Planning interval in which changes require special authorization.

Hierarchical planning: Coordinated planning across strategic, aggregate, master, material, capacity, scheduling, and execution levels.

Master production schedule (MPS): Time-phased plan for production of specific end items.

Material requirements planning (MRP): System for calculating time-phased component requirements and planned orders.

Net requirement: Additional quantity needed after considering inventory and scheduled receipts.

Projected available balance (PAB): Expected inventory balance after planned receipts and demand.

Rough-cut capacity planning: Preliminary capacity check applied to an aggregate or master schedule.

Scheduled receipt: Supply already due or committed in a specified period.

Shop-floor control: Operational control of released jobs, priorities, progress, and exceptions.

Time fence: Boundary that governs the degree of permitted schedule change.

Work centre: Resource or group of resources performing specified operations.

Review Questions

Short Answer Questions

1. Why is the MPS considered a bridge between aggregate planning and MRP?
2. Define projected available balance and identify its principal components.
3. What is the purpose of a frozen time fence?
4. Distinguish MRP from CRP.
5. Why must execution transactions be updated promptly?

Long Answer Questions

1. Explain the complete flow from an aggregate production plan to shop-floor execution.
2. Discuss the role of BOMs, lead times, lot sizes, and scheduled receipts in MRP.
3. Explain how CRP can reveal infeasibility that is hidden in an aggregate plan.
4. Critically evaluate the advantages and risks of using AI in hierarchical production planning.
5. Develop a governance framework for an integrated MPS review process.

Analytical / Application Questions

1. Using the MPS example in this chapter, propose two alternatives for resolving the week-3 projected shortage and evaluate their operational consequences.
2. A work centre has three machines, 32 scheduled hours per machine, and an availability factor of 0.90. Calculate available hours and explain what additional information is needed before accepting a planned load of 86 hours.
3. Design an integrated response when a supplier reports that a constrained component will arrive two weeks late but a confirmed customer order is due next week. Consider MPS, MRP, CRP, procurement, distribution, and customer communication.

Multiple Choice Questions with Answers

Question 1

Which planning document specifies quantities and timing for specific end items?

A. Aggregate production plan

B. Master production schedule

C. Bill of material

D. Dispatch list

Correct Answer: B. Master production schedule

Question 2

If opening inventory is 80 units, the MPS receipt is 100 units, scheduled receipt is 20 units, and demand is 70 units, what is ending projected available balance?

A. 90 units

B. 110 units

C. 130 units

D. 170 units

Correct Answer: B. 110 units

Question 3

Which MRP activity moves a planned order receipt backward by its lead time?

A. Explosion

B. Netting

C. Offsetting

D. Allocation

Correct Answer: C. Offsetting

Question 4

A finished product requires two units of component A. An MPS quantity of 150 units creates what gross requirement for A, before other adjustments?

A. 75 units

B. 150 units

C. 152 units

D. 300 units

Correct Answer: D. 300 units

Question 5

What is the primary purpose of CRP?

A. To calculate customer forecasts

B. To compare planned workload with work-centre capacity

C. To define the product BOM

D. To record supplier invoices

Correct Answer: B. To compare planned workload with work-centre capacity

Question 6

Which zone normally permits the least schedule change without authorization?

A. Liquid zone

B. Forecast zone

C. Frozen zone

D. Planning zone

Correct Answer: C. Frozen zone

Question 7

A work centre has two machines, 40 scheduled hours per machine, and an availability factor of 0.80. Its nominal available capacity is:

A. 32 hours

B. 64 hours

C. 80 hours

D. 100 hours

Correct Answer: B. 64 hours

Question 8

Which system primarily records operational completion, downtime, and shop-floor status?

A. Manufacturing execution system

B. Customer relationship management system

C. General ledger

D. Sales forecasting worksheet

Correct Answer: A. Manufacturing execution system

Question 9

What does ATP principally identify?

A. Total plant capacity

B. Uncommitted supply available for customer promises

C. Total BOM component count

D. Historical supplier quality

Correct Answer: B. Uncommitted supply available for customer promises

Question 10

Why can a feasible aggregate plan become infeasible after disaggregation?

A. Aggregation always eliminates inventory

B. Product mix, setups, materials, and bottlenecks may be hidden at family level

C. The MPS does not use time periods

D. MRP ignores all customer orders

Correct Answer: B. Product mix, setups, materials, and bottlenecks may be hidden at family level