Lean Operations and Just-in-Time Systems

Introduction and context

Lean operations is a management system for delivering customer value with the least necessary resources while controlling variation, defects, delay and risk. Just-in-Time (JIT) is a central lean principle: materials, information and work should be available when needed, in the quantity needed and at the required quality. Lean is not simply a cost-cutting programme or a collection of tools. It is a socio-technical system involving purpose, process discipline, employee participation, supplier relationships and continual learning.

The most influential model is the Toyota Production System (TPS), developed through the work of Toyota leaders including Taiichi Ohno and Shigeo Shingo. TPS is commonly represented by a house. Its foundation is stability and standardisation; its pillars are Just-in-Time and jidoka, or built-in quality; and its roof represents customer value, quality, delivery, safety, morale and cost. Continuous improvement and respect for people support the whole system.

Lean is especially important in Total Quality Management because it links quality to process design. Quality is not inspected into a product after production. It is created by capable processes, clear standards, immediate response to abnormalities and learning from root causes.

Learning objectives

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

1. Explain the principles and structure of TPS, lean operations and JIT.
2. Identify customer value and classify operational activities as value-adding or waste.
3. Calculate takt time, cycle time, throughput and basic kanban requirements.
4. Design and evaluate pull systems, flow, cellular layouts and standard work.
5. Apply 5S, visual management, mistake-proofing, setup reduction and kaizen.
6. Analyse the benefits, risks and implementation conditions of lean systems.
7. Evaluate lean performance using operational, quality, people and customer measures.

1. Customer value and the lean transformation

Customer value

Value is the capability or feature for which the customer is willing to pay, delivered at the right quality, time and price. Value must be defined from the customer’s perspective rather than from the convenience of the producer. For example, machining a component to the required specification is value-adding; moving it repeatedly between buildings is not.

An activity is usually classified as value-adding when it changes the product or service in a way the customer requires, is done correctly the first time and the customer would recognise as part of the paid offering. Necessary but non-value-adding activities, such as regulatory inspection or certain transport activities, may be unavoidable, but they should be reduced. Pure waste consumes resources without creating customer value.

The eight wastes

The traditional TPS categories are often remembered by the acronym TIMWOODS:

Waste categories overlap. Excess inventory may conceal defects, waiting and scheduling problems. Therefore, reducing one visible waste without addressing its cause may merely move the problem elsewhere.

Lean principles

A widely used five-principle model is:

1. Specify value.
2. Identify the value stream.
3. Make value flow.
4. Establish pull.
5. Pursue perfection through continual improvement.

Value-stream thinking examines the complete sequence from demand to delivery, including information flows, suppliers, production, quality decisions and customer feedback. It discourages local optimisation, such as maximising the output of one machine while increasing total inventory and lead time.

2. The Toyota Production System

TPS combines two complementary ideas. Just-in-Time coordinates the timing and quantity of work. Jidoka stops or contains a process when an abnormality occurs, preventing defects from moving downstream. JIT without quality creates rapid production of defective items; quality without flow may create excessive inspection and inventory.

Stability as the foundation

Flow and pull systems require relatively stable equipment, materials, methods, demand information and workforce capability. Stability does not mean that demand is perfectly predictable. It means that known sources of variation are controlled and abnormalities are made visible. Preventive maintenance, reliable suppliers, standard work, training and controlled schedules support stability.

The principle of respect for people

Lean requires employees to identify problems and improve work. Respect does not mean lowering standards; it means providing safe work, fair treatment, development, authority to raise concerns and involvement in problem-solving. A system that uses lean tools to intensify work, hide defects or punish problem reporting is not a mature lean system.

The improvement cycle

Kaizen follows a disciplined cycle: define the problem, observe the current condition, identify causes, test countermeasures, check results, standardise successful changes and continue learning. The PDCA cycle—Plan, Do, Check, Act—is commonly used. A3 reports provide a concise structure for problem definition, analysis, countermeasures and follow-up.

3. Flow, takt and capacity

Takt time

Takt time expresses the rate at which a customer requires output. It is not the same as the process cycle time.

Available production time must exclude planned breaks, meetings and other unavailable periods. The units of both terms must be identical. If a plant has 420 available minutes per shift and demand is 140 units per shift:

This means one completed unit is required every three minutes on average. Takt is a demand signal, not a guarantee that every unit takes exactly three minutes.

Cycle time and lead time

Cycle time is the elapsed time between completion of successive units at a process. Lead time is the total time from customer order to delivery, including waiting, transport, processing and decision time. A process can have a cycle time below takt but still have long lead time because of queues and batch sizes.

For a stable process, the approximate theoretical capacity is:

The result is valid only when the process is available for the stated time and cycle time is representative. Actual output may be lower because of downtime, changeovers, defects, staffing and material shortages.

Line balancing

The minimum theoretical number of workstations is:

where Σt_i is total work content, t_i is the time for task i, and C is the required cycle time, commonly set no greater than takt time. Since a fraction of a workstation is impossible, the practical number must be rounded up and must also respect task precedence and ergonomic constraints.

If total work content is 10.5 minutes and required cycle time is 3 minutes, N_min = 3.5, so at least four stations are theoretically required. Actual balancing may require more than four.

Continuous flow and batch reduction

One-piece or small-lot flow reduces waiting, inventory and feedback time. It is not always appropriate: some processes have technical batch constraints, long curing times, economical minimum runs or safety requirements. In such cases, the objective is to reduce batch size and decouple only where justified, not to pursue one-piece flow as an absolute rule.

Figure: Customer demand pulls replenishment through connected processes; physical work moves downstream while the signal returns upstream.

4. Pull systems and kanban

A push system releases work according to a forecast or central schedule, often creating queues when downstream capacity is unavailable. A pull system authorises upstream replenishment in response to actual downstream consumption. Pull does not eliminate planning; forecasts remain useful for capacity, purchasing and long-term decisions. Pull controls short-term release and replenishment.

Kanban is a visible or electronic signal that authorises production, movement or replenishment. A kanban may be a card, container, barcode scan, electronic message or location signal. Effective kanban rules specify the item, quantity, source, destination, trigger and response to abnormalities.

A basic kanban quantity relationship is:

where n is the number of kanban containers or cards, D is demand per time period, L is replenishment lead time in the same time unit, α is the safety factor expressed as a decimal, and C is the standard quantity per container. The result is rounded up to a whole number. The formula assumes reasonably stable demand, a defined replenishment lead time and containers of consistent size.

For demand of 480 units per day, replenishment lead time of 0.25 day, a safety factor of 10% and 30 units per container:

Therefore, five containers are required. The safety factor should not permanently conceal unreliable processes; it should be reviewed as lead time and variation improve.

Kanban is unsuitable or must be adapted where demand is extremely intermittent, product variety is very high, replenishment lead time is highly unstable, items are unique projects, or safety and regulatory controls require a different release mechanism.

5. Core lean practices

5S

5S creates an orderly, visual and stable workplace:

• Sort: remove unnecessary items.
• Set in order: locate necessary items for efficient retrieval.
• Shine: clean while inspecting for abnormalities.
• Standardise: define and maintain the desired condition.
• Sustain: develop habits, audits and leadership routines.

5S is not cosmetic housekeeping. It exposes leaks, wear, missing tools, unsafe conditions and abnormal inventory. Audits should assess meaningful conditions rather than reward superficial tidiness.

Visual management

Visual management allows a person unfamiliar with the area to understand normal and abnormal conditions quickly. Examples include floor markings, material limits, status boards, colour coding, andon signals, process measures and escalation rules. Visual displays should lead to action; a dashboard that no one uses is decoration rather than management.

Standard work

Standard work defines the current best-known method for safely producing a quality result at the required rate. It commonly includes:

1. Takt or required cycle time.
2. Work sequence.
3. Standard work-in-process needed to keep the sequence operating.

Standard work is not a restriction on improvement. It is the baseline from which improvement can be measured. If every operator uses a different method, performance differences cannot be analysed reliably.

Mistake-proofing

Poka-yoke prevents an error or makes it immediately detectable. Examples include keyed connectors, presence sensors, interlocks, templates, software validation and colour-coded components. Prevention is stronger than detection. However, sensors and alarms can introduce new failure modes, so they require verification, maintenance and clear responses.

Setup reduction and SMED

Single-Minute Exchange of Die (SMED) is a structured approach to reducing changeover time. Its logic is to separate internal activities, which require the machine to be stopped, from external activities, which can occur while it runs; convert internal activities to external where possible; simplify and standardise both types; and eliminate adjustment through guides, settings and repeatable fixtures. Benefits include smaller economic batches, greater flexibility and reduced inventory.

Cellular operations

A manufacturing cell groups equipment and people according to product family or process sequence rather than functional departments alone. Cells can shorten transport, reveal quality problems and support flow. Their design must consider demand, product families, equipment capability, ergonomics, maintenance and staffing flexibility. Cellular layouts are not automatically lean; poor family grouping may create bottlenecks or duplicate expensive equipment.

6. Problem-solving and kaizen

A lean improvement begins with a clearly observable gap between current and required performance. Teams should go to the gemba—the actual place where work occurs—and observe rather than rely only on reports. Useful tools include process mapping, value-stream mapping, Pareto analysis, cause-and-effect diagrams, the five whys, check sheets and control charts.

Root-cause analysis must distinguish symptoms from causes. For example, a late order is a symptom. The cause may be unreliable equipment, an inaccurate routing, an approval queue or poor demand information. The five whys is a questioning discipline, not a requirement to produce exactly five answers. Causes should be verified with evidence.

A good countermeasure is specific, testable and linked to the cause. After implementation, teams should compare performance with the baseline and check for unintended effects, such as improved speed accompanied by more defects or ergonomic strain. Successful changes are incorporated into standard work and training.

7. Lean performance measurement

No single measure represents lean performance. A balanced set may include:

• lead time and on-time delivery;
• first-pass yield, defects per unit and customer complaints;
• work-in-process and finished-goods inventory;
• changeover time, downtime and overall equipment effectiveness;
• productivity and capacity utilisation;
• safety incidents, ergonomic indicators and absenteeism;
• employee suggestions, participation and implemented improvements.

Overall Equipment Effectiveness (OEE) is often expressed as:

Availability is the proportion of planned production time during which equipment is available; performance compares actual speed with the ideal speed; quality is good output divided by total output. The factors must be expressed as decimals or percentages consistently. OEE is useful for diagnosing losses, but it should not be used to encourage overproduction or maximise isolated machine utilisation. A lower machine utilisation may be appropriate when demand is lower or when capacity is deliberately reserved for flexibility.

8. Implementation approach

A practical implementation sequence is:

1. Clarify customer requirements and business purpose.
2. Select a value stream and establish baseline measures.
3. Stabilise safety, quality, equipment, materials and standard work.
4. Observe the gemba and map current information and material flows.
5. Define takt, identify bottlenecks and design future-state flow.
6. Introduce workplace organisation, visual controls and mistake-proofing.
7. Reduce setup time and create smaller, controlled flows.
8. Pilot pull and kanban at a manageable boundary.
9. Establish daily management, escalation and problem-solving routines.
10. Audit, learn, standardise and extend the system.

Leadership should create conditions for improvement rather than merely announce targets. Supervisors need time and capability to coach problem-solving. Finance should recognise inventory reduction, quality improvement and lead-time gains without rewarding local overproduction. Human-resources policies should support training, multi-skilling and psychological safety.

Illustrative case: Northstar Medical Components

Background

Northstar Medical Components, a fictional manufacturer, produces sterile plastic housings for diagnostic equipment. Customer demand is 240 housings per day over two shifts. Available production time is 720 minutes per day after planned breaks. The value stream contains moulding, trimming, inspection and packing.

Problem

Customers report late deliveries despite high machine utilisation. Average lead time is 6.5 days, while actual processing time is only 18 minutes per unit. Work-in-process is concentrated between moulding and inspection. Defects are usually found at final inspection, and changeovers cause large batches.

Analysis

The team calculates takt time:

Observation shows that moulding produces in batches of 600 units, while downstream demand averages 240 units daily. The team identifies waiting, overproduction, transport, late defect detection and long setup as the major wastes. A Pareto review shows that two mould-change activities account for most changeover time.

Approach and solution

The team first establishes standard work for trimming and packing, introduces first-piece approval at changeover, and installs a visual abnormality signal. A supermarket holding a controlled quantity of accepted moulded parts is placed between moulding and downstream operations. Two-bin kanban replenishment is used, with quantities reviewed weekly. The team applies SMED: tools are prepared externally, clamps are replaced with quick-release devices, and settings are recorded on a standard sheet.

Poka-yoke fixtures prevent incorrect orientation during packing. Daily meetings review safety, quality, delivery, cost and people measures. Operators receive training in problem-solving and are authorised to stop the process when a defect is detected.

Outcome

After three months, illustrative results are: lead time reduced from 6.5 to 2.1 days; first-pass yield increased from 94% to 98.5%; average changeover time reduced by 55%; and work-in-process reduced by 43%. The results are not attributed to kanban alone. They depend on standard work, quality at source, setup reduction, reliable information and employee involvement.

Learning points

The case demonstrates that high utilisation can coexist with poor delivery, that inventory can hide process problems, and that pull works only when processes are stable enough to replenish reliably. It also shows why lean should be evaluated using a balanced set of customer, quality, flow and people outcomes.

9. Limitations, risks and governance

Lean systems can fail when organisations copy tools without understanding the system. Excessive inventory reduction may increase vulnerability to supplier disruption, transport failure or sudden demand changes. JIT therefore requires risk-based buffers, qualified suppliers, contingency plans and visibility beyond the immediate facility.

Demand smoothing, or heijunka, can reduce instability by sequencing products and quantities more evenly. It cannot eliminate genuine market volatility. In critical healthcare, food, aerospace and public-service operations, safety stock and traceability may be essential. The correct question is not whether inventory is good or bad, but what level protects service at an acceptable total cost and risk.

Ethical risks include using lean as a pretext for unsafe work intensification, removing jobs without consultation, manipulating performance data or suppressing stop-the-line authority. Governance should include safety reviews, worker participation, privacy controls for digital monitoring, transparent measures and escalation of quality concerns. Regulatory requirements for traceability, validation, documentation and product safety must never be bypassed in the name of speed.

Digital lean systems increasingly use sensors, real-time location data, electronic kanban, process mining and artificial intelligence for demand forecasting or predictive maintenance. AI can identify patterns and prioritise inspection, but its recommendations require data-quality controls, human accountability, cybersecurity and validation. Automation should remove waste and improve capability, not merely conceal poor process design.

Best-practice principles

• Begin with customer value and a measurable problem.
• Stabilise before optimising.
• Design flow around demand and quality requirements.
• Use pull to control release, not to compensate for unreliable processes.
• Make abnormal conditions visible and respond quickly.
• Build quality into the process and stop defects at source.
• Treat standard work as the current baseline for improvement.
• Involve the people who perform the work.
• Reduce setup and batch size before reducing protective inventory.
• Use balanced measures and review unintended consequences.
• Protect safety, ethics, resilience and regulatory compliance.
• Sustain gains through leadership routines, auditing and learning.

Chapter Summary

Lean operations is a customer-centred system for creating value while reducing waste, variation and unnecessary resource use. TPS integrates Just-in-Time flow and pull with jidoka, standardisation, respect for people and continuous improvement. Takt time connects demand to operating rhythm; cycle time, capacity and line balancing help determine whether processes can meet that rhythm. Kanban authorises replenishment based on consumption, while flow, cellular layouts and setup reduction reduce waiting and batch-related waste.

5S, visual management, standard work, poka-yoke and kaizen make abnormalities visible and support reliable improvement. Lean performance must be evaluated through a balanced set of measures covering delivery, quality, inventory, productivity, safety and employee involvement. Successful implementation requires process stability, capable leadership, worker participation, risk management and ethical governance. Lean is not the indiscriminate removal of resources; it is the disciplined design and improvement of a system that reliably delivers customer value.

Key Terms

Lean operations: A system for creating customer value with minimal waste and controlled variation.

Just-in-Time: Providing the right item, quantity and timing when required.

Toyota Production System: An integrated system based on flow, pull, built-in quality and continual improvement.

Jidoka: Building quality into the process and stopping when an abnormality occurs.

Customer value: Product or service capability that fulfils a customer requirement.

Waste: Resource consumption that does not create required customer value.

Value stream: The complete sequence of activities and information needed to deliver value.

Takt time: Available production time divided by customer demand.

Cycle time: Time between successive completed units at a process.

Lead time: Total elapsed time from order or request to delivery.

Pull system: A system in which downstream consumption authorises upstream replenishment.

Kanban: A signal authorising production, movement or replenishment.

5S: Sort, Set in order, Shine, Standardise and Sustain.

Standard work: The documented current best-known method for performing a task.

Poka-yoke: Mistake-proofing that prevents or detects errors.

SMED: A method for reducing setup and changeover time.

Kaizen: Structured, continuous, incremental improvement.

Gemba: The actual place where work and problems occur.

Heijunka: Smoothing production volume and product mix.

Overall Equipment Effectiveness: A combined measure of availability, performance and quality.

Review Questions

Short Answer Questions

1. Define customer value and distinguish value-adding from non-value-adding activity.
2. Explain the difference between a push system and a pull system.
3. What are the five elements of 5S?
4. Why is standard work important in kaizen?
5. State the relationship between takt time and customer demand.

Long Answer Questions

1. Explain the Toyota Production System house and the relationship between JIT and jidoka.
2. Discuss the eight wastes and show how one waste can conceal another.
3. Describe how SMED, cellular operations and standard work can improve flow.
4. Explain the design and governance requirements of a kanban system.
5. Evaluate the benefits and limitations of lean implementation in a volatile supply chain.

Analytical / Application Questions

1. A process has 600 available minutes per day and customer demand of 180 units. Calculate takt time and explain how it should influence line design.
2. A value stream has high machine utilisation but late deliveries and rising inventory. Diagnose likely causes and propose a staged lean intervention.
3. Design a balanced performance-measurement system for a healthcare or manufacturing process. Explain how it would prevent lean improvements from damaging safety or quality.

Multiple Choice Questions with Answers

Question 1

Which principle most directly means that upstream production is authorised by downstream consumption?

A. Push scheduling

B. Pull production

C. Functional layout

D. Final inspection

Correct Answer: B. Pull production

Question 2

A facility has 480 available minutes per day and demand for 160 units. What is the takt time?

A. 2 minutes per unit

B. 3 minutes per unit

C. 160 minutes per unit

D. 640 minutes per unit

Correct Answer: B. 3 minutes per unit

Question 3

Which TPS concept requires a process to stop or signal when an abnormality occurs?

A. Jidoka

B. Heijunka

C. Mura

D. Kanban

Correct Answer: A. Jidoka

Question 4

Which activity is an example of overproduction?

A. Producing to a confirmed customer requirement

B. Repairing a defective component

C. Making products earlier than they are required

D. Reducing unnecessary walking

Correct Answer: C. Making products earlier than they are required

Question 5

What is the primary purpose of standard work?

A. To prevent all future changes

B. To establish a reliable baseline for safe, repeatable performance and improvement

C. To maximise batch size

D. To replace employee judgement in every situation

Correct Answer: B. To establish a reliable baseline for safe, repeatable performance and improvement

Question 6

In the basic kanban formula, why is the calculated number of containers rounded upward?

A. Containers cannot normally be fractional

B. Demand is always increasing

C. Cycle time is ignored

D. Quality is guaranteed by extra containers

Correct Answer: A. Containers cannot normally be fractional

Question 7

Which SMED action is most appropriate?

A. Perform all preparation after the machine stops

B. Increase batch size to avoid changeovers

C. Move preparation activities outside the machine stoppage period

D. Eliminate setup documentation

Correct Answer: C. Move preparation activities outside the machine stoppage period

Question 8

Which statement about 5S is correct?

A. It is only a cleaning campaign

B. It removes the need for standard work

C. It makes normal and abnormal workplace conditions more visible

D. It applies only to factories

Correct Answer: C. It makes normal and abnormal workplace conditions more visible

Question 9

If total work content is 10 minutes and the required cycle time is 3 minutes, the theoretical minimum number of stations is:

A. 2

B. 3

C. 4

D. 10

Correct Answer: C. 4

Question 10

Why should OEE not be used as the sole lean performance measure?

A. It cannot include quality

B. It may encourage local machine utilisation or overproduction at the expense of customer flow

C. It has no relationship to equipment

D. It is only applicable to office work

Correct Answer: B. It may encourage local machine utilisation or overproduction at the expense of customer flow