9. Detailed Scheduling, Sequencing and Line Balancing

Introduction and Context

Production planning determines what is to be produced, in what quantity, and within which broad time horizon. Detailed scheduling converts these decisions into an executable timetable. It specifies when each job should start and finish, which machine or work centre should process it, the order in which competing jobs should be undertaken, and how labour and equipment should be coordinated.

Scheduling is therefore the point at which plans encounter the physical realities of production: limited machine capacity, alternative routings, setup times, material availability, labour skills, maintenance interruptions, due dates, and uncertain processing times. A schedule that appears efficient on paper may be impossible to execute if it ignores these constraints. Production control subsequently compares actual progress with the schedule and initiates corrective action.

Sequencing is closely related but more specific. Scheduling determines the timing and allocation of work, whereas sequencing determines the order in which jobs are processed. Line balancing applies similar logic to repetitive and assembly operations, where tasks must be distributed among workstations so that production proceeds at a required rate with minimal idle time.

The appropriate method depends on the production environment. Job shops require flexible sequencing among diverse jobs; flow shops require coordination across a fixed process route; batch systems must consider campaign sizes and setup costs; and continuous systems depend heavily on stable rates, process control, and synchronization.

Learning Objectives

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

1. Distinguish scheduling, sequencing, dispatching, and line balancing.

2. Explain the characteristics of job-shop, flow-shop, batch, and continuous scheduling.

3. Select and evaluate priority rules using appropriate performance measures.

4. Construct and interpret Gantt charts.

5. Apply Johnson’s rule to a two-machine flow-shop problem.

6. Explain finite-capacity scheduling and its managerial implications.

7. Balance an assembly line using cycle time, precedence relationships, and workstation assignment.

8. Evaluate scheduling decisions in terms of delivery, utilization, work-in-process, flexibility, and operational risk.

9. Recognize the role of digital scheduling systems, analytics, and human judgment.

The Logic of Detailed Scheduling

A schedule normally contains five elements: jobs, resources, processing times, sequence, and time. A job may be a customer order, a batch, a maintenance activity, or an internal replenishment order. Resources include machines, labour groups, tools, inspection facilities, and material-handling equipment.

Several measures are used to judge schedule quality. If job j has completion time Cj, processing time pj, release time rj, and due date dj, then common measures include:

• Flow time: Cj − rj. This indicates the time spent in the system.
• Lateness: Lj = Cj − dj. A negative value means early completion and a positive value means late completion.
• Tardiness: Tj = max(0, Cj − dj). Unlike lateness, tardiness does not reward early completion with a negative value.
• Makespan: The completion time of the final job in the schedule.
• Utilization: The proportion of available resource time actually used for productive work.
• Work-in-process: Jobs released but not yet completed.

No single schedule optimizes all these measures. A sequence that minimizes average flow time may cause an important customer order to be late. A sequence that protects due-date performance may reduce machine utilization because of additional setups. Scheduling is consequently a managerial trade-off rather than a purely mathematical exercise.

Types of Production Environment

In a job shop, products follow different routings and may require different machines. Examples include tool rooms, repair workshops, and customized engineering production. The principal scheduling difficulty is the interaction of multiple routes and competing orders.

In a flow shop, all jobs generally pass through machines in the same order, although processing times may differ. Food processing, appliance manufacture, and many component operations have flow-shop characteristics. Sequencing is comparatively structured, and methods such as Johnson’s rule can be useful in special cases.

A batch system produces a quantity of identical or similar units before changing to another product. Scheduling must balance setup frequency, inventory, delivery requirements, and campaign efficiency. Long campaigns reduce setup losses but may delay other products.

A continuous system produces material through a highly integrated, usually uninterrupted process. Petroleum refining, cement, electricity generation, and chemical processing are examples. The central concerns are rate synchronization, capacity stability, maintenance windows, quality control, and safe operation. Conventional job-by-job sequencing has less relevance, although production-rate and changeover decisions remain important.

Scheduling Relationships and Constraints

A feasible schedule must satisfy several types of constraints. Precedence constraints require one activity to occur before another. For example, drilling must precede painting. Capacity constraints prevent a machine or employee from processing two jobs simultaneously. Material constraints prevent an operation from starting before required components are available. Labour constraints arise when only qualified personnel can perform particular tasks. Tooling constraints occur when fixtures, dies, or gauges are limited.

A further distinction is made between infinite-capacity and finite-capacity scheduling. Infinite-capacity scheduling assigns work according to required dates without initially checking whether the necessary resource capacity exists. It is useful for rough planning but often creates unrealistic overloads. Finite-capacity scheduling incorporates actual capacity and does not place work on a resource when that resource is unavailable. It may move work to a later time, use an alternative resource, split a lot, or alter the priority sequence.

Finite scheduling is usually more realistic, but it also requires better data. Processing times, calendars, setup rules, alternative routings, labour availability, and maintenance periods must be reasonably accurate. A sophisticated scheduling algorithm cannot compensate for unreliable master data.

Priority Rules in Practice

Priority rules provide a practical way to choose the next job when several jobs are waiting. Common rules include:

• First come, first served (FCFS): Jobs are processed in arrival order. It is simple and perceived as fair, but it may perform poorly when a long job blocks several short jobs.
• Shortest processing time (SPT): The job with the smallest processing time is processed first. SPT often reduces average flow time and work-in-process, but long jobs may be continually postponed.
• Earliest due date (EDD): The job with the nearest due date is selected first. EDD is useful when due-date performance is important.
• Critical ratio (CR): CR = (Due date − Current time) / Remaining processing time. A lower ratio indicates greater urgency. The ratio changes as time passes, making it useful in dynamic environments.
• Slack time: Slack = Due date − Current time − Remaining processing time. The smallest slack receives priority.
• Least slack per operation: Slack is divided by the number of remaining operations, giving attention to jobs with limited scheduling flexibility.
• Setup-oriented rules: Jobs may be grouped by product family, colour, material, tooling, or temperature to reduce changeovers.

Priority rules should not be treated as universal laws. A hospital equipment order, a safety-related repair, or a customer order with contractual penalties may properly override SPT or FCFS. Effective dispatching combines quantitative rules with organizational priorities and exception management.

Gantt Charts as Scheduling Instruments

A Gantt chart displays jobs or activities against a time scale. Each horizontal bar shows the planned duration of an operation; gaps indicate idle time, waiting, or unavailable capacity. Separate rows can represent machines, work centres, employees, or orders.

A basic Gantt chart supports four managerial activities. First, it communicates the intended sequence. Second, it reveals overloads and idle periods. Third, it helps supervisors compare planned and actual progress. Fourth, it makes the consequences of disruptions visible. If a machine fails, the affected bars can be shifted and downstream effects examined.

A Gantt chart is not necessarily an optimization method. It is a visualization and control tool. Its quality depends on the logic used to create it. A visually attractive chart may still contain impossible overlaps, missing setups, or unrecognized material constraints. Digital systems can update charts automatically, but supervisors must still validate assumptions and respond to real operating conditions.

Johnson’s Rule for Two-Machine Flow Shops

Johnson’s rule determines an optimal sequence for minimizing makespan when every job passes through two machines in the same order, processing times are known, machines process only one job at a time, and jobs are not interrupted. Let the machines be M1 and M2.

The procedure is:

1. Identify the smallest processing time among all unscheduled jobs on both machines.

2. If the smallest time belongs to M1, place that job in the earliest available position in the sequence.

3. If it belongs to M2, place that job in the latest available position.

4. Remove the job and repeat until all jobs are sequenced.

Illustrative Example

Suppose four jobs have the following processing times:

| Job | M1 | M2 |

|---|---:|---:|

| A | 6 | 3 |

| B | 2 | 8 |

| C | 5 | 4 |

| D | 7 | 1 |

The smallest time is 1 for D on M2, so D is placed last. The next smallest is 2 for B on M1, so B is placed first. The next smallest is 3 for A on M2, so A is placed in the latest unfilled position before D. C occupies the remaining position. The sequence is therefore B–C–A–D.

The completion calculation is:

| Job | M1 completion | M2 start | M2 completion |

|---|---:|---:|---:|

| B | 2 | 2 | 10 |

| C | 7 | 10 | 14 |

| A | 13 | 14 | 17 |

| D | 20 | 20 | 21 |

The makespan is 21 time units. Johnson’s rule is powerful within its assumptions, but it does not directly solve three-machine problems, unrelated routings, sequence-dependent setups, or environments with significant interruptions. Extensions and heuristic methods may be required in those situations.

Scheduling in Job Shops and Batch Systems

Job-shop scheduling is difficult because jobs visit different machines in different orders. A job may wait for a lathe, move to a milling machine, return for inspection, and then require assembly. The schedule must coordinate both machine priorities and technological precedence.

Two broad approaches are common. Forward scheduling begins with the earliest possible start and schedules operations toward completion. It supports high utilization and early completion but may create excessive finished-goods inventory. Backward scheduling begins with the due date and works backward, placing operations as late as possible. It reduces unnecessary early production but is vulnerable to disruption and may leave little protective capacity.

Batch scheduling adds the question of lot size. Small batches reduce waiting and improve responsiveness but increase setup frequency. Large batches reduce setup cost per unit but increase cycle time and inventory. A practical schedule often uses product families and sequence-dependent setup matrices. For example, a paint operation may sequence light colours before dark colours, or a food line may schedule allergen-free products before products containing allergens, reducing cleaning and contamination risk.

In job shops, useful performance controls include queue length, bottleneck utilization, overdue order count, average flow time, and schedule adherence. The bottleneck should receive particular attention because lost time at a constraint cannot normally be recovered elsewhere in the system without additional capacity or overtime.

Assembly-Line Balancing

Assembly-line balancing assigns tasks to workstations so that each station has an acceptable workload and all precedence relationships are respected. The fundamental quantity is cycle time, the maximum time allowed at each station to meet the required output rate:

Cycle time = Available production time per period / Required output per period.

If total task time is T and cycle time is C, the theoretical minimum number of workstations is:

Nmin = ceiling(T / C).

The actual number may be higher because tasks cannot always be divided freely and precedence restrictions may prevent perfect allocation. A common efficiency measure is:

Line efficiency = Total task time / (Number of stations × Cycle time) × 100.

Illustrative Balancing Example

Assume a line must produce one unit every 10 minutes. The tasks are: A = 4 minutes, B = 3, C = 5, D = 2, E = 4, and F = 2. The precedence relationships are A before B, A before C, B and C before D, D before E, and E before F. Total task time is 20 minutes, so the theoretical minimum is two stations.

A feasible assignment is:

• Station 1: A, B = 7 minutes
• Station 2: C, D, E = 11 minutes, which is infeasible because it exceeds the 10-minute cycle time.

The tasks must therefore be rearranged. One possible arrangement is:

• Station 1: A, C = 9 minutes
• Station 2: B, D, E = 9 minutes
• Station 3: F = 2 minutes

Three stations are required because precedence and task indivisibility prevent a two-station solution. Efficiency is 20/(3 × 10) × 100 = 66.7 percent. The unused time is balance delay. A different assignment may improve the distribution, but it cannot place a task before its predecessors or exceed the cycle time.

Balancing methods include ranked positional weight, largest candidate rule, and trial-and-error supported by precedence diagrams. In practice, ergonomic considerations, walking distance, skill requirements, quality checks, and worker rotation must also be included. A mathematically balanced line can remain poor if it causes fatigue or makes inspection ineffective.

A Detailed Illustrative Case: Scheduling at Orion Precision Components

Background

Orion Precision Components is an illustrative medium-sized manufacturer producing customized hydraulic components. Its system contains a CNC turning section, a milling section, heat treatment, inspection, and final assembly. Orders vary in routing and quantity, making the plant a hybrid job shop and batch operation.

Situation and Problem

On-time delivery had declined even though average machine utilization exceeded 85 percent. Supervisors frequently released urgent orders into already crowded work centres. Jobs were often started early, creating large queues before heat treatment and inspection. The production manager initially proposed increasing overtime across all departments.

Analysis

A review of four weeks of data showed that the inspection section was the principal constraint. The dispatching rule at machining was effectively FCFS, but several long jobs occupied machines while short, nearly complete jobs waited. Setup information was not included in the informal schedule. The plant also used infinite-capacity planning, so several operations were promised for the same inspection period.

The team mapped routings, measured actual setup and processing times, and introduced a finite-capacity schedule. At machining, SPT was used for ordinary orders, while EDD and customer-penalty flags took precedence for committed deliveries. Similar materials were grouped to reduce setups, but grouping was restricted where it threatened due dates. Inspection capacity was protected by releasing work according to its available slots rather than merely according to machining completion.

A rolling Gantt chart was displayed for the next ten working days. Each morning, supervisors reviewed exceptions: overdue jobs, material shortages, broken tools, and capacity changes. The schedule was frozen for the immediate shift except in cases involving safety, quality, or contractual urgency.

Outcome

The revised system reduced unnecessary early release, exposed the inspection bottleneck, and improved coordination between machining and inspection. The central lesson was not that one priority rule solved the problem. Improvement came from combining reliable data, finite capacity, bottleneck protection, selective sequencing rules, and daily rescheduling discipline.

Learning Points

1. High utilization does not guarantee good delivery performance.

2. A bottleneck determines the practical pace of the system.

3. Releasing more work can increase congestion rather than output.

4. Priority rules should reflect order characteristics and business consequences.

5. Human review remains essential when data or conditions change.

Technology, Analytics, and Contemporary Practice

Modern manufacturing execution systems and advanced planning and scheduling systems integrate orders, routings, calendars, inventory, machine status, and labour information. Finite-capacity algorithms can evaluate alternative sequences and identify bottlenecks. Real-time data from machines can update actual completion times and trigger schedule revisions.

Optimization models may use mixed-integer programming, constraint programming, simulation, or metaheuristics such as genetic algorithms and tabu search. These tools are especially useful for complex job shops, sequence-dependent setups, and multi-objective scheduling. However, algorithmic solutions must be explainable to production personnel. A schedule that cannot be understood or followed may be rejected on the shop floor.

Artificial intelligence can support estimated processing times, disruption prediction, anomaly detection, and dynamic rescheduling. Its value depends on data quality and governance. Historical data may reflect past biases, such as systematically prioritizing certain customers or underestimating setup time. Human approval, audit trails, cybersecurity, and protection of operational data are therefore important.

Emerging practice emphasizes dynamic scheduling, in which schedules are updated as conditions change, while preserving a short frozen horizon to maintain stability. Digital twins and simulation allow managers to test alternative rules before implementation. Increasingly, scheduling is also linked with energy management, carbon reporting, maintenance planning, and workforce capability. The objective is shifting from isolated machine utilization toward resilient and sustainable flow.

Challenges, Risks, and Common Mistakes

Scheduling systems commonly fail for reasons that are managerial rather than mathematical. Typical problems include:

• using inaccurate processing times;
• ignoring setup, cleaning, inspection, or material movement;
• releasing work without checking downstream capacity;
• changing priorities too frequently;
• measuring utilization without considering delivery and inventory;
• overloading bottleneck resources;
• treating due dates as fixed without negotiating realistic commitments;
• ignoring worker fatigue, safety, or skill limitations;
• creating schedules so detailed that normal variation makes them obsolete.

Excessive rescheduling can create nervousness throughout the plant. Employees stop trusting the schedule, materials are moved repeatedly, and jobs are started merely to satisfy changing priorities. A stable frozen period, clear escalation rules, and disciplined exception management reduce this risk.

Ethically and professionally, planners should not conceal capacity shortages by assigning impossible dates or manipulating completion records. Safety-critical, medical, and regulated products require traceability and authorized release procedures. Schedule pressure must never justify bypassing quality inspections, maintenance requirements, labour protections, or environmental controls.

Best-Practice Guidelines

Effective scheduling generally follows these principles:

1. Maintain accurate routings, standard times, calendars, and setup data.

2. Identify and protect bottleneck resources.

3. Use finite capacity for executable schedules.

4. Match the priority rule to the performance objective.

5. Distinguish normal priorities from genuine emergencies.

6. Use Gantt charts to communicate and control, not merely to decorate reports.

7. Include setup, material, labour, maintenance, and quality constraints.

8. Freeze the near-term schedule while allowing controlled revisions beyond it.

9. Measure delivery, flow time, work-in-process, utilization, and schedule adherence together.

10. Review actual performance and revise standards through a structured learning process.

Chapter Summary

Detailed scheduling transforms aggregate production plans into time-phased assignments of jobs and resources. Sequencing determines the order of competing jobs, while line balancing distributes tasks among workstations in repetitive production. Job shops require attention to alternative routings and bottlenecks; flow shops benefit from structured sequencing such as Johnson’s rule; batch systems must balance setup economies with responsiveness; and continuous systems depend on rate synchronization and operational stability.

Priority rules such as FCFS, SPT, EDD, critical ratio, and slack provide practical dispatching methods, but each produces different performance consequences. Gantt charts make schedules visible and support shop-floor control. Finite-capacity scheduling improves realism by recognizing actual resource limits. Assembly-line balancing uses cycle time, precedence relationships, workstation assignments, and efficiency measures to achieve required output.

Contemporary scheduling combines optimization, simulation, real-time data, and human judgment. The most effective systems do not pursue utilization in isolation. They coordinate delivery reliability, flow, capacity, quality, safety, workforce capability, and resilience.

Key Terms

Detailed scheduling: Assigning jobs, resources, sequences, and times for execution.

Sequencing: Determining the order in which jobs are processed.

Dispatching: Releasing or selecting the next job for processing.

Makespan: Completion time of the final job in a schedule.

Flow time: Time a job spends in the production system.

Lateness: Difference between completion time and due date.

Tardiness: Positive lateness when a job finishes after its due date.

Utilization: Proportion of available resource time used productively.

Job shop: Production environment with varied products and alternative routings.

Flow shop: Production environment in which jobs generally follow the same machine sequence.

Batch production: Production of a quantity of similar units before a changeover.

Continuous production: Integrated production operating at a sustained process rate.

Priority rule: A rule used to select the next job for processing.

Gantt chart: Time-scaled visual representation of scheduled activities.

Finite-capacity scheduling: Scheduling that respects actual resource capacity and availability.

Cycle time: Maximum time allowed at a workstation to meet required output.

Precedence relationship: Requirement that one task be completed before another begins.

Line efficiency: Ratio of total task time to available station time.

Balance delay: Unused workstation time caused by imperfect task allocation.

Bottleneck: Resource whose limited capacity restricts overall system output.

Review Questions

Short Answer Questions

1. Distinguish scheduling from sequencing.

2. What is the difference between lateness and tardiness?

3. State the formula for assembly-line cycle time.

4. What is the purpose of a Gantt chart?

5. Define finite-capacity scheduling.

Long Answer Questions

1. Explain the characteristics and scheduling problems of job shops, flow shops, batch systems, and continuous systems.

2. Compare FCFS, SPT, EDD, critical ratio, and slack-based priority rules.

3. Describe Johnson’s rule and explain the assumptions under which it is applicable.

4. Explain assembly-line balancing using cycle time, precedence relationships, theoretical minimum stations, and line efficiency.

5. Discuss how digital scheduling systems and real-time analytics are changing production control.

Analytical / Application Questions

1. Four jobs require processing on two machines. Explain how you would use Johnson’s rule and identify the information needed before applying it.

2. A factory reports 95 percent machine utilization but worsening delivery performance. Diagnose at least four possible scheduling causes and recommend corrective actions.

3. An assembly line has a cycle time of 12 minutes and total task time of 43 minutes. Calculate the theoretical minimum number of stations and explain why the actual number may be higher.

22. Multiple Choice Questions with Answers

Question 1

Which measure ignores early completion by assigning zero value to jobs completed before their due dates?

A. Flow time

B. Tardiness

C. Lateness

D. Makespan

Correct Answer: B. Tardiness

Question 2

Under Johnson’s rule, a job with the smallest processing time on Machine 2 is placed:

A. In the earliest available position

B. In the latest available position

C. Immediately before the longest job

D. At the position of its due date

Correct Answer: B. In the latest available position

Question 3

Which priority rule generally tends to reduce average flow time when processing times are known?

A. Shortest processing time

B. First come, first served

C. Random selection

D. Longest processing time

Correct Answer: A. Shortest processing time

Question 4

A schedule that assigns work without checking whether the required resource capacity exists is an example of:

A. Finite-capacity scheduling

B. Backward scheduling

C. Infinite-capacity scheduling

D. Bottleneck scheduling

Correct Answer: C. Infinite-capacity scheduling

Question 5

If available production time is 480 minutes and required output is 60 units, the assembly-line cycle time is:

A. 6 minutes

B. 8 minutes

C. 10 minutes

D. 12 minutes

Correct Answer: B. 8 minutes

Question 6

Which situation most clearly represents a precedence constraint?

A. Two jobs require the same machine

B. A machine is unavailable during maintenance

C. Painting must occur after surface preparation

D. A worker is qualified for only one task

Correct Answer: C. Painting must occur after surface preparation

Question 7

The main purpose of grouping similar products in batch scheduling is usually to:

A. Increase the number of changeovers

B. Reduce setup losses

C. Eliminate all work-in-process

D. Guarantee shortest processing time

Correct Answer: B. Reduce setup losses

Question 8

A line has total task time of 36 minutes, four stations, and a cycle time of 10 minutes. Its line efficiency is:

A. 72%

B. 80%

C. 90%

D. 144%

Correct Answer: C. 90%

Question 9

Why can very high machine utilization coexist with poor delivery performance?

A. Utilization always measures customer satisfaction

B. High utilization may create queues and overload bottlenecks

C. High utilization eliminates work-in-process

D. Utilization is calculated only from due dates

Correct Answer: B. High utilization may create queues and overload bottlenecks

Question 10

Which practice best supports stable dynamic scheduling?

A. Changing every priority continuously

B. Ignoring setup and maintenance time

C. Using a short frozen horizon with controlled revisions

D. Releasing all available orders immediately

Correct Answer: C. Using a short frozen horizon with controlled revisions