Chapter 4 of 4

Designing for Resilience in Global Chains

3 sections · ~7 min read · 3 practice questions · part of Global Supply Chains

Learning objectives

  • Quantify the expected cost of a disruption on a global lane.
  • Compare dual-sourcing, near-shoring, and safety-stock as resilience strategies.
  • Identify the operational signals that a chain is becoming fragile.

Disruption Cost as an Operating Variable

Disruption cost is the operating cost of a supply shock: lost sales, expedited freight, premium overtime, customer goodwill erosion, and recovery inventory. It is not a hypothetical; it is a recurring expense that varies with chain design. The expected disruption cost is roughly P(disruption) × severity. The probability of disruption is a function of geography, supplier, and category; severity is a function of inventory buffer, alternate capacity, and customer tolerance. Both halves of the equation are design choices. A well-designed global chain has low probability of disruption (diversified suppliers, audited compliance) and low severity when disruption does happen (safety stock, qualified alternates, expedite budget). The right discipline is to size resilience investments the same way you size capacity: model expected cost under a range of disruption scenarios, and choose the level of insurance the company is willing to fund. This chapter deliberately avoids legal and policy commentary; the question is operational design, not regulation.

Resilience Levers: Dual-Source, Near-Shore, Stock

Three levers dominate resilience strategy. Dual-source qualifies a second supplier in another geography; it raises complexity and may reduce scale economies, but limits single-point-of-failure exposure. Near-shore moves a portion of spend closer to demand; it raises unit cost but shortens pipeline and reduces some categories of geographic risk. Safety stock holds buffer inventory at strategic decoupling points; it raises carrying cost but absorbs short shocks without changing the supplier base. The right mix depends on the time horizon of the disruption the chain is most worried about. Short, recurring shocks (3–7 days of port closure, a quality event at one supplier) are best handled by safety stock and expedite budget. Long, structural shocks (a regional capacity loss, a sustained policy shift) are best handled by dual-source or near-shore. The cheapest mix is the one that matches the lever to the shock profile; companies that buy only safety stock are exposed to structural shocks; companies that buy only dual-source are overpaying for short shocks.

Fragility Signals

Some signals indicate a global chain is becoming fragile before a major disruption exposes it. Expedites as a share of freight spend is usually the first to move: when that percentage climbs, the chain is buying last-minute capacity because buffer has run out somewhere. Rising forecast error at the SKU level matters more on long lanes than short ones, because the error has to be covered across a longer horizon. Concentration ratios, meaning the share of category spend held by the top one or two suppliers or regions, indicate shrinking optionality as they climb; the threshold at which that becomes uncomfortable is a matter of company risk appetite and category criticality rather than a universal number, so set it deliberately and write it down instead of adopting a figure from elsewhere. Increasing lead-time variance on inbound lanes indicates that carriers and ports are losing reliability. Average supplier audit scores drifting downward indicate that supplier oversight is weakening. None of these signals alone is decisive; together they form a fragility profile that leadership can act on. The discipline is to monitor these signals in normal operations, not only after a shock. A chain that manages resilience only in crisis mode is a chain that pays the full cost of every shock.

Worked example

Problem

Cascadia Components sources a custom sensor from a single supplier in one country. Annual spend on the sensor is $24M, and it feeds a product line with annual revenue of $180M ($15M per month) at a 35% contribution margin. Two disruption classes are estimated. Class 1: an 8% annual probability of a 30-day outage. Class 2: a separate 2% annual probability of a 90-day outage, from a regional rather than a supplier-specific cause. A 30-day outage costs one month of contribution ($5.25M) plus $1.2M expedite and $0.6M goodwill, so about $7.0M. A 90-day outage costs three months of contribution ($15.75M) plus about $2.25M of extras, so about $18.0M. Compare the status quo against (a) qualifying a second supplier at $1.5M/year fixed plus a 4% unit-cost premium, and (b) holding 30 days of component safety stock at a 22% holding rate. Cost the residual risk under each option, not just the option itself.

Step by step

  1. Status quo expected disruption cost = (0.08 × $7.0M) + (0.02 × $18.0M) = $560,000 + $360,000 = $920,000/year. Both classes must be counted; pricing only the more likely one understates the exposure by 39%.
  2. Option (b), 30 days of buffer. Buffer value at cost = $24,000,000 × 30/365 = $1,973,000. Annual carrying cost = 0.22 × $1,973,000 = $434,000.
  3. Residual risk under option (b): 30 days of stock fully absorbs the 30-day outage, so class 1 expected cost falls to zero. It covers the first 30 days of a 90-day outage, leaving a 60-day event costing roughly two months of contribution ($10.5M) plus $2.0M of extras, so $12.5M. Class 2 expected cost = 0.02 × $12.5M = $250,000.
  4. Option (b) total = $434,000 carrying + $250,000 residual = $684,000/year, against the status quo's $920,000. Net saving = $236,000/year.
  5. Option (a), dual source. Annual cost = $1,500,000 fixed + (0.04 × $24,000,000) = $1,500,000 + $960,000 = $2,460,000. Even if it eliminated disruption risk entirely, it would cost $2.46M to avoid $0.92M, so it loses by roughly $1.54M/year.
  6. Break-even for dual-sourcing: it would need to remove at least $2,460,000 of expected disruption cost, which is 2.46 / 0.92 = 2.7 times the current exposure. That would require severity or probability to roughly triple, for instance if the product line were three times larger or the outage probability rose from 8% to over 20%.
  7. A hybrid of 15 days of buffer ($217,000 carrying) plus partial qualification of the critical tooling step ($400,000/year) costs $617,000 before residual risk, and leaves both outage classes only partly covered, so its all-in figure lands above option (b). At these parameters, buffer stock alone is the efficient answer and the hybrid is not worth its complexity.

Answer. Status quo expected disruption cost is $920,000/year. Option (b), 30 days of buffer, costs $434,000 to carry and leaves $250,000 of residual tail risk, for an all-in $684,000 and a net saving of $236,000/year. Option (a), dual-sourcing, costs $2,460,000 and cannot pay for itself at this exposure; it needs roughly triple the severity or probability to break even. Buffer stock wins here, and the reason is worth generalising: buffer is a variable, right-sizable spend that scales with the days of cover bought, while qualification is a large fixed commitment that only pays back at high exposure. Two limitations. The probability estimates carry more uncertainty than the arithmetic implies, and small changes in them move the ranking; and buffer stock protects only against duration, so it does nothing about a supplier that fails permanently, which may require a qualified alternative supplier.

Practice

Work each question before opening the solution.

  1. What two variables determine expected disruption cost?

    Show solution for question 1

    Probability of disruption and severity when disruption occurs. Both halves are design variables, affected by supplier diversification, geographic spread, safety stock, and expedite capability.

  2. Why is dual-source sometimes more expensive than the disruption cost it avoids?

    Show solution for question 2

    Dual-source adds fixed qualification cost and a per-unit premium from reduced scale economies. For categories with low disruption severity, the cost can exceed the expected avoided cost. Dual-source pays when severity or probability is high enough to justify the structural spend.

  3. Why is expedite spend as a share of total freight a useful early fragility signal, and what can it fail to catch?

    Show solution for question 3

    It is useful because it is a financial measurement of an operational shortfall the organisation is already paying to hide. When buffer, capacity, or lead-time reliability falls short, someone books premium transport to protect the customer, so the service metrics stay green while the cost quietly migrates onto the freight line. That makes it an earlier indicator than fill rate, which only moves once expediting stops working. What it fails to catch is fragility that has no expedite option: a sole-source component with a six-month qualification cycle produces no expedite spend at all, because there is nothing to expedite, and the chain can look calm right up until the supplier fails. Read it alongside concentration and qualification-coverage measures rather than on its own.