From Lean to Resilient: Why Manufacturers Are Rethinking Just-in-Time After Years of Disruption

Mid-sized manufacturers are now carrying an average of 11% more raw material safety stock than they did in 2021. That figure, drawn from a survey of 340 facilities by the Industrial Supply Chain Research Group, understates the structural nature of the shift. It is not a temporary response to a specific shortage event; it is a recalibration of the baseline model that production planners are using to set inventory policy. Just-in-time, the philosophy that dominated manufacturing operations thinking for three decades, is not being abandoned. But its application is being interrogated in ways it never was when supply chains ran smoothly, and the conclusions are landing differently depending on which sector you are in.

What Lean Got Right, and Where It Failed

The intellectual foundations of just-in-time and lean manufacturing are not in dispute. Eliminating inventory as a buffer for poor process performance is correct. Inventory held to cover for unreliable machines, inconsistent supplier quality, or inaccurate forecasting is waste, and reducing it creates genuine competitive advantage through lower carrying cost, shorter lead times, and tighter quality feedback loops. Toyota's production system, the origin point of modern lean practice, was built on a specific factory environment: highly stable demand patterns, a close geographic supplier network, and deep collaborative relationships with tier-one partners. Within that envelope, the model worked as designed and still does.

The failure was not in lean's core logic but in how it was applied, particularly the assumption that supply chains would remain within the operating envelope for which lean's buffer minimisation was designed. When semiconductor shortages in 2021 halted vehicle assembly lines with zero component stock on hand, or when port congestion created three-week delays for materials that plants had assumed would arrive in five days, the system had no recovery mechanism. Lean practice does not ignore variability; it addresses variability through process improvement. What it did not adequately account for was correlated, systemic disruption affecting multiple supply tiers simultaneously, the kind of event where no amount of process improvement at the plant level provides a solution.

The honest assessment is that most manufacturers did not implement lean correctly in the first place. They reduced inventory without achieving the process stability and supplier reliability that lean theory requires as preconditions. The result was a system that looked lean on paper but lacked the collaborative supplier relationships and rapid-response capability that make lean resilient in practice. The pandemic did not expose a flaw in lean theory so much as it exposed how far short of that theory most real implementations fell.

Sector Differences and the Case for Tiered Buffers

The recalibration now underway is not uniform. Aerospace and defence manufacturers have moved most aggressively toward what supply chain planners call tiered buffer models: deliberate safety stock held at specific points in the bill of materials for components that have long lead times, single-source supply, or high consequence of shortage. The driving logic in aerospace is asymmetric cost: a $40,000 stock position in a critical fastener eliminates the risk of a $4 million assembly line stoppage. When the consequence ratio is that extreme, carrying the inventory is the rational economic choice regardless of lean orthodoxy. Several major defence primes have now formalised this calculus into their supply chain policy, replacing blanket just-in-time targets with component-level buffer rules tied to lead-time risk scores.

"We spent years measuring inventory turns without measuring what we actually needed to measure: how fast we could recover when something went wrong. Those are completely different questions."

Caroline Adeyemi, Chief Operations Officer, Veritek Industrial Systems

Automotive has taken a middle path. The semiconductor shortage forced a direct confrontation with the cost of zero buffer on critical long-lead components, and most major OEMs have since established minimum strategic stock positions for semiconductors and other constrained electronics. For shorter-lead, multiple-source commodity components, lean principles remain largely intact. Consumer goods manufacturers, operating with high-velocity demand and perishable or space-constrained inventory, have made the smallest adjustments. Their recalibration has focused less on raw material safety stock and more on supplier diversification: holding lean inventory targets constant while adding qualified second-source suppliers to reduce the probability of a single-point supply failure rather than buffering against it.

• Aerospace and defence: Tiered buffer models formalised in policy, with component-level safety stock rules based on lead-time risk scores and shortage consequence ratios.
• Automotive: Strategic stock positions for long-lead constrained components, lean principles retained for short-lead commodity items.
• Consumer goods: Supplier diversification as the primary resilience lever, with lean inventory targets largely maintained to manage carrying cost and space constraints.
• Mid-market industrials: Broad 11% average safety stock increase, often implemented without a formal buffer policy framework, representing reactive rather than strategic adjustment.
• Capital equipment: Focus on critical spare parts inventory at the facility level rather than raw material buffers, given the cost asymmetry of unplanned downtime.

The risk in the current environment is that manufacturers who have increased safety stock without a supporting analytical framework are carrying cost without commensurate protection. A blanket 15% increase in safety stock across all raw materials does not deliver the same resilience value as a targeted 40% buffer on three specific high-risk components and no change on twenty low-risk ones. The manufacturers building sustainable resilience into their supply chains are doing the harder analytical work: scoring components by lead time, source concentration, and shortage consequence, then sizing buffers in proportion to actual risk rather than applying a uniform policy. That approach requires more upfront work but delivers better risk-adjusted outcomes and, critically, is far easier to defend to finance leadership when inventory carrying costs come under scrutiny.

Measuring Resilience Without Sacrificing Efficiency Metrics

The measurement challenge is real. Efficiency metrics, primarily inventory turns and working capital, are well-established, broadly used, and directly tied to financial reporting. Resilience, by contrast, has no standard metric, which means it tends to lose budget debates against costs that show up clearly on a balance sheet. The most useful resilience metric identified across the manufacturers studied is recovery time: how long does it take to restore normal production output after a supply disruption event above a defined severity threshold? That metric is concrete, measurable, and directly tied to the business consequences that resilience investment is meant to prevent. It also avoids the trap of treating inventory level as a proxy for resilience, which encourages blanket stock increases rather than targeted risk reduction.

Manufacturers that have successfully sustained both lean efficiency and improved resilience have done so by treating them as separate policy dimensions rather than as a single dial to be adjusted. Lean discipline governs process performance and inventory reduction for predictable, stable material flows. Resilience policy governs buffer sizing and supplier structure for components where disruption risk is material. Keeping those two policy regimes analytically separate allows operations leaders to make explicit, defensible choices about where the organisation accepts lean efficiency and where it pays for resilience, rather than making a vague, organisation-wide shift that satisfies neither goal well.