Lean, JIT and Pull Production Systems

Introduction and Context

Production Planning and Control (PPC) seeks to deliver the right product, in the right quantity, at the required time, with efficient use of resources. Traditional systems often protect production from uncertainty by creating inventories, large batches and schedules based on forecasts. Lean production takes a different approach: it exposes problems, removes activities that do not create customer value and synchronises operations with actual demand.

Lean, Just-in-Time (JIT) and pull production are closely related but not identical. Lean is a management philosophy and operating system for creating value with minimum waste. JIT is the principle of producing and delivering what is needed, when it is needed, and in the quantity needed. A pull system authorises production through downstream demand rather than allowing each process to produce according to an independent forecast. Successful application requires more than Kanban cards or rearranged machines. It depends on stable processes, capable employees, reliable suppliers, disciplined planning, visual control and continuous improvement.

This chapter examines the practical application of lean tools in production planning and control. It compares push, pull and hybrid systems and evaluates their implementation requirements, risks and performance implications.

Learning Objectives

After studying this chapter, you should be able to:

1. Explain lean thinking, JIT and pull production.
2. Identify and reduce the eight major forms of waste.
3. Calculate takt time and relate it to capacity and work content.
4. Design standard work, cellular flow and basic Kanban systems.
5. Explain heijunka, SMED and value-stream mapping.
6. Compare push, pull and hybrid PPC systems.
7. Assess implementation requirements, risks, performance measures and governance considerations.
8. Apply lean principles to an operational case.

1. Lean Thinking and the Meaning of Value

Lean thinking originated in the Toyota Production System (TPS), although its underlying ideas have wider historical roots in flow production, quality management and industrial engineering. Lean begins with the customer. Value is defined by the customer’s requirements for function, quality, delivery and price. An activity is value-adding when it transforms the product or service in a way the customer requires and is performed correctly the first time.

Non-value-adding activities consume resources without increasing customer value. Some are necessary under current conditions, such as regulatory inspection, but they should still be challenged and reduced. Lean therefore distinguishes:

• Value-adding work: directly changes the product or service in accordance with customer requirements.
• Necessary but non-value-adding work: required by current technology, control or regulation, but not valued by the customer.
• Pure waste: unnecessary activity that should be eliminated immediately.

The five commonly used lean principles are:

1. Specify value from the customer’s perspective.
2. Identify the value stream, including material and information flows.
3. Make value flow by reducing interruptions, queues and rework.
4. Establish pull so that output responds to actual consumption.
5. Pursue perfection through continuous improvement, or kaizen.

Lean is not synonymous with cutting headcount or minimising inventory at any cost. Inventory reduction is useful because it reveals problems, but removing buffers without improving process capability can create shortages and instability.

2. The Eight Wastes

The traditional seven wastes are often expanded to eight. A useful mnemonic is TIMWOODS:

Overproduction is often regarded as the most serious waste because it creates or conceals other wastes. It consumes capacity, occupies space and makes defects harder to detect. In a lean system, the preferred response to poor quality is not more inspection alone, but prevention through process capability, mistake-proofing (poka-yoke) and rapid feedback.

3. Takt Time, Capacity and Line Balance

Takt time expresses the pace of production required to meet customer demand. It is not the same as cycle time. Cycle time is the actual time between completed units at a process; takt time is the allowable interval determined by demand.

Available production time must exclude planned breaks, meetings and other non-production periods. Both terms must use the same time unit, such as seconds per shift and units per shift. The calculation assumes that demand is represented over the selected planning period and that the available time is realistically usable.

Worked example. A factory has 7.5 hours of net production time per shift and demand is 300 units per shift.

Net time = 7.5 × 60 × 60 = 27,000 seconds

The line must therefore complete one acceptable unit every 90 seconds on average. A process with a cycle time above 90 seconds is a constraint unless work is redistributed, capacity is added or demand is changed.

For a task or work centre, utilisation relative to takt can be expressed as:

If a station requires 72 seconds per unit and takt time is 90 seconds, utilisation is 80%. The remaining 18 seconds is not necessarily waste: it may be required for variation, minor stoppages or operator movement. Excessive unused time, however, can indicate imbalance.

A theoretical minimum number of stations is:

where N_min is the minimum integer number of stations under ideal balancing assumptions. Actual stations may be greater because tasks cannot always be divided, precedence restrictions exist, and worker or equipment capabilities differ.

Takt time should not be confused with a demand forecast. If demand changes materially, takt must be recalculated. If demand is volatile, heijunka, flexible staffing, overtime rules or a hybrid planning approach may be needed.

4. Standard Work and Flow

Standard work is the current best-known method for performing a task safely, consistently and efficiently. It is not an attempt to suppress employee judgement. Rather, it establishes a stable baseline from which improvement can be measured.

A standard-work combination sheet commonly records manual work time, machine time and walking or waiting. A standard-work chart shows task sequence and timing. Standard work normally specifies:

• takt time;
• work sequence;
• standard in-process stock;
• quality and safety checkpoints;
• abnormal-condition responses; and
• required tools and materials.

Without standard work, managers cannot distinguish process variation from operator variation, and improvement experiments cannot be evaluated reliably. Standards must be visible, trained, audited and revised when a better method is proven.

Cellular Manufacturing

A cellular layout groups dissimilar machines and people needed to make a family of products. Compared with a functional layout, a cell can reduce transport, queue time and hand-offs. A U-shaped cell may allow one operator to serve several processes and improve communication, although the design must respect ergonomic, safety and skill requirements.

Cell design requires product-family analysis. Products should have sufficiently similar processing routes, volumes and quality requirements. A cell may be inappropriate where products are highly customised, equipment is extremely expensive or demand is too low to justify dedicated flow. In such cases, a focused hybrid arrangement may be more suitable.

5. Pull Production and Kanban

In a push system, planned orders are released using forecasts, a master production schedule or material requirements planning (MRP). Each process attempts to meet its schedule, regardless of whether the downstream process is ready. In a pull system, a downstream withdrawal or customer order authorises replenishment from the preceding process.

Kanban is a signalling method for controlling replenishment. A Kanban may be a card, electronic message, container, barcode scan or other authorised signal. The signal normally identifies the part, quantity, source, destination and replenishment rules.

A basic Kanban quantity equation is:

where:

• n = number of Kanban containers or production signals;
• D = demand rate in units per time;
• L = replenishment lead time in the same time unit;
• S = safety factor expressed as a decimal;
• C = units per container; and
• ⌈x⌉ = round upward to the next whole number.

This formula assumes relatively stable demand, a defined replenishment lead time and a fixed container quantity. The safety factor should reflect measured variation, not arbitrary padding. Excessive safety stock weakens the problem-revealing function of pull.

Worked example. Demand is 240 units per day, replenishment lead time is 0.5 day, the safety factor is 20%, and each container holds 30 units.

Five containers provide a nominal controlled loop of 150 units. The system should be reviewed when demand, lead time, container size or process reliability changes.

Kanban rules commonly include:

1. Downstream processes withdraw only what they need.
2. Upstream processes produce only against an authorised signal.
3. Defective material is not passed forward.
4. The number of signals is controlled and visible.
5. Abnormal conditions are escalated rather than hidden by unauthorised production.

Kanban is less effective when demand is highly erratic, products are engineered to order, lead times are unreliable or quality is unstable. Electronic Kanban can improve traceability, but digital signalling does not correct poor process discipline.

6. Heijunka: Smoothing the Production Schedule

Heijunka means production levelling. It aims to smooth product mix and volume over time so that processes and suppliers experience a more stable workload. Suppose weekly demand is 500 units: 300 of product A and 200 of product B. A batch schedule that produces all A first and all B later creates uneven material and labour requirements. A level sequence might distribute production as A-B-A-A-B repeatedly, subject to feasible quantities and setup times.

Heijunka does not eliminate demand variation. It decouples short-term production rhythm from every fluctuation in customer orders by using a controlled finished-goods or supermarket buffer. It is most effective when:

• products can be produced in repeatable families;
• setup times are short enough to permit mixed-model production;
• demand data are reliable over the planning horizon; and
• quality and equipment availability are stable.

A level schedule can conflict with urgent customer priorities. Therefore, the organisation needs explicit rules for exceptions, allocation of scarce capacity and management of the frozen planning window.

7. SMED and Changeover Reduction

Single-Minute Exchange of Die (SMED) is a method for reducing changeover time, often targeting a changeover completed in fewer than ten minutes, although the underlying principle is more important than the literal target.

Changeover work is classified as:

• Internal: can be performed only while the machine is stopped.
• External: can be performed while the machine is operating.

SMED typically involves observing and recording the current changeover, separating internal and external activities, converting internal work to external work, simplifying adjustments, using quick-release devices, preparing tools in advance and standardising the new method.

Shorter changeovers reduce the economic and operational pressure to produce large batches. They support heijunka, lower finished-goods inventory and make problems visible sooner. Safety cannot be compromised; stored energy, guarding, lifting and verification requirements remain mandatory.

8. Value-Stream Mapping

Value-stream mapping (VSM) is a visual method for representing material and information flow from supplier through production to customer. A current-state map records process times, changeover times, uptime, inventory quantities, information signals and waiting between processes. A future-state map proposes a better flow and pull design.

Important distinctions include:

• Processing time: time during which the product is actively transformed.
• Lead time: elapsed time from release or receipt to completion.
• Value-added ratio:

The ratio is descriptive, not a complete measure of performance. Some necessary activities may not be customer-value-adding, and a high ratio alone does not guarantee quality or resilience.

A VSM event should have a defined product family and scope. The team walks the process, uses observed data rather than assumptions, identifies bottlenecks and information delays, then converts the future-state design into an implementation plan. VSM should not become a drawing exercise: each improvement should have an owner, timing, expected result and review method.

Figure: A simplified pull value stream showing forward material flow and backward demand signalling.

9. Push, Pull and Hybrid PPC Systems

The choice between push and pull should be based on product characteristics, demand behaviour, process capability and supply-chain risk.

A hybrid system may use MRP to plan purchased components and capacity, a supermarket pull system between fabrication and assembly, and final assembly triggered by customer orders. The decoupling point is the location where forecast-driven production changes to order-driven fulfilment. Good hybrid design makes ownership, signal logic and exception rules explicit.

10. Implementation Requirements and Risks

Lean implementation is a socio-technical change programme. Essential conditions include:

• senior leadership that reinforces principles rather than demanding isolated cost cuts;
• reliable measurement of demand, lead time, quality and downtime;
• process capability and preventive maintenance;
• trained, multi-skilled and empowered employees;
• supplier collaboration and agreed replenishment standards;
• visual management and rapid escalation of abnormalities;
• disciplined problem-solving using methods such as A3, five whys and cause-and-effect analysis; and
• a controlled pilot before wider deployment.

Common risks include:

1. Inventory removal without stability: shortages and missed deliveries increase.
2. Local optimisation: one department improves its utilisation while total lead time worsens.
3. Failure to protect people: work intensification, unsafe motions or inadequate staffing undermine sustainability.
4. Poor data: incorrect container quantities and takt calculations create false confidence.
5. Demand volatility: a rigid pull loop cannot absorb extreme variation without a buffer or capacity response.
6. Supplier fragility: frequent small deliveries may increase transport exposure and disruption.
7. Technology-first implementation: software is installed before process rules are understood.
8. Backsliding: signals multiply, standards become outdated and unauthorised production returns.

Lean performance should therefore be balanced across delivery, quality, cost, flexibility, safety, employee wellbeing and resilience. Measures may include first-pass yield, on-time delivery, lead time, WIP, schedule adherence, changeover time, equipment availability, safety incidents and improvement closure rate. No single metric should dominate. For example, maximising labour utilisation may encourage overproduction, while minimising inventory may reduce resilience during a supply disruption.

11. Technology, Digital Lean and Governance

Barcode scanning, RFID, manufacturing execution systems, machine monitoring and electronic Kanban can improve visibility and traceability. Analytics can identify recurring downtime, predict maintenance needs and compare actual cycle time with takt. Artificial intelligence may support demand sensing, anomaly detection and sequencing, but its outputs require data governance, human review and clear accountability.

Digital systems should preserve the logic of the operating system. A poorly designed electronic Kanban can automate excess production just as efficiently as manual cards. Data definitions, access controls, cybersecurity, audit trails and system availability are important governance requirements. Employees should understand how performance data are used; monitoring should not become an unjustified surveillance mechanism. Lean transformation should also consider environmental effects, including transport emissions, packaging, energy use, scrap and disposal of obsolete inventory.

12. Applied Case: Illustrative Electronics Assembly Cell

Background

The following is an illustrative case. Orion Devices assembles a standard control module in two shifts. Customer demand averages 640 modules per day. Net available production time is 14 hours per day after breaks and planned meetings. The assembly area is arranged by function, and finished modules are produced in large batches.

Problem

The plant reports a 12-day production lead time, although the actual assembly work is approximately 18 minutes per module. Work-in-process is high, changeovers take 95 minutes and final delivery performance is 82%. Operators frequently wait for components, while finished goods of low-demand variants accumulate.

Analysis

Net time is:

14 × 60 × 60 = 50,400 seconds per day.

Takt time is:

The team maps the product family and observes that material travels between four departments. It identifies long queues, duplicated inspection and a schedule that releases work according to monthly forecasts rather than assembly consumption. A 95-minute changeover encourages large batches.

Approach and solution

The team first creates a cross-trained U-shaped cell and establishes standard work. Total work content is 300 seconds per module, so the theoretical minimum number of stations is:

Four stations are feasible only after task precedence and ergonomic constraints are checked. The team performs a SMED workshop, preparing fixtures externally and using quick-release clamps. Changeover falls to 28 minutes. A supermarket is placed before final assembly, and a Kanban loop is introduced. Demand is 640 units per day, replenishment lead time is 0.25 day, safety factor is 15%, and container size is 40 units:

The production schedule is levelled across the main product variants. Quality is moved closer to the source through first-piece approval, poka-yoke connectors and immediate escalation of defects.

Outcome

After three months, illustrative results are: production lead time reduced from 12 days to 4.5 days; changeover reduced from 95 to 28 minutes; delivery performance increased from 82% to 96%; and WIP reduced by 48%. The team does not eliminate all buffers. A controlled supermarket remains because supplier lead time is variable and a critical component has a long replenishment period.

Learning points

The improvement came from combining tools rather than applying Kanban in isolation. Takt exposed the required pace, standard work stabilised tasks, cellular flow reduced movement, SMED enabled smaller batches, heijunka reduced schedule volatility and Kanban controlled replenishment. Management also retained a carefully governed buffer where supply risk justified it. The case demonstrates that lean is a system of mutually reinforcing practices, not a collection of labels.

Best-Practice Checklist

Before implementing a pull system, managers should confirm that:

• customer demand and available time are measured consistently;
• the product family and value-stream scope are defined;
• process quality and equipment reliability are understood;
• standard work exists at the relevant operations;
• replenishment lead times and container quantities are verified;
• abnormal conditions have escalation rules;
• safety stock and decoupling points have explicit justification;
• employees and suppliers have been trained;
• performance measures include safety, quality, delivery, cost and resilience; and
• a review process will adjust signals as conditions change.

Chapter Summary

Lean production creates customer value with fewer resources by identifying and removing waste, improving flow and developing people. JIT seeks to provide what is needed, when it is needed and in the required quantity. Pull production uses downstream consumption to authorise replenishment, while push systems use forecasts or planned orders. Hybrid systems combine both approaches at suitable decoupling points.

Takt time converts demand into a required production rhythm. Standard work creates a stable baseline; cellular layouts shorten movement and improve flow. Kanban controls replenishment, heijunka smooths volume and mix, SMED reduces changeover time, and value-stream mapping links material and information analysis to a future-state design. Implementation requires reliable processes, accurate data, employee involvement, supplier coordination and balanced performance measures. Lean tools cannot compensate for unsafe work, unstable quality or unmanaged demand volatility. Sustainable lean PPC is therefore both an operational and managerial system.

Key Terms

Lean production: A system for creating customer value while systematically reducing waste.

Just-in-Time (JIT): Producing and delivering the required item, quantity and timing with minimal unnecessary inventory.

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

Push system: A system in which production is released according to forecasts or planned orders.

Takt time: Available production time divided by customer demand.

Cycle time: Actual elapsed time between completed units at a process.

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

Cellular manufacturing: Grouping equipment and people to support flow for a product family.

Kanban: A visual or electronic signal authorising replenishment or production.

Heijunka: Production levelling by volume and product mix.

SMED: A structured method for reducing setup or changeover time.

Value-stream mapping: A visual analysis of material and information flow across a value stream.

Kaizen: Continuous, incremental improvement involving those who perform the work.

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

Supermarket: A controlled inventory point from which downstream users withdraw material.

Work-in-process (WIP): Items currently between the start and completion of production.

Decoupling point: The location separating forecast-driven activity from order-driven activity.

First-pass yield: The proportion of units meeting requirements without rework.

Review Questions

Short Answer Questions

1. Distinguish takt time from cycle time.
2. List the eight wastes represented by TIMWOODS.
3. What is the purpose of a Kanban signal?
4. Explain the difference between internal and external changeover work.
5. Why is standard work important for continuous improvement?

Long Answer Questions

1. Explain the five principles of lean thinking and discuss how they influence PPC.
2. Compare push, pull and hybrid production systems, including suitable application contexts.
3. Describe how heijunka, SMED and Kanban reinforce one another.
4. Explain the stages and managerial value of a value-stream mapping exercise.
5. Discuss the organisational, technical and human requirements for sustainable lean implementation.

Analytical / Application Questions

1. A plant has 8 hours of net production time per shift and demand of 480 units. Calculate takt time and explain the consequence for a workstation with a 70-second cycle time.
2. Demand is 360 units per day, replenishment lead time is 0.4 day, the safety factor is 25% and container capacity is 24 units. Calculate the required number of Kanban containers, showing your assumptions.
3. A company has unstable suppliers, highly variable demand and long changeovers. Recommend whether it should use push, pull or a hybrid system. Justify your answer using risk, inventory, responsiveness and implementation requirements.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes takt time from cycle time?

A. Takt time is based on customer demand, while cycle time is the actual process interval.

B. Takt time measures setup duration, while cycle time measures transport distance.

C. Takt time is always shorter than cycle time.

D. Takt time applies only to service operations.

Correct Answer: A. Takt time is based on customer demand, while cycle time is the actual process interval.

Question 2

A plant has 28,800 seconds of available time and demand for 360 units. What is the takt time?

A. 60 seconds per unit

B. 80 seconds per unit

C. 100 seconds per unit

D. 120 seconds per unit

Correct Answer: B. 80 seconds per unit

Question 3

Which action is most directly associated with SMED?

A. Increasing the number of finished-goods batches

B. Separating internal and external changeover activities

C. Replacing all pull signals with forecasts

D. Increasing inspection after production

Correct Answer: B. Separating internal and external changeover activities

Question 4

In a pull system, production is normally authorised by:

A. Downstream consumption or an authorised replenishment signal

B. Maximum machine utilisation

C. An annual budget alone

D. The largest possible batch size

Correct Answer: A. Downstream consumption or an authorised replenishment signal

Question 5

What is the main purpose of heijunka?

A. To eliminate all inventory immediately

B. To level production volume and product mix

C. To increase functional departmental separation

D. To replace standard work with operator discretion

Correct Answer: B. To level production volume and product mix

Question 6

Demand is 200 units per day, lead time is 0.5 day, safety factor is 10%, and container size is 20 units. How many Kanban containers are required using the stated formula?

A. 4

B. 5

C. 6

D. 10

Correct Answer: C. 6

Question 7

Which activity is normally classified as pure non-value-adding waste?

A. Transforming material into a required product feature

B. Waiting for an avoidable approval or material shortage

C. Meeting a legally required safety specification

D. Performing a customer-required product test

Correct Answer: B. Waiting for an avoidable approval or material shortage

Question 8

What is a principal benefit of cellular manufacturing?

A. It increases distance between successive operations.

B. It reduces flow visibility by separating related work.

C. It can reduce transport, queues and hand-offs for a product family.

D. It requires every product to use identical processing routes.

Correct Answer: C. It can reduce transport, queues and hand-offs for a product family.

Question 9

Which statement about value-stream mapping is most accurate?

A. It analyses only machine depreciation.

B. It represents material and information flow from supplier through customer.

C. It is useful only after all improvement projects are complete.

D. It replaces the need for observed process data.

Correct Answer: B. It represents material and information flow from supplier through customer.

Question 10

Why might a hybrid PPC system be appropriate?

A. It uses forecasts and pull controls at different stages according to operating conditions.

B. It prevents any use of inventory or buffers.

C. It requires every process to follow customer orders directly.

D. It eliminates the need for supplier coordination.

Correct Answer: A. It uses forecasts and pull controls at different stages according to operating conditions.