Lean, Agile and Resilient Production Systems

Introduction and Context

Production systems operate under simultaneous pressure to reduce cost, improve quality, respond quickly to customers and continue operating during disruption. Lean production focuses on creating customer value with fewer resources by eliminating waste and stabilising processes. Agile production emphasises speed, flexibility and responsiveness when demand, product mix or technology changes rapidly. Resilient production extends these ideas by developing the capability to absorb shocks, recover quickly and adapt to new conditions.

These approaches are complementary rather than mutually exclusive. A lean system without adequate flexibility may become vulnerable to supply interruptions or demand volatility. An agile system without process discipline may respond quickly but at excessive cost. A resilient system combines operational efficiency, visibility, flexibility, risk management and organisational learning.

For production managers, the central challenge is not simply to adopt tools such as Kanban or 5S. It is to design a coherent operating system in which material, information, people, equipment and decisions flow reliably from suppliers to customers.

Learning Objectives

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

1. Explain the principles and strategic purpose of lean, agile and resilient production.
2. Identify the eight common forms of waste and distinguish value-adding from non-value-adding work.
3. Construct and interpret a value-stream map.
4. Apply Just-in-Time, Kanban, 5S, SMED, poka-yoke, kaizen and standardized work.
5. Compare lean and agile production systems and identify when each is appropriate.
6. Calculate basic takt time, cycle-time capacity and Kanban quantities.
7. Design responses to demand variability, supplier failure, equipment breakdown and other disruptions.
8. Evaluate implementation risks, performance measures and managerial responsibilities.

1. Foundations of Lean Production

1.1 Value and waste

Lean begins with value as defined by the customer. An activity is value-adding when it transforms the product or service in a way the customer requires, is performed correctly the first time and is done at the right time. Inspection, movement, waiting and correction may be necessary under current conditions, but they do not normally add customer value.

The traditional seven wastes are:

A frequently added eighth waste is underused human capability: failing to use employees’ knowledge, creativity and problem-solving ability.

Waste is often a symptom of instability. Large inventories may conceal unreliable suppliers; rework may conceal poor process capability; and long queues may conceal unbalanced capacity. Lean improvement therefore attacks causes rather than merely reducing visible quantities.

1.2 The five lean principles

The five widely used principles are:

1. Specify value: determine what the customer needs and is willing to recognise as valuable.
2. Map the value stream: examine all activities and information flows required to deliver the product.
3. Create flow: remove interruptions, queues, backtracking and batch barriers.
4. Establish pull: produce in response to actual downstream consumption rather than speculative forecasts.
5. Pursue perfection: use continuous improvement to reduce waste and variation indefinitely.

Lean is therefore a management system, not a collection of isolated techniques. Its technical practices require leadership behaviour, employee involvement, problem-solving discipline and respect for people.

2. Value-Stream Mapping

Value-stream mapping (VSM) is a visual method for representing material and information flows from supplier to customer. It distinguishes processing time from waiting time and helps identify improvement priorities.

A current-state map normally records customer demand, production-control signals, suppliers, processes, inventories, transportation, operating time and information flows. A future-state map redesigns the system around flow, pull and a defined production rhythm.

Important measures include:

Takt time is the available production time divided by customer demand:

Available production time and demand must refer to the same time period. Breaks, planned maintenance and meetings should be excluded if they genuinely make production unavailable. Takt time is a demand rhythm, not necessarily the actual cycle time of every operation.

Lead time is the elapsed time from order initiation to delivery. Processing time is the time during which work is actively transformed. The ratio of processing time to lead time is sometimes called process-cycle efficiency:

This ratio is meaningful only when the numerator and denominator use the same units and the definition of value-adding time is consistent.

Worked example: interpreting a value stream

A factory has 420 available minutes per shift and customer demand of 140 units per shift. Its takt time is:

A product spends 3 minutes in actual value-adding processing but remains in the system for 180 minutes. Therefore:

The result does not mean that 98.33% of time is automatically removable. Some waiting may be required for safety, curing, transport or legal control. It does, however, indicate a strong opportunity to investigate queues, batch sizes, approvals and material movement.

3. Just-in-Time and Pull Control

Just-in-Time (JIT) aims to provide the right item, in the right quantity, at the right time and with the required quality. It reduces inventory and exposes problems, but it does not mean eliminating all buffers. A practical JIT system requires reliable processes, short changeovers, capable suppliers, accurate information and rapid problem escalation.

3.1 Kanban

Kanban is a visual authorisation mechanism that controls replenishment. A downstream process withdraws a defined quantity; the withdrawal signal authorises the upstream process or supplier to replenish it. Kanban may use cards, containers, electronic signals or barcode transactions.

A commonly used sizing relationship is:

where:

• n = number of Kanban containers or cards;
• D = demand rate in units per time period;
• L = replenishment lead time in the same time unit;
• S = safety factor expressed as a decimal;
• C = quantity per container.

The result is normally rounded up to the next whole container because a fraction of a physical container is not operationally usable. The formula assumes relatively stable demand, a known replenishment lead time, consistent container quantity and a safety factor that reasonably represents variability.

Worked example: Kanban quantity

Demand is 240 units per day. Replenishment lead time is 0.5 day, the safety factor is 10% and each container holds 30 units:

The system requires 5 containers after rounding up. The result should be reviewed after observing actual shortages, excess inventory, demand variation and supplier performance.

Kanban is unsuitable as a substitute for solving severe instability. If demand changes sharply, lead time is unreliable or quality is poor, the organisation may need temporary protective capacity, revised policies or a different planning method.

4. Core Lean Practices

4.1 5S workplace organisation

5S is a disciplined method for organising the workplace:

1. Sort: remove unnecessary items.
2. Set in order: locate tools and materials for easy identification and access.
3. Shine: clean equipment and inspect for abnormal conditions.
4. Standardize: establish visual and procedural standards.
5. Sustain: audit, learn and maintain the discipline.

5S is not merely housekeeping. Cleaning can reveal leaks, loose fasteners, wear, contamination and unsafe conditions. Effective 5S uses employee participation and makes abnormality visible. A superficial campaign that relies on photographs and periodic clean-ups will not produce lasting improvement.

4.2 Standardized work

Standardized work defines the current best-known method for performing a task safely, consistently and at the required pace. It normally specifies work sequence, standard in-process stock and takt or cycle-time expectations. It provides a baseline for training, auditing and improvement.

Standardization does not mean preventing innovation. Once a better method has been verified, the standard should be updated. Without a standard, it is difficult to determine whether a change has improved the process or merely changed it.

4.3 SMED

Single-Minute Exchange of Die (SMED) reduces setup or changeover time, with the original ambition of reducing it to single-digit minutes where practical. Its logic is:

1. Observe and document the complete changeover.
2. Separate internal activities, which require the machine to stop, from external activities, which can occur while it runs.
3. Convert internal activities to external activities where possible.
4. Simplify, parallelise and standardize the remaining work.
5. Verify safety and quality after the change.

Examples include preparing tools in advance, using quick-release fixtures, presetting settings, using standard fasteners and assigning parallel tasks. Lower setup time permits smaller batches, better product variety and lower inventory.

4.4 Poka-yoke

Poka-yoke means mistake-proofing. A device or method either prevents an error or detects it immediately before the error is passed to the next stage. Examples include asymmetric fixtures that prevent incorrect orientation, connectors that fit only one way, presence sensors and software interlocks.

A strong poka-yoke addresses the source of error rather than depending on operator memory. It must not create new hazards, disable legitimate work or encourage false confidence.

4.5 Kaizen and structured problem solving

Kaizen is continuous, incremental improvement involving people who perform the work. It may address safety, quality, delivery, cost, morale or environmental performance. A useful improvement cycle is Plan–Do–Check–Act (PDCA): define the problem and target, test a countermeasure, examine evidence and standardize or revise the response.

Good kaizen begins with facts. Common tools include process observation, Pareto analysis, cause-and-effect diagrams, the five-whys technique, check sheets and control charts. The five whys is not a guarantee of causal truth; it is a questioning aid that must be supported by evidence.

5. Agile Production Systems

Agility is the ability to respond rapidly and economically to changing customer requirements, product variety, volume fluctuations and technological conditions. Agile production commonly includes modular product design, flexible equipment, cross-trained employees, rapid changeover, postponement, configurable processes and close customer information.

Lean and agile systems overlap but emphasise different conditions:

The choice is not binary. A business may use lean flow for stable components and agile postponement for final configuration. This hybrid is often called a leagile system. The decoupling point is the location where forecast-driven production changes to customer-order-driven production.

Agility should be measured by response time, schedule adherence under change, mix flexibility, time to introduce products and cost of change—not merely by the number of product variants offered.

6. Resilient Production Systems

Resilience is the capability to anticipate, withstand, adapt to and recover from disruptive events while maintaining critical performance. Disruptions include supplier insolvency, transport interruption, cyberattack, epidemic, energy failure, natural disaster, equipment breakdown, labour shortage and sudden regulatory change.

A resilient system does not attempt to eliminate every risk. It identifies critical vulnerabilities and designs proportionate responses. Useful capabilities include:

• Visibility: accurate information about suppliers, inventory, orders, capacity and dependencies.
• Redundancy: alternative suppliers, routes, equipment or skills where the consequence of failure justifies the cost.
• Flexibility: capacity, labour and process options that can be redeployed.
• Modularity: products and processes designed so that components or operations can be substituted.
• Collaboration: coordinated planning and transparent communication across the supply network.
• Recovery discipline: tested continuity plans, escalation rules and post-event learning.

A basic risk-priority approach considers likelihood, consequence and detectability. Numerical risk scores can help organise discussion, but they should not be treated as objective truth. A low-frequency event with catastrophic consequences may deserve priority even when its simple score is modest.

Balancing efficiency and resilience

Extreme inventory reduction can remove useful protection against long replenishment times. Conversely, excessive safety stock can conceal process problems and consume capital. The appropriate response is segmentation. Critical, high-risk items may require dual sourcing, strategic stock or redesign, while predictable, low-risk items can remain under tight JIT control.

Safety stock should be based on observed variability, service requirements and replenishment capability rather than arbitrary rules. Managers should also evaluate time-to-recover and time-to-survive: a system is vulnerable when it will exhaust available protection before supply can be restored.

7. Performance Measurement and Control

Lean, agile and resilient systems require balanced measures. A narrow focus on labour utilisation or machine utilisation may encourage overproduction and inventory. Useful measures include:

• first-pass yield and defect rate;
• on-time-in-full delivery;
• lead time and queue time;
• changeover duration;
• schedule adherence;
• inventory turns and days of supply;
• overall equipment effectiveness (OEE);
• employee safety and near misses;
• supplier delivery and quality performance;
• recovery time after disruption;
• customer complaints and response time.

OEE is often represented as:

Availability, performance and quality are expressed as proportions between 0 and 1 when multiplied. Availability commonly reflects operating time compared with planned production time; performance compares actual speed with ideal speed; quality compares good output with total output. Organisations must define the measures consistently, because different treatment of minor stops, planned downtime or rework can make comparisons misleading.

Metrics should be used for learning and system improvement, not for blaming individuals. A target without a problem-solving process may produce data manipulation, unsafe speed or hidden inventory.

8. Integrated Implementation Roadmap

A practical implementation can follow these stages:

1. Establish purpose and sponsorship. Link improvement to customer value, safety, business strategy and workforce capability.
2. Define the value stream. Select a product family and map current material and information flow.
3. Stabilize basic conditions. Address safety, quality, equipment reliability, skills, housekeeping and data accuracy.
4. Measure the baseline. Record lead time, takt, cycle times, inventory, defects, changeover and delivery performance.
5. Improve flow and pull. Balance work, remove bottlenecks, reduce batch sizes and introduce controlled replenishment.
6. Develop people. Train employees in standard work, visual management and structured problem solving.
7. Pilot and learn. Test changes in a contained area, verify results and identify unintended effects.
8. Standardize and spread. Update procedures, train affected personnel and replicate only proven practices.
9. Build resilience. Conduct disruption scenarios, review dependencies and define recovery actions.
10. Sustain improvement. Use audits, leadership routines, employee suggestions and periodic value-stream reviews.

Technology can support this roadmap through real-time equipment data, digital Kanban, traceability, simulation and predictive maintenance. Artificial intelligence may identify patterns in demand, quality or equipment condition, but it depends on reliable data and appropriate human oversight. An algorithm that optimizes local utilization may increase total inventory or reduce resilience. Digital tools should therefore support, not replace, process understanding and managerial judgement.

9. Illustrative Case: Orion Medical Devices

Background

Orion Medical Devices assembles a family of portable diagnostic units. Demand has become less predictable because hospitals increasingly request customised configurations. The plant experiences long changeovers, late supplier deliveries, excess work-in-process and occasional shortages of electronic modules.

Problem

The assembly area produces in large weekly batches based on a forecast. Average customer lead time is 12 working days, although actual processing time is less than one day. Changeovers take 150 minutes. A single overseas supplier provides a critical module, and planners compensate for uncertainty by holding large quantities of less critical components.

Analysis

A value-stream map shows that most elapsed time is waiting between test, assembly and final configuration. The forecast-driven batch system creates overproduction of standard configurations while customised orders wait. Operators use different methods, generating variation and rework. The supplier is judged mainly on unit price, not total risk or delivery performance.

Approach and solution

Orion first establishes standardized work and 5S in the assembly cells. It records the changeover sequence and moves preparation, tooling and documentation outside machine downtime. Quick-release fixtures and preset programs reduce changeover from 150 to 45 minutes.

The company then introduces a pull loop for common modules, using containers sized from measured demand and replenishment lead time. Final configuration is postponed until the customer order is confirmed. Two qualified regional suppliers are developed for the critical electronic module, while a modest strategic buffer is retained because qualification and replenishment cannot occur immediately after a disruption.

Cross-training enables employees to move between test and configuration. Daily tier meetings review safety, quality, delivery, cost and risk. A poka-yoke fixture prevents incorrect cable orientation, and electronic traceability links test results to each unit.

Outcome

After a controlled pilot, average lead time falls, changeover capacity increases and work-in-process is reduced. The plant is able to accommodate more configuration variety without returning to large finished-goods inventories. The second-source arrangement costs slightly more per module, but management accepts the cost because the previous single-source exposure threatened hospital deliveries.

The result is not a pure lean system. Stable upstream work uses pull and low inventory, while the final stage uses agile postponement and flexible labour. Resilience is improved through supplier diversification, visibility, standard work and tested recovery procedures.

Learning points

1. Mapping the complete value stream prevents local improvements from shifting waste elsewhere.
2. SMED supports both lean inventory and agile product variety.
3. Resilience may justify selective buffers and supplier redundancy.
4. Employee skills and problem-solving routines are strategic capabilities.
5. Performance improvement must include delivery and risk, not only cost.

10. Challenges, Ethics and Governance

Lean implementation can fail when managers treat it as a headcount-reduction programme. Removing buffers without improving capability transfers pressure to employees and suppliers. Work intensification may create ergonomic injury, fatigue and unsafe shortcuts. Ethical implementation requires consultation, safe staffing, fair redeployment and respect for worker knowledge.

Supplier relationships also require governance. Excessive price pressure can undermine supplier resilience, labour conditions and quality. Responsible production managers consider modern slavery risks, environmental impacts, data security and continuity across the supply network.

Digital monitoring creates additional concerns. Data collection should have a legitimate purpose, access controls and transparent rules. Employee performance data should not be interpreted without context, and automated decisions should be reviewable. Cybersecurity is part of production resilience because connected machines and planning systems can become disruption channels.

Common technical limitations include unstable demand, inaccurate master data, unreliable equipment, poor process capability, inadequate maintenance and insufficient leadership time. Lean tools cannot compensate for unsafe design, impossible schedules or structurally inadequate capacity. Managers should distinguish a temporary abnormality from a system-design problem and avoid applying one tool to every situation.

Chapter Summary

Lean production creates customer value by removing waste, stabilizing processes and improving flow. Its principal practices include value-stream mapping, JIT, Kanban, 5S, standardized work, SMED, poka-yoke and kaizen. Takt time connects available capacity with customer demand, while process-cycle efficiency reveals the relationship between processing and total lead time.

Agile production prioritizes rapid and economical response to variety and uncertainty through flexibility, postponement, modularity, cross-training and fast changeover. Resilient production adds the capability to anticipate, absorb and recover from disruption. Effective systems often combine lean methods for stable activities with agile and resilience measures at points of uncertainty.

Successful implementation is socio-technical. It requires reliable data, employee involvement, leadership routines, balanced measures, ethical governance and continuous learning. Inventory, supplier redundancy and flexible capacity should be designed selectively according to risk rather than treated as universally good or bad. The strategic objective is a production system that is efficient in normal conditions, responsive to customers and capable of recovery when conditions change.

Key Terms

Lean production: A system for creating customer value with minimal waste and stable processes.

Waste: Any activity or resource use that does not create required customer value.

Value stream: The complete set of activities and information flows required to deliver a product or service.

Value-stream mapping: A visual analysis of material and information flow from supplier to customer.

Takt time: The available production time divided by customer demand for the same period.

Just-in-Time: Supplying or producing the required item, quantity and timing with minimal unnecessary inventory.

Kanban: A visual signal authorizing withdrawal and replenishment.

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

Standardized work: The current best-defined method for performing a task safely and consistently.

SMED: A method for reducing setup and changeover time.

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

Kaizen: Continuous, participative and incremental improvement.

Agility: The capability to respond rapidly and economically to change.

Postponement: Delaying final configuration until demand is known.

Resilience: The capability to withstand, adapt to and recover from disruption.

Decoupling point: The location where forecast-driven activity changes to customer-order-driven activity.

OEE: A composite measure of availability, performance and quality.

Process-cycle efficiency: Value-adding processing time divided by total lead time.

Review Questions

Short Answer Questions

1. What distinguishes value-adding work from non-value-adding work?
2. State the five principles of lean production.
3. What is the purpose of a Kanban signal?
4. Explain the difference between takt time and cycle time.
5. Name three capabilities that contribute to production resilience.

Long Answer Questions

1. Explain the eight forms of waste and show how value-stream mapping can expose them.
2. Describe the stages of SMED and explain how setup reduction supports both lean and agile production.
3. Compare lean and agile production systems in terms of demand, inventory, workforce and equipment.
4. Discuss how 5S, standardized work, poka-yoke and kaizen reinforce one another.
5. Explain why a resilient production system may require selective inventory or supplier redundancy even when lean principles favour low inventory.

Analytical / Application Questions

1. A plant has 450 available minutes per shift and demand of 150 units per shift. Calculate takt time and explain how the result should be used in line balancing.
2. A component has daily demand of 300 units, replenishment lead time of 0.4 day, safety factor of 15% and container size of 40 units. Calculate the required number of Kanban containers, rounding up, and state the assumptions behind the calculation.
3. Design a lean–agile–resilient improvement plan for a manufacturer facing volatile demand, a single-source critical component and frequent changeovers. Justify the location of buffers, flexibility measures and risk controls.

Multiple Choice Questions with Answers

Question 1

Which activity is normally classified as waste in lean production?

A. Transforming a component to the required specification

B. Waiting for material because of an upstream delay

C. Testing a legally required safety characteristic

D. Installing a customer-required feature

Correct Answer: B. Waiting for material because of an upstream delay

Question 2

A factory has 420 available minutes per shift and demand of 140 units per shift. What is the takt time?

A. 2 minutes per unit

B. 3 minutes per unit

C. 4 minutes per unit

D. 5 minutes per unit

Correct Answer: B. 3 minutes per unit

Question 3

What is the primary function of Kanban?

A. Authorise replenishment in response to consumption

B. Increase production regardless of demand

C. Replace all preventive maintenance

D. Measure employee satisfaction

Correct Answer: A. Authorise replenishment in response to consumption

Question 4

Which 5S stage establishes and documents consistent workplace practices?

A. Sort

B. Shine

C. Standardize

D. Sustain

Correct Answer: C. Standardize

Question 5

What is the central purpose of SMED?

A. Increase inspection frequency

B. Reduce setup and changeover time

C. Increase finished-goods inventory

D. Eliminate all product variety

Correct Answer: B. Reduce setup and changeover time

Question 6

Which example best represents poka-yoke?

A. A fixture that prevents a part from being inserted incorrectly

B. A monthly report of total production

C. A larger warehouse for finished goods

D. A forecast of annual demand

Correct Answer: A. A fixture that prevents a part from being inserted incorrectly

Question 7

Which characteristic is most strongly associated with agile production?

A. Rigid schedules and large batches

B. Rapid response to changing product mix

C. Elimination of all buffers under every condition

D. Dependence on a single supplier

Correct Answer: B. Rapid response to changing product mix

Question 8

Why might a resilient system retain a strategic buffer of a critical component?

A. To encourage overproduction of every item

B. To protect against disruption when replacement supply is slow

C. To avoid measuring supplier performance

D. To eliminate the need for quality control

Correct Answer: B. To protect against disruption when replacement supply is slow

Question 9

In OEE, which three factors are multiplied?

A. Cost, revenue and profit

B. Availability, performance and quality

C. Demand, inventory and price

D. Safety, training and morale

Correct Answer: B. Availability, performance and quality

Question 10

A process has 4 minutes of value-adding time and 200 minutes of total lead time. What is its process-cycle efficiency?

A. 0.5%

B. 2%

C. 5%

D. 50%

Correct Answer: B. 2%