Lean, Just-in-Time and Replenishment Systems

Introduction and context

Lean inventory management is the disciplined design of material flows so that customers receive value with minimal waste, delay and unnecessary stock. Just-in-Time (JIT) is a central lean approach: materials are supplied in the required quantity, at the required time and at the required point of use. Replenishment systems translate this philosophy into operating rules for purchasing, production, warehousing and distribution.

Lean does not mean simply reducing inventory. Inventory is a buffer against uncertainty, but it can also conceal unreliable processes, poor quality, long changeovers and inaccurate records. Effective lean management removes the underlying causes of excess stock while preserving adequate protection against genuine variation. The managerial challenge is therefore to reduce inventory without making the supply chain fragile.

Learning objectives

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

• explain lean inventory, JIT, pull systems and the seven forms of waste;
• distinguish push, pull, Kanban, supermarket, vendor-managed inventory, consignment and continuous replenishment;
• calculate basic Kanban quantities and interpret their assumptions;
• evaluate the benefits, conditions and risks of low-inventory systems;
• design practical implementation and control measures;
• assess resilience, governance, ethical and technology implications.

1. Lean inventory principles

Lean thinking focuses on customer-defined value and the elimination of activities that consume resources without creating value. In materials management, waste commonly appears as:

1. Overproduction—making or ordering earlier or in greater quantity than required.
2. Waiting—materials, people or information idle between processes.
3. Transport—unnecessary movement of goods.
4. Over-processing—more handling, inspection or specification than required.
5. Inventory—raw materials, work-in-process or finished goods beyond what is needed.
6. Motion—unnecessary movement by employees.
7. Defects—scrap, rework, returns and associated disruption.

Many organisations also identify underused human capability as an additional waste. Lean therefore combines flow improvement, standard work, problem solving, employee involvement, visual management and continuous improvement (kaizen).

A useful distinction is between necessary stock and symptom stock. Necessary stock protects against lead-time variation, demand variation, batch constraints or strategic risk. Symptom stock compensates for poor quality, unreliable suppliers, inaccurate bills of material, weak maintenance or inefficient scheduling. Lean improvement targets symptom stock first.

Push and pull control

A push system releases work according to a forecast, plan or schedule, regardless of whether the downstream process is ready. A pull system authorises replenishment in response to actual consumption or a valid downstream signal. Pull does not eliminate planning: forecasts remain important for capacity, procurement and aggregate planning, while actual demand governs short-term replenishment.

A practical lean system often combines:

• push at strategic and capacity levels, where long-term forecasts are necessary; and
• pull at execution levels, where consumption signals control replenishment.

2. Just-in-Time as an operating system

JIT is more than frequent deliveries. It is an integrated operating philosophy involving stable processes, short lead times, small batches, dependable quality, reliable equipment, trained employees and close supplier relationships. Its objectives include reduced throughput time, lower work-in-process, rapid problem detection and improved responsiveness.

JIT exposes problems rather than solving them automatically. When excess stock is removed, a machine breakdown or supplier defect can stop production. This is intentional only when the organisation has the capability to identify and correct the cause quickly. Reducing stock without improving process capability is inventory starvation, not lean management.

Important enabling practices include:

• takt-based planning, aligning production pace with customer demand;
• level scheduling (heijunka), smoothing product mix and volume where feasible;
• quick changeover, reducing the economic and operational penalty of small batches;
• total productive maintenance, improving equipment availability;
• quality at source, preventing defects from moving downstream;
• standard work, defining the safest and most repeatable method;
• supplier development, improving quality, lead time and communication;
• visual management, making abnormal conditions immediately visible.

Takt time is a planning rate, not a guaranteed processing time. Under stable demand, it is calculated as:

Available production time must exclude planned breaks and other unavailable periods, and both terms must use compatible time units. Takt time should not be confused with cycle time: cycle time is the actual time to complete one unit at a process. For a feasible flow, the effective cycle time of each required operation should generally be no greater than takt time, after allowing for utilisation and variability.

3. Kanban and pull systems

Kanban is a signalling method that authorises production, movement or purchase. A Kanban may be a card, electronic message, barcode, container, label or approved signal. The signal is not merely a stock record; it represents permission to replenish a defined quantity under defined conditions.

Common Kanban types include:

• withdrawal Kanban, authorising movement from a supplying process;
• production Kanban, authorising production of a specified item and quantity;
• supplier Kanban, transmitting a replenishment requirement to an external supplier;
• electronic Kanban, integrating the signal with an information system.

A basic container-based Kanban quantity can be estimated as:

where:

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

The result should be rounded up to a whole number because a partial container or signal is normally impractical. The formula assumes reasonably stable demand, a known replenishment lead time, fixed container quantity and a safety factor that adequately represents variation. If demand or lead time is highly variable, the result should be validated using service-level analysis or simulation rather than treated as exact.

Worked example

A line consumes 480 components per eight-hour shift. Replenishment takes 0.25 shift. Management uses a 10% protection factor and containers hold 30 components.

The operating requirement is therefore 5 containers/signals, subject to trial and review. Five containers provide 150 units of nominal capacity. The calculated demand during lead time plus protection is 132 units; rounding up is necessary because 4.4 containers cannot be operated as a complete-container system. If the line consumes 480 units over eight hours, the average demand rate is 60 units per hour, and the assumed lead time is two hours.

Kanban discipline requires standard quantities, clear routes, defined collection frequency, accurate locations, rapid escalation of shortages and a rule that no production occurs without a valid signal. Lost cards, unauthorised substitutions and unrecorded emergency issues undermine the system.

4. Supermarkets and decoupling points

A supermarket is a controlled inventory point holding standard quantities of frequently used items. The downstream process withdraws what it needs; the upstream process replenishes exactly what was withdrawn. It is analogous to a retail supermarket: the customer takes from a visible stock location and replenishment restores the consumed quantity.

Supermarkets are useful when processes have different operating rhythms, geographical separation, setup constraints or transport frequencies. They decouple processes without relying on unlimited work-in-process. However, they do not remove inventory; they locate and control it. The stock level should reflect demand, replenishment lead time, variability and the required service level.

A supermarket should have:

• defined item identity and location;
• fixed minimum and maximum quantities or container counts;
• clear ownership and replenishment responsibility;
• FIFO or expiry control where relevant;
• cycle-counting and discrepancy investigation;
• an escalation rule for empty or overfilled locations.

5. Replenishment models beyond Kanban

Vendor-managed inventory

Under vendor-managed inventory (VMI), the supplier monitors the customer’s inventory or consumption data and decides when and how much to replenish within agreed rules. The customer gains lower administrative workload and potentially better availability; the supplier gains demand visibility and may coordinate production and transport more effectively.

VMI requires accurate, timely data; agreed ownership and service rules; compatible information systems; trust; and clearly defined responsibility for forecast exceptions, obsolete stock, quality and transportation. VMI is not automatically supplier-owned inventory. Ownership, payment and risk must be specified contractually.

Consignment inventory

In consignment, the supplier places stock at the customer’s site but retains ownership until the customer uses it or an agreed transfer event occurs. Consignment can reduce the customer’s working-capital burden and protect availability. The supplier, however, carries financing, obsolescence and demand risk. Contracts should define title transfer, insurance, shrinkage, counting, deterioration, expiry, returns and liability for obsolete goods.

VMI and consignment may be combined, but they are conceptually different: VMI concerns replenishment decision authority, while consignment concerns ownership and payment timing.

Continuous replenishment

Continuous replenishment uses frequent, data-based signals to maintain agreed stock levels or respond to consumption. It is common in retail and distribution, where point-of-sale data, inventory positions and shipment records support regular replenishment. Compared with traditional periodic ordering, it can reduce order spikes and improve product availability, but it may increase transport frequency unless deliveries are consolidated.

A simple order-up-to policy illustrates the logic. If IP is inventory position and T is the target level, the replenishment quantity is:

Inventory position normally includes on-hand stock plus scheduled receipts minus backorders or committed demand. The policy is meaningful only if the definitions are consistent and the target level is periodically reviewed. It should not be confused with a universal optimal-order formula.

6. Designing a low-inventory system

Implementation should proceed as a controlled redesign rather than a stock-reduction campaign.

Step 1: Understand demand and process capability

Segment items by demand pattern, value, criticality, lead time and supply risk. Map material and information flows. Measure actual consumption, replenishment lead time, defects, changeovers, downtime and schedule adherence. A high-value or critical item may require a different policy from a stable, low-value component.

Step 2: Stabilise the process

Correct inaccurate bills of material, standardise containers and locations, improve equipment reliability, reduce defects and establish accurate transaction discipline. Do not launch Kanban while basic records are unreliable.

Step 3: Choose the control method

Use Kanban where demand is repetitive and item specifications are stable. Use planning systems, reorder-point controls or engineer-to-order processes where demand is irregular or the item is highly customised. Use hybrid systems where some components are pull-controlled and others are planned.

Step 4: Set parameters and pilot

Define container quantity, number of signals, replenishment route, collection frequency, safety allowance, escalation rules and ownership. Pilot one product family or value stream. Observe shortages and excesses, then adjust based on evidence rather than informal pressure.

Step 5: Integrate suppliers and technology

Share consumption, schedules and exception information with suppliers. Electronic Kanban, barcode scanning, RFID, mobile devices and ERP integration can improve visibility, but automation cannot compensate for incorrect master data or poor process discipline. Artificial intelligence can support anomaly detection, lead-time prediction and parameter review, but decisions should remain auditable and subject to human oversight.

7. Performance measurement and economics

Lean performance should be assessed across service, flow, cost, quality and resilience. Useful measures include:

Inventory turns are commonly expressed as:

The numerator and denominator must use the same valuation basis. A higher turns ratio is not automatically better: it may reflect excessive stock reduction, poor availability or emergency purchasing. Balanced measures and trend analysis are therefore essential.

8. Resilience risks and safeguards

Low inventory reduces carrying cost and can expose problems quickly, but it also reduces time to respond to disruption. Risks include single sourcing, geopolitical events, transport interruption, cyberattack, labour shortage, quality failure, demand surges and inaccurate data. Lean systems are particularly vulnerable when managers interpret every buffer as waste without distinguishing protection from inefficiency.

Resilience safeguards include:

• dual or geographically diversified sources for critical items;
• qualified alternatives and approved substitute specifications;
• strategic reserves for long-lead or life-critical materials;
• supplier financial and operational monitoring;
• realistic lead-time and capacity assumptions;
• business continuity and recovery plans;
• scenario analysis and stress testing;
• transparent exception escalation;
• cybersecurity controls for connected replenishment systems.

The appropriate objective is not minimum inventory but minimum total system cost at an acceptable risk and service level. A small increase in stock may be rational when the consequence of a stockout is very large.

9. Ethical, governance and regulatory considerations

Inventory decisions affect workers, suppliers, customers and communities. Extreme delivery schedules can transfer risk to suppliers or encourage unsafe transport and excessive overtime. Responsible management therefore evaluates labour conditions, supplier payment practices, environmental effects of frequent deliveries and the disposal of obsolete stock.

Governance should assign responsibility for parameter approval, master-data changes, emergency overrides, supplier access to data and performance reporting. In regulated sectors, traceability, expiry control, segregation, recall capability and documented authorisation may take priority over pure flow efficiency. Data shared through VMI must be protected and used only for legitimate, agreed purposes.

Illustrative case: lean replenishment in a medical-device assembly plant

Background

A fictional medical-device plant assembled diagnostic kits using 1,200 purchased components. It experienced high inventory, frequent line-side shortages and urgent courier deliveries. Inventory records were only 88% accurate, and suppliers received changing schedules with little notice.

Problem analysis

A material-flow study found that excess stock concealed inaccurate bills of material, unrecorded withdrawals and long changeovers. The plant also relied on one supplier for a sterilised component with a six-week lead time. A blanket reduction in inventory would have created unacceptable service and regulatory risk.

Approach

The plant first corrected item masters, introduced barcode scanning at issue points, standardised containers and established FIFO controls. It created supermarkets for 60 stable, frequently used components and introduced withdrawal Kanban. For the sterilised component, it retained a strategic reserve and qualified a second supplier. The purchasing department and key suppliers shared consumption and exception data through an electronic portal. VMI was used for standard packaging, while ownership of consigned items remained with suppliers until use.

Kanban parameters were calculated from measured consumption and replenishment lead time, then tested for eight weeks. Quality, stockouts and emergency orders were reviewed daily during the pilot and weekly thereafter.

Outcome

After six months, average inventory fell by 24%, line-side shortages fell by 70%, and emergency freight was reduced by 45%. The plant did not apply the same reduction to every item: critical sterilised components retained higher protection, and the second source improved resilience. The improvements depended more on data accuracy, process discipline and supplier collaboration than on the Kanban cards themselves.

Learning points

1. Inventory reduction should follow process stabilisation.
2. Pull control is appropriate for repetitive, visible consumption, not every item.
3. Criticality and disruption consequences must influence safety stock.
4. VMI, consignment and Kanban solve different control problems.
5. Lean results require balanced measures covering service and resilience.

Chapter Summary

Lean inventory management seeks customer value with less waste, but it is not equivalent to indiscriminate stock reduction. JIT links frequent, reliable replenishment with quality at source, short lead times, stable processes, small batches and continuous improvement. Pull systems use actual consumption to authorise replenishment; Kanban provides a disciplined signal, while supermarkets provide controlled decoupling points. Basic Kanban sizing depends on demand, lead time, protection and container quantity, and assumptions must be tested in practice.

VMI transfers replenishment decision responsibility to a supplier, consignment changes ownership timing, and continuous replenishment uses frequent data-based signals. Successful implementation requires accurate data, capable processes, supplier collaboration, clear governance and appropriate technology. Low-inventory systems can improve cost, flow and responsiveness, but they increase exposure to disruption. Resilient design therefore combines lean flow with selective buffers, alternative sources, scenario planning and balanced performance measures.

Key Terms

Lean inventory: Inventory management that removes waste while preserving required service and flow.

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

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

Push system: A system in which production or purchasing is released by a plan or forecast.

Kanban: A physical or electronic signal authorising movement, production or replenishment.

Supermarket: A controlled stock point replenished in response to withdrawals.

Takt time: Available production time divided by required customer quantity.

Cycle time: Actual time required to complete one unit or process cycle.

Heijunka: Production smoothing by volume and product mix.

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

Vendor-managed inventory: Supplier-managed replenishment based on customer inventory or consumption data.

Consignment inventory: Stock located at a customer site but owned by the supplier until an agreed transfer event.

Continuous replenishment: Frequent replenishment driven by current inventory and consumption information.

Inventory position: On-hand inventory plus scheduled receipts minus backorders or commitments.

Safety factor: Additional protection applied to a calculated replenishment requirement.

Lead time: Time between initiating and receiving a replenishment order or signal.

Quality at source: Preventing and correcting defects at the point where they arise.

Resilience: The ability of a supply system to withstand, respond to and recover from disruption.

Review Questions

Short Answer Questions

1. What distinguishes a pull system from a push system?
2. Identify five forms of waste relevant to materials management.
3. What is the purpose of a supermarket in a lean flow?
4. Distinguish VMI from consignment inventory.
5. Why can reducing inventory expose operational problems?

Long Answer Questions

1. Explain the relationship between JIT, quality at source, quick changeover and total productive maintenance.
2. Describe the design and governance requirements of a Kanban system.
3. Compare Kanban, continuous replenishment and reorder-point control, including suitable applications.
4. Evaluate the benefits and limitations of VMI and consignment from both customer and supplier perspectives.
5. Discuss how an organisation can pursue lean inventory while maintaining supply-chain resilience.

Analytical / Application Questions

1. A component has demand of 600 units per shift, replenishment lead time of 0.2 shift, a 15% protection factor and a container size of 40 units. Calculate the required number of Kanban containers, rounding up, and state two assumptions that should be tested.
2. A retailer reports higher inventory turns but also rising stockouts and expedite costs. Diagnose why the turns measure may be misleading and recommend a balanced evaluation framework.
3. Design a replenishment approach for a critical, long-lead medical component and a stable, high-volume packaging component. Explain why the two policies should differ.

Multiple Choice Questions with Answers

Question 1

Which statement best describes a pull system?

A. Production is released solely from an annual forecast.

B. Replenishment is authorised by downstream consumption or a valid signal.

C. All items are purchased in the largest economical batch.

D. Inventory is increased to prevent all possible shortages.

Correct Answer: B. Replenishment is authorised by downstream consumption or a valid signal.

Question 2

Which activity is an example of overproduction waste?

A. Correcting a defective component at the workstation

B. Moving material to its point of use

C. Making products before there is a customer or downstream requirement

D. Training an employee in standard work

Correct Answer: C. Making products before there is a customer or downstream requirement

Question 3

A process has demand of 480 units per shift, lead time of 0.25 shift, a 10% protection factor and containers of 30 units. Using N = (D × L × (1 + S)) ÷ C, how many containers are required after rounding up?

A. 3

B. 4

C. 5

D. 6

Correct Answer: C. 5

Question 4

What is the primary function of a Kanban signal?

A. To authorise replenishment, movement or production

B. To calculate annual financial profit

C. To replace all long-term capacity planning

D. To guarantee that demand will never vary

Correct Answer: A. To authorise replenishment, movement or production

Question 5

Which statement correctly distinguishes VMI from consignment?

A. VMI concerns ownership, while consignment concerns forecasting only.

B. VMI concerns replenishment decision authority, while consignment concerns ownership and payment timing.

C. Both terms always mean exactly the same arrangement.

D. Consignment requires the customer to own stock before receipt.

Correct Answer: B. VMI concerns replenishment decision authority, while consignment concerns ownership and payment timing.

Question 6

A supermarket in a lean system is best described as:

A. An uncontrolled reserve of emergency stock

B. A controlled stock point replenished in response to withdrawals

C. A production schedule based only on forecasts

D. A warehouse used exclusively for obsolete items

Correct Answer: B. A controlled stock point replenished in response to withdrawals

Question 7

Which condition most strongly supports the use of basic Kanban?

A. Highly irregular demand and unique engineered products

B. Stable, repetitive demand and reliable replenishment lead time

C. Unknown item specifications and inaccurate records

D. A supply chain with no defined replenishment route

Correct Answer: B. Stable, repetitive demand and reliable replenishment lead time

Question 8

If inventory position is 250 units and the target level is 700 units, what order quantity results from Q = max(0, T − IP)?

A. 250 units

B. 450 units

C. 700 units

D. 950 units

Correct Answer: B. 450 units

Question 9

Why can a higher inventory-turns ratio be misleading?

A. It always indicates lower cost of goods sold.

B. It may result from excessive stock reduction that causes stockouts and expediting.

C. It measures supplier quality directly.

D. It is calculated using demand forecasts only.

Correct Answer: B. It may result from excessive stock reduction that causes stockouts and expediting.

Question 10

Which is an appropriate resilience safeguard for a critical long-lead component?

A. Eliminating all protection stock regardless of consequence

B. Using a single source to simplify administration

C. Qualifying alternative sources and retaining a strategically justified reserve

D. Ignoring supplier financial and operational risk

Correct Answer: C. Qualifying alternative sources and retaining a strategically justified reserve