Aggregate Planning, Scheduling and Workforce Operations

Introduction

Operations systems convert resources into goods and services over time. Their performance depends not only on process design and capacity, but also on deciding how much to produce, when to produce it, which jobs to process first, and how to deploy people and equipment. Aggregate planning, master scheduling, workforce planning and detailed scheduling connect strategic capacity decisions with day-to-day execution.

Aggregate planning normally covers a medium-term horizon, often three to eighteen months. It uses product families, labour categories and broad capacity measures rather than individual orders. The master production schedule (MPS) translates the aggregate plan into end-item quantities and dates. Detailed scheduling assigns jobs to specific machines, employees and time periods. Workforce operations ensure that the required skills, staffing levels, working patterns and employment arrangements are available.

These decisions are interdependent. A schedule that ignores labour availability is infeasible; a workforce plan that ignores demand variability creates excessive cost; and an aggregate plan that ignores bottlenecks may promise output that the system cannot deliver.

Learning Objectives

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

1. Explain the purpose and hierarchy of aggregate planning, master scheduling and detailed scheduling.
2. Formulate and evaluate level, chase and mixed aggregate plans.
3. Prepare a master schedule using demand, inventory and available capacity information.
4. Apply sequencing rules, Gantt charts and finite-capacity scheduling.
5. Explain the Theory of Constraints and use bottleneck logic in scheduling decisions.
6. Develop workforce strategies that balance cost, flexibility, capability and employee well-being.
7. Distinguish scheduling approaches for manufacturing, services and projects.
8. Evaluate technological, ethical and operational risks in workforce and scheduling decisions.

1. The Planning and Scheduling Hierarchy

Operations planning is commonly organized into three linked levels:

Aggregate planning reconciles forecast demand with available capacity. Its outputs may include regular-time production, overtime, subcontracting, inventory, backorders and workforce changes. The MPS then specifies what is to be produced or delivered. Material requirements planning may use the MPS to calculate component requirements, while shop-floor scheduling determines the exact sequence of operations.

A useful planning principle is progressive aggregation: decisions become more detailed as the time of execution approaches. Earlier decisions should preserve flexibility because forecasts are less accurate at long horizons. Near-term decisions require stronger commitment because materials, employees and customer promises have already been arranged.

2. Aggregate Planning

2.1 Demand, capacity and decision variables

An aggregate plan uses forecasts of demand for a product family or service category. For period t, common variables include:

• D_t: forecast demand in units;
• P_t: regular-time production;
• O_t: overtime production;
• S_t: subcontracted production;
• I_t: ending inventory;
• B_t: ending backorders;
• W_t: workforce level;
• H_t and F_t: hires and separations.

The fundamental inventory balance is:

Here, all quantities must be expressed in the same units and period. The equation assumes that production, demand and inventory are measured consistently and that spoilage, yield losses and returns are either negligible or incorporated into the relevant terms. If backorders are not permitted, B_t = 0 for every period.

A simplified total-cost objective is:

where the cost coefficients represent regular production, overtime, subcontracting, inventory holding, backorders, hiring and separation costs. This objective is useful when costs can be estimated reliably. In practice, managers also consider service, quality, employee relations, risk and strategic capability.

Typical constraints include:

where C_t^R and C_t^O are regular and overtime capacities. If each worker provides h productive hours per period and each unit requires a labour hours, labour-based capacity is approximately:

This relationship assumes stable productivity, sufficient materials and no other binding constraint. It should not be used as the sole capacity measure where machines, skills, space or regulatory requirements are limiting.

2.2 Main aggregate-planning strategies

Level strategy. The organization maintains a relatively stable workforce and production rate. Demand variation is absorbed through inventory, backorders, overtime or temporary capacity. Level plans support stable employment and learning, but can create high inventory or service problems when demand falls sharply.

Chase strategy. Production and workforce levels are adjusted to follow demand. Inventory is usually low, but frequent hiring, layoffs, transfers or schedule changes can increase cost, reduce morale and damage capability.

Mixed strategy. Most organizations use a combination: stable core employment, overtime during peaks, temporary workers, limited inventory and selective subcontracting. Mixed strategies are often more realistic because they recognize that labour and capacity are only partly flexible.

2.3 Worked example: evaluating a simple plan

An organization forecasts demand of 900, 1,200 and 1,000 units in three months. It begins with 200 units of inventory. Regular capacity is 1,000 units per month. A level plan produces 1,000 units each month, and ending inventory is required to be non-negative.

The inventory calculation is:

The plan is feasible under the stated assumptions and ends with 100 units. Total available units are 200 + 3(1,000) = 3,200, while total demand is 900 + 1,200 + 1,000 = 3,100. The difference is the final inventory of 100 units.

If the organization instead used a chase plan with no inventory and no backlog, production would need to equal monthly demand: 900, 1,200 and 1,000 units. The second month would exceed regular capacity by 200 units and therefore require overtime, subcontracting or a different workforce arrangement.

2.4 Aggregate planning process

A professional planning cycle generally includes:

1. Develop a demand forecast and identify uncertainty.
2. Aggregate products into families with similar capacity requirements.
3. Audit opening inventory, workforce, suppliers and bottlenecks.
4. Identify feasible alternatives: regular time, overtime, inventory, subcontracting and demand management.
5. Compare plans using cost, service, risk, quality and workforce criteria.
6. Test sensitivity to forecast error and capacity disruption.
7. Approve the plan and review it at defined intervals.

Sales and operations planning (S&OP) provides a governance process for this cycle. Sales, operations, finance, procurement and human resources jointly review assumptions and agree on one integrated plan. The purpose is not to produce a mathematically perfect forecast, but to create coordinated decisions and visible trade-offs.

3. Master Scheduling

The MPS converts an aggregate family plan into specific end items. It states the quantity to be completed in each time bucket and supports customer-promise decisions, material planning and capacity checks.

A simplified projected-available-balance calculation is:

where PAB_t is projected available balance and MPS_t is the scheduled receipt or production quantity. In an environment with known customer orders, gross requirements may include both forecast and orders, but the exact treatment depends on the organization’s forecast-consumption rule. The MPS must also respect lot sizes, lead times, due dates, safety stock and capacity.

A schedule is not feasible merely because total weekly output equals total weekly demand. Capacity must be checked by work centre and time period. For example, an MPS requiring 500 hours at a work centre with only 420 available hours is infeasible, even if other resources are idle.

Important MPS principles include:

• use time fences to distinguish frozen, slushy and liquid planning zones;
• avoid unnecessary nervousness by limiting frequent schedule changes;
• maintain visibility of available-to-promise inventory;
• check critical materials and bottleneck capacity before promising dates;
• use exception reports rather than manually inspecting every item.

4. Scheduling Fundamentals

Scheduling assigns activities to resources and time. A schedule must address four questions: what job is processed, on which resource, at what time, and in what sequence?

4.1 Gantt charts

A Gantt chart displays activities against time. It can show planned start and finish, actual progress, waiting time, setup and machine availability. Gantt charts are easy to communicate, but they can conceal resource conflicts unless all relevant resources are displayed.

Figure: A simple Gantt chart showing the planned timing of three jobs.

4.2 Sequencing rules

Sequencing rules are dispatching priorities used when several jobs compete for a resource.

• First come, first served (FCFS): processes jobs in arrival order. It is simple and perceived as fair, but may produce long waiting times.
• Shortest processing time (SPT): processes the shortest job first. It minimizes average flow time in a single-machine setting when all jobs are available at time zero and no interruptions or priority constraints apply. It may delay long jobs.
• Earliest due date (EDD): processes the job with the nearest due date. It is useful for reducing lateness risk, particularly when due dates are important.
• Critical ratio (CR):

The numerator and denominator must use the same time unit. A value below 1 indicates that the job is behind the time required to meet its due date, assuming no disruption. The lowest CR is often given priority, but the rule must be recalculated as conditions change.

No sequencing rule is universally best. SPT may improve average flow time while EDD improves due-date performance. Safety, setup families, customer priority, material availability and skill restrictions may override a simple rule.

4.3 Worked sequencing example

Four jobs are available at time zero on one machine.

SPT gives the sequence B–D–C–A. Completion times are 2, 5, 9 and 15 hours. EDD gives B–C–A–D, with completion times 2, 6, 12 and 15 hours. The choice depends on the performance objective. SPT produces a lower average completion time: (2+5+9+15)/4 = 7.75 hours. EDD produces lateness values of −5, −4, 0 and 0 hours, so no job is late. This illustrates why managers should select a rule based on the operational objective rather than habit.

5. Finite-Capacity Scheduling

Infinite-capacity planning assumes that required work can be loaded into a period regardless of actual resource limits. Finite-capacity scheduling explicitly prevents a resource from being assigned more work than it can perform.

For resource r in period t:

where a_jr is the amount of resource r required by job j, x_jrt is the assigned quantity or processing decision, and C_rt is available capacity. The units must match, such as machine-hours or qualified labour-hours.

Finite-capacity scheduling must incorporate:

• setup and changeover time;
• maintenance and planned downtime;
• skill and certification restrictions;
• material and tooling availability;
• sequence-dependent processing times;
• transportation and queue time;
• breaks, shifts and legal working-time limits.

A practical schedule is therefore an iterative process. The planner loads work, identifies overloads, changes sequence or timing, and checks whether the revised plan creates another conflict elsewhere. Advanced systems may use constraint programming, mixed-integer optimization, simulation or heuristic search. These tools support decisions but do not remove the need for accurate data and managerial judgement.

6. Theory of Constraints and Bottleneck Scheduling

The Theory of Constraints (TOC) argues that system performance is limited by at least one constraint. The constraint may be a machine, labour skill, policy, supplier, market or information process. Improving non-constraints while the bottleneck remains unchanged may increase local efficiency without increasing total throughput.

The five focusing steps are:

1. Identify the system constraint.
2. Exploit it: use existing capacity effectively, reducing avoidable downtime, poor quality and unnecessary setups.
3. Subordinate other activities to the constraint’s requirements.
4. Elevate the constraint by adding capacity, technology, shifts, outsourcing or redesign.
5. If the constraint moves, return to step one.

TOC scheduling commonly uses a drum-buffer-rope logic. The drum is the pace set by the constraint. The buffer protects the constraint from starvation and protects customer delivery from variation. The rope authorizes upstream release in line with the constraint’s consumption, limiting excess work-in-process.

Throughput accounting distinguishes three central measures:

• throughput: the rate at which the system generates money through sales;
• inventory: money invested in items intended for sale;
• operating expense: money spent to convert inventory into throughput.

TOC is powerful because it directs attention to system flow. Its limitations include difficulty identifying multiple interacting constraints, risk of oversimplifying financial decisions and the need to integrate it with quality, safety and workforce considerations.

7. Workforce Operations

Workforce operations concern the number, skills, deployment, motivation and protection of employees needed to execute the plan. A workforce strategy should balance flexibility and stability rather than treat labour as a purely variable cost.

7.1 Workforce options

Common options include:

• permanent core employees;
• overtime and shift extensions;
• part-time or temporary employees;
• cross-training and multiskilling;
• flexible shifts and annualized hours;
• subcontracting or managed services;
• redeployment between departments;
• automation and job redesign.

Cross-training increases flexibility, but it requires investment and may reduce short-term productivity while employees learn. Overtime can respond quickly to peaks, but excessive overtime increases fatigue, error, absenteeism and safety risk. Subcontracting can provide capacity, but may create quality, confidentiality, dependency and labour-standard concerns.

7.2 Workforce capacity calculation

Suppose 20 employees work 160 paid hours per month. If planned productive availability is 85%, effective hours are:

If each unit requires 2.4 labour-hours, theoretical labour capacity is:

This is an estimate, not a guaranteed output. It assumes the required skills, materials, equipment and quality levels are available. Capacity should be reduced where training, meetings, maintenance, absence or legal restrictions are not already reflected in the availability factor.

7.3 Workforce governance and ethics

Scheduling decisions affect income, family responsibilities, health and dignity. Good practice includes transparent rules for overtime allocation, adequate notice of shift changes, compliance with employment law, rest periods, non-discrimination and protection of personal data. Algorithmic scheduling systems require governance because optimization may unintentionally penalize particular workers, conceal bias or make decisions that employees cannot challenge.

Managers should monitor not only labour cost and utilization but also injury rates, fatigue indicators, turnover, absenteeism, schedule stability, training completion and employee voice. A plan that achieves output by creating unsafe or inequitable conditions is operationally unsustainable.

8. Scheduling in Manufacturing, Services and Projects

Manufacturing

Manufacturing schedules emphasize routing, machines, materials, setups, batch sizes and work-in-process. Job-shop environments need flexible dispatching, while repetitive lines rely more on takt-based balancing and sequence control. Takt time is the available production time divided by customer demand for the period:

It is a demand-pacing measure, not automatically the processing time of every operation. If a workstation’s cycle time exceeds takt time, the line requires redesign, parallel capacity, task redistribution or demand adjustment.

Services

Services cannot generally be inventoried in the same way as physical products. Capacity and demand must therefore be matched through appointments, queue design, staffing, self-service, pricing, reservations and priority rules. Examples include nurses assigned to patient demand, contact-centre agents scheduled against interval forecasts and airport staff deployed according to flight banks.

Service schedules must recognize variability in arrival and service times. A schedule based only on average demand may fail during peaks. Managers use queue measures, scenario analysis and reserve capacity. Service quality and employee emotional demands are also important; a highly utilized service system may produce unacceptable waiting and burnout.

Projects

Project scheduling uses activities, precedence relationships, resources and milestones. The critical path is the longest-duration path through a precedence network; it determines the minimum project completion time under the stated durations and assumptions. Activities on the critical path have zero total float in the basic deterministic model.

For activity i, if ES_i and EF_i are earliest start and finish, and LS_i and LF_i are latest start and finish, total float is:

This formula assumes a consistent network and deterministic activity durations. In uncertain projects, probabilistic estimates, simulation and risk reserves are more appropriate. Compressing a project by adding resources is called crashing; it should be applied selectively because not every activity can be shortened and additional resources may create coordination losses.

9. Technology and Contemporary Developments

Enterprise resource planning, advanced planning and scheduling systems, manufacturing execution systems, workforce-management platforms and digital twins support integrated planning. Real-time data from sensors and operational systems can reveal downtime, queue growth and attendance changes earlier than periodic reports.

Artificial intelligence can assist demand forecasting, anomaly detection, schedule generation and what-if analysis. However, AI-generated schedules are only as reliable as the data and objective function. Risks include forecast bias, opaque recommendations, overfitting to historical patterns and prioritizing utilization over safety or service. Human review, audit trails, access controls, model monitoring and clear escalation rules are essential.

Resilient planning has also become more important. Organizations increasingly use scenario planning for supply disruption, geopolitical risk, climate events, cyber incidents and labour shortages. Rather than optimize one forecast, managers may evaluate robust plans that perform acceptably across several plausible futures.

10. Illustrative Case: Northstar Medical Devices

Background

Northstar Medical Devices produces two families of diagnostic cartridges. Demand is seasonal, and cartridges require a specialized testing station. The station is the system bottleneck. Northstar has 30 permanent operators, limited overtime and strict regulatory requirements for trained staff.

Problem

The aggregate plan promised a peak-season increase of 25%, but the MPS loaded the testing station beyond available hours. Managers considered adding temporary workers, although training and certification would take six weeks. Customer service also observed increasing lateness.

Analysis

The planning team aggregated demand by cartridge family, calculated testing hours, and compared them with available bottleneck hours. It discovered that the MPS was feasible at assembly but required 1,180 testing hours against 1,000 available hours. In addition, 8% of testing time was lost to changeovers and rework. The apparent labour shortage was therefore not the only problem; the bottleneck was being used inefficiently.

The team applied TOC. It first identified the testing station as the constraint. It then exploited the constraint by grouping compatible batches, scheduling preventive maintenance outside peak periods, improving first-pass quality and ensuring that trained operators were available at every shift. Upstream release was subordinated to testing capacity, reducing excess work-in-process. Finally, Northstar elevated capacity by authorizing a controlled overtime block and qualifying an external laboratory for a limited, audited volume.

Approach and outcome

The revised aggregate plan used a stable permanent workforce, targeted overtime, a small safety inventory of approved cartridges and limited subcontracting. The MPS was rebuilt with a bottleneck capacity check. Jobs were sequenced using due-date priority within compatible setup families. A Gantt chart was reviewed daily, while a frozen window protected near-term stability.

The plan reduced testing overload, improved delivery reliability and avoided unsafe last-minute hiring. It also exposed a longer-term need to invest in a second testing station. The outcome was not simply higher utilization; it was improved system throughput and more reliable customer service.

Learning points

1. Aggregate plans must be checked against the actual bottleneck.
2. Adding labour at a non-bottleneck does not necessarily increase throughput.
3. Schedule stability can be as valuable as theoretical flexibility.
4. Workforce capability and certification are capacity constraints.
5. Quality and safety improvements can create capacity without adding equipment.

Challenges and Limitations

Aggregate plans depend on forecasts, cost estimates and assumptions about capacity. Forecast error can make an apparently optimal plan inappropriate. Linear cost models may ignore nonlinear effects such as fatigue, learning, congestion and minimum staffing. Sequencing rules can optimize one measure while damaging another. Finite-capacity schedules can become unstable when disruptions occur. TOC may oversimplify systems with several near-simultaneous constraints. Finally, software can create false confidence when master data, routings, calendars or skill records are inaccurate.

Managers should therefore use rolling reviews, scenario analysis, exception management and disciplined data governance. Performance measures should include throughput, on-time delivery, average flow time, schedule adherence, inventory, overtime, quality, safety and employee outcomes.

Best-Practice Checklist

• Integrate sales, operations, finance, procurement and human resources.
• Aggregate only when products have sufficiently similar capacity requirements.
• Separate forecast assumptions from firm customer orders.
• Check capacity by resource, skill and time period.
• Use finite-capacity logic at bottlenecks and near-term work centres.
• Select sequencing rules according to the performance objective.
• Protect the short-term schedule with appropriate time fences.
• Maintain a stable core workforce while developing controlled flexibility.
• Treat safety, fairness and legal compliance as constraints, not optional costs.
• Use technology to support, not replace, accountable managerial judgement.

Chapter Summary

Aggregate planning balances medium-term demand and capacity through production, inventory, workforce, overtime, subcontracting and backlog decisions. Level, chase and mixed strategies offer different trade-offs between stability, cost, inventory and responsiveness. The inventory balance and capacity constraints provide a basic analytical foundation.

The master production schedule translates aggregate decisions into end-item quantities and dates. Detailed scheduling assigns jobs to resources and sequences them using rules such as FCFS, SPT, EDD and critical ratio. Gantt charts communicate timing, while finite-capacity scheduling prevents resource overloads. The Theory of Constraints focuses attention on the system bottleneck through identification, exploitation, subordination and elevation.

Workforce operations connect capacity planning to people. Cross-training, overtime, temporary labour, shift design and automation can provide flexibility, but each involves cost, capability, ethical and safety implications. Manufacturing, services and projects require different scheduling approaches, yet all depend on accurate data, explicit priorities and continual review. Modern planning technologies and AI can improve visibility and analysis, but governance and human accountability remain essential.

Key Terms

Aggregate plan: A medium-term plan for demand, production, inventory, workforce and capacity.

Backorder: Demand that is accepted but not fulfilled by the requested date.

Bottleneck: A resource or condition that limits system throughput.

Chase strategy: An aggregate strategy that adjusts production or workforce to follow demand.

Critical path: The longest precedence path determining minimum project duration.

Critical ratio: The time remaining until due date divided by remaining processing time.

Drum-buffer-rope: A TOC method linking constraint pace, protective buffer and controlled release.

Earliest due date: A sequencing rule that prioritizes the job with the nearest due date.

Finite-capacity scheduling: Scheduling that respects actual resource limits.

Gantt chart: A time-scaled chart showing activity timing and duration.

Level strategy: An aggregate strategy using relatively stable workforce and production.

Master production schedule: A time-phased plan for specific end items.

Projected available balance: Expected inventory after scheduled receipts and requirements.

Sequencing rule: A priority rule used to determine job order.

S&OP: A cross-functional process for integrating demand, supply and business plans.

Takt time: Available production time divided by customer demand.

Throughput: The rate at which a system generates output for sale.

Total float: The time an activity can be delayed without delaying project completion.

Workforce flexibility: The ability to adjust staffing, skills, hours or deployment as requirements change.

Review Questions

Short Answer Questions

1. What is the purpose of aggregate planning?
2. Distinguish a level strategy from a chase strategy.
3. What information does a master production schedule provide?
4. Define a bottleneck in the Theory of Constraints.
5. What is the difference between infinite-capacity and finite-capacity scheduling?

Long Answer Questions

1. Explain how aggregate planning, master scheduling and detailed scheduling are connected.
2. Compare FCFS, SPT, EDD and critical-ratio sequencing rules, including an advantage and limitation of each.
3. Explain the five focusing steps of the Theory of Constraints.
4. Discuss how workforce strategies can provide flexibility while protecting quality and employee well-being.
5. Compare scheduling requirements in manufacturing, services and projects.

Analytical / Application Questions

1. A company begins with 150 units of inventory and forecasts demand of 700, 900 and 800 units. If it produces 800 units each month, calculate ending inventory after each month and assess whether the plan avoids backorders.
2. Four jobs require 5, 2, 7 and 3 hours and have due dates of 9, 6, 15 and 8 hours respectively. Construct the SPT sequence and the EDD sequence, calculate completion times, and explain which performance objective each sequence supports.
3. A bottleneck work centre has 360 available hours but an MPS requires 410 hours. Recommend a response using the Theory of Constraints and discuss at least two workforce or scheduling risks of your recommendation.

Multiple Choice Questions with Answers

Question 1

Which planning level normally deals with product families and a medium-term horizon?

A. Aggregate planning

B. Dispatching

C. Preventive maintenance

D. Inspection scheduling

Correct Answer: A. Aggregate planning

Question 2

In the inventory balance, what generally increases ending inventory, all else equal?

A. Higher demand

B. Higher production

C. Higher backorders

D. Lower opening inventory

Correct Answer: B. Higher production

Question 3

Which strategy keeps production and workforce relatively stable while using inventory or backlog to absorb variation?

A. Chase strategy

B. Level strategy

C. Random strategy

D. Zero-capacity strategy

Correct Answer: B. Level strategy

Question 4

Which sequencing rule prioritizes the job with the shortest processing time?

A. EDD

B. FCFS

C. SPT

D. Critical ratio

Correct Answer: C. SPT

Question 5

A critical ratio below 1 generally indicates that a job is:

A. Ahead of the time required for its due date

B. Behind the time required for its due date

C. Already completed

D. Unrelated to its due date

Correct Answer: B. Behind the time required for its due date

Question 6

What is the main purpose of finite-capacity scheduling?

A. To ignore machine limitations

B. To prevent assignments exceeding available resource capacity

C. To maximize inventory regardless of demand

D. To eliminate all setup times

Correct Answer: B. To prevent assignments exceeding available resource capacity

Question 7

In the Theory of Constraints, what should be done after identifying a constraint?

A. Exploit it before considering elevation

B. Ignore it and improve non-bottlenecks

C. Increase every resource equally

D. Stop measuring throughput

Correct Answer: A. Exploit it before considering elevation

Question 8

If 2,000 available minutes are divided by demand of 500 units, what is takt time?

A. 2 minutes per unit

B. 4 minutes per unit

C. 250 minutes per unit

D. 2,500 minutes per unit

Correct Answer: B. 4 minutes per unit

Question 9

Which workforce practice most directly increases the ability to move employees among tasks?

A. Cross-training

B. Reducing all training

C. Eliminating skill records

D. Increasing setup time

Correct Answer: A. Cross-training

Question 10

In basic project scheduling, total float can be calculated as:

A. ES + EF

B. LS − ES

C. Demand ÷ capacity

D. Inventory − demand

Correct Answer: B. LS − ES