Life science supply chain resilience: how to plan
General
Key takeaways
Resilience can be measured at each node by comparing time to survive with time to recover, counted until customers receive normal supply.
Dual sourcing shortens recovery, but a second site can take years to qualify, and FDA reporting categories depend on the product and the change.
Buffer stock should cover the recovery gap for the disruption you plan for, capped by the dating customers can actually use.
A cold chain plan works lane by lane, with packaging time, triggers, a named decision maker, an approved hold point and a rehearsal.
Tariffs, FDA shortage notifications, DSCSA and the BIOSECURE Act can each change which sources and backups remain usable.
Life science supply chain resilience is the ability to keep medicines, biologics, samples and critical materials reaching patients and study sites when a supplier, plant, shipping lane or trade rule fails. It can be measured. For each weak point, compare how long you could keep supplying without it against how long it would take to restore normal supply. Where recovery takes longer, you have a gap, and dual sourcing, buffer stock and cold chain backup plans are the ways to close it.
This guide applies that test to each lever, using FDA rules, WHO guidance and published supply research. It stays at the planning level. Tariff rates, GPS tracking, storage requirements and partner selection each have a dedicated Mercury guide, linked where they come up.
What is life science supply chain resilience?
Life science supply chain resilience is the margin at each point in the network between the time you can keep meeting demand after a failure and the time a full recovery takes. MIT researchers working with Ford use two measures for this. Time to survive is the maximum period supply can match demand after a disruption. Time to recover is how long a node takes to return to full function.
One finding from that work carries over to a life science supply chain. Ford's largest exposure did not sit with its biggest strategic suppliers. It tended to come from tiny suppliers of parts costing about 10 cents, so a review based on spend would have missed it.
Recovery also has to be measured all the way to the customer, as the 2024 IV fluid shortage shows. Hurricane Helene flooded Baxter's North Cove, North Carolina plant in late September 2024, and the American Hospital Association said the site made 60% of the country's IV solutions. Baxter reported a return to pre-hurricane production levels in February 2025, less than five months after the storm, and announced on May 13, 2025 that allocations were lifted for virtually all North Cove IV solution codes. FDA declared the end of the national shortage of 0.9% sodium chloride IV solutions on August 8, 2025, with other IV fluids still in shortage. The plant's recovery date and the market's recovery date were months apart, and a plan built on the first number understates the gap.
Where a biotech supply chain differs
A biotech supply chain often has less slack than a large manufacturer's network. One CDMO may hold the only validated process, one cell bank may sit behind every batch, and clinical supplies may move on small, irregular lanes with little stock at the site. The consequences differ as well. A commercial product risks a shortage, while a clinical program can also lose dosing windows, samples and data.
Biotech supply chains also change quickly. A program moving from Phase I toward commercial supply changes its volumes, sites and countries, so a gap estimate from last year may no longer describe the network you run today.
How do life science companies build a resilient supply chain?
Life science companies build a resilient supply chain by mapping the nodes that can stop supply, measuring the gap at each one, closing the largest gaps with the lever that fits, and rehearsing the response. A workable life sciences supply chain strategy follows this order:
Map the nodes that can stop supply. Include suppliers, CDMOs, test labs, packaging components, storage sites and transport lanes, and go past tier one.
Estimate time to survive. Count stock that is released, in range and reachable. Leave out quarantined batches and stock with too little remaining dating to sell or use.
Estimate time to recover. Run the clock until customers receive normal supply, including filings, qualification, release testing and transit.
Rank the gaps by consequence. Put life-supporting products, active dosing and irreplaceable samples ahead of products with substitutes.
Choose a lever for each gap. Compare the options in the table below on cost, lead time and what they leave exposed.
Name owners and rehearse. Decide who declares a disruption, who can approve spending and who speaks to regulators, then review the plan when a site, lane or trade rule changes.
This sequence is close to what FDA's 2022 draft guidance on risk management plans recommends, with risk assessment, risk control and scheduled review built on ICH Q9 principles. The guidance is a non-binding draft. The statute behind it, section 506C(j) of the FD&C Act, does require manufacturers of certain critical drugs and their active ingredients to maintain a redundancy risk management plan, and FDA can inspect it. ICH Q9(R1) adds that the formality of this work should match the risk of losing the product's availability, so a small biotech can keep the plan short as long as it is specific.
The levers do different jobs. This table compares what each one buys and what limits it.
Lever | What it buys | What limits it | Where it fits |
|---|---|---|---|
Second qualified source | Shorter recovery after a supplier or site failure | Filing and qualification time, shared inputs | Long recovery, high patient impact |
Risk mitigation inventory | Longer time to survive | Shelf life, cold capacity, cash | Stable demand, long dating |
Reserve capacity at the primary site | Faster make-up after a short stoppage | No help if the site itself is lost | Short, contained interruptions |
Backup lane or carrier | Shorter recovery after a transit failure | Commodity acceptance, qualified packaging | Every critical temperature-controlled lane |
These choices interact, so they belong in one decision. A 2017 study in Omega, built around a large drug company's risk problem, found in its model that inventory and reserve capacity can substitute for each other while no second source exists. Once a second source exists, reserve capacity and the second source appear to substitute, and the optimal inventory level falls. The result comes from a model and a worked example, so treat it as a reason to budget the levers together, with your own figures.
Dual sourcing and what a second source takes
What is dual sourcing in life sciences?
Dual sourcing in life sciences means qualifying two suppliers or manufacturing sites for the same material, product or service, so that either can supply under your quality system and, for regulated products, your filings. The second source may run a share of volume every month or stay qualified on standby. Running volume keeps the process and the working relationship active, while standby costs less and can drift out of date.
A second source mainly shortens time to recover, and creating one can take a long time. The Omega study describes establishing and registering a second manufacturing site for all or major markets as a strategic project that often takes up to five years because of regulatory complexity. A second supplier of a simple consumable can be far quicker, so estimate each case separately.
Filings set the timeline
For an approved US drug, 21 CFR 314.70 sorts changes by their potential to harm identity, strength, quality, purity or potency. Major changes need an approved supplement before distribution. Moderate changes can usually use a supplement that allows distribution no sooner than 30 days after FDA receives it. FDA's guidance on changes to an approved application recommends a prior approval supplement when a new site lacks a satisfactory CGMP inspection for the type of operation being moved, and it lists certain aseptic transfers as major changes.
That guidance excludes specified biotechnology products, so biologics teams start from 21 CFR 601.12. Its listed examples of changes needing approval before distribution include a change to the source material or cell line and a new master cell bank. In clinical development, sponsors report new chemistry and manufacturing information to an IND through information amendments.
The practical step is to ask regulatory affairs for the filing route and its likely duration before procurement commits. That estimate belongs in your time to recover, and the category depends on the product and the change, so it should not be assumed from a previous project.
Check that the two sources are independent
Two sources that buy the same starting material or sit in the same flood zone can fail together. ICH Q9(R1) makes a similar point: supply chain diversity can support availability, but complex chains create interdependencies that can introduce systemic risk. Ask each source where its critical inputs come from, then compare the lists.
Prioritize by gap rather than spend. FDA's draft guidance asks manufacturers to check whether container-closure suppliers are sole sources, and a missing stopper can hold up a batch as surely as a missing active ingredient.
The same check applies to logistics. A sudden provider closure shows the dependency at the carrier level, and Mercury's first-mile planning guide recommends naming a secondary carrier for each critical lane and confirming it accepts the commodity.
When a full second source is not justified
An early-stage biotech may not have the volume or budget for a second CDMO. ICH Q9(R1) mentions a pre-qualified alternative option as a possible response to supplier problems. A completed audit, a draft quality agreement and a technology transfer package kept current can shorten recovery at far lower cost than running two sites, though the filing and validation work still has to happen when the switch is made.
Sizing buffer stock from the recovery gap
How much buffer stock does a life science company need?
A life science company needs enough buffer stock to cover the gap between time to recover and time to survive in the disruption it chooses to plan for, capped by what can be used before it expires. No single number fits every product.
Ordinary safety stock handles normal swings in demand and lead time. The buffer discussed here sits on top of it. The Omega study calls it risk mitigation inventory and defines it as stock beyond cycle inventory, work in progress and safety stock. The company in that study set it to cover the 95th percentile of a disruption risk profile, which assesses a manufacturing site's ability to recover from different events. That is one company's practice, and it shows stock tied to recovery evidence instead of a flat rule.
FDA's draft guidance asks manufacturers to identify weaknesses in inventory management and does not set a number of months. France takes a different approach. Since September 2021 it has required a minimum two-month safety stock for medicines of major therapeutic interest intended for French patients, and the regulator can raise it to as much as four months for products with recurring shortages or shortage risks. Those figures are a legal floor for one market. They are not a benchmark for a biologic with a short shelf life.
A worked example
Consider a hypothetical refrigerated biologic with one fill-finish CDMO, demand of 1,000 units a month and a 24-month shelf life. All figures here are illustrative. The planning scenario is a stopped fill line, and the team estimates eight months from that decision until the first released batch from a backup site reaches customers. Current stock covers two months, so the gap is six months and the buffer is 6,000 units, for 8,000 units in total.
Next, test the plan against dating. If customers need at least six months of dating on receipt, 18 months are usable. At 1,000 units a month with oldest stock going first, 8,000 units turn over in eight months, so the plan fits. With a 12-month shelf life, usable dating falls to six months and part of that stock would expire. The team would then shorten recovery, hold material upstream as drug substance if stability data support it, or accept a smaller buffer and record the remaining risk.
Where the buffer sits
Rotation is a compliance matter as well as a cost one. 21 CFR 211.150 requires the oldest approved stock to be distributed first, with deviation permitted only if it is temporary and appropriate.
Location matters as much as quantity. A buffer in the same building as the primary site can fail in the same flood or power cut. A second temperature-controlled location brings its own pharmaceutical cold storage requirements, and the stock is only useful if a qualified lane can reach customers from there.
Cold chain contingency by lane
How do you plan for a cold chain disruption?
You plan for a cold chain disruption by giving each critical lane a written plan that states how long the payload stays in range, which events trigger action, who can book the alternative, where product can wait under approved conditions and who decides its disposition afterward. Then you rehearse it.
WHO's model guidance on storage and transport sets the baseline. It asks for contingency arrangements covering extended power outages, equipment failure and vehicle breakdown, action plans for product exposed to excursions, and staff who know and have rehearsed the response. It also asks that transport routes be profiled and qualified.
The same test applies here on a shorter clock. For a lane, time to survive is the remaining qualified packaging duration plus any refrigerant you can add. Time to recover is how soon product can reach the receiver or an approved hold point. The moment to act is when the second number closes on the first, which can be well before a logger shows a breach. This table sets out common scenarios.
Disruption | Time check | Backup action | Record to keep |
|---|---|---|---|
Missed flight or connection | Packaging time left vs next departure | Rebook or reroute the shipment | New routing and timestamps |
Customs hold | Expected release vs coolant left | Broker works the hold, agent adds coolant | Hold reason and release time |
Dry ice running low | Sublimation rate vs arrival time | Replenish at a planned point | Weight, time and logger data |
Carrier or courier failure | Time to the pickup deadline | Move to an onboarded secondary | Custody handoff |
Cold room power failure | Time until the room leaves range | Backup power or move stock | Excursion log and stock moved |
Each backup only works if it was arranged in advance, so qualify the secondary packaging and carrier before the first shipment. Mercury's pharma cold chain guide covers lane qualification, and the first-mile guide suggests running one disruption scenario end to end during onboarding. What happens to product after an excursion is a quality decision based on stability data, which Mercury's guide to preventing temperature excursions walks through.
Product-specific rules can change the plan. The MMR vaccine example shows two products for the same use with different storage labels, so a contingency written for one may be wrong for the other.
Visibility that buys response time
Visibility has two layers, and they answer different questions. Network visibility shows who makes what, where, and what they depend on. Shipment visibility shows where a package is and what condition it is in.
The network layer is often the thinner one. USP's 2026 drug shortages report says 44 percent of drugs in shortage have at least one key starting material made solely in a single country, typically China or India, and that the average US shortage now lasts more than five years. USP tracks marketed medicines in the US, not clinical programs. Even so, it shows that concentration in a biopharma supply chain can start well above the supplier you contract with. A usable network map records:
Each manufacturing and testing site by tier and country
Sole-source flags for critical materials, components and services
Lead times and the stock each partner holds for you
Inputs that two or more of your sources share
Contract terms requiring notice of site or supplier changes
Shipment visibility has its own Mercury guides, on GPS tracking and control-center response and on identifiers and handoffs. For resilience, one test matters most: an alert should reach someone with authority to act while time remains in the lane's margin. A live map with no named responder adds little to your time to recover.
Tariff and regulatory risk as supply events
Tariffs can change the cost of a source faster than you can add another one. Mercury's guide to US pharma tariffs sets out the position as of October 2026, where rates depend on patent status, manufacturer, country and HTSUS line. The 2026 Section 232 proclamation gave importers about four to six months before its duties started, while qualifying a new site can take years. Model duty for each qualified source in your tariff planning, and check how a change of route or mode affects cost and transit time before relying on it.
Regulatory duties shape the response as well. FDA says applicants and manufacturers of certain finished drugs and biologics must generally notify it at least six months before a permanent discontinuance or an interruption likely to cause a meaningful supply disruption. If that is not possible, notice is due as soon as practicable, and no later than five business days after the event for finished products. Similar duties cover the active ingredient. A plan should name who makes that call and what triggers it.
Traceability affects backups too. Under DSCSA, companies may trade only with authorized partners and must pass transaction data with each change of ownership, so a backup distributor has to be authorized and connected before the day you need it.
Supplier restrictions are the newer risk for biotech supply chains. The BIOSECURE Act, signed on December 18, 2025, restricts federal procurement from designated biotechnology companies of concern and federal contracts with entities that use their equipment or services in performing the contract. The bar also extends to federal loan and grant funds. One legal analysis says it does not stop a US company from receiving federal money simply because it does business with a listed company. The first list from the Office of Management and Budget is due by December 2026. Federally funded biotechs should check CDMOs and service vendors against it when it appears, and any resulting switch becomes a second-source project with the timelines above.
What to ask of life sciences supply chain services
Life sciences supply chain management solutions, such as planning software and risk dashboards, can hold the map and the numbers. They do not close a gap by themselves, so test a tool on whether it stores time to survive and time to recover by node and surfaces them during a disruption.
Life sciences supply chain services on the logistics side are tested on execution. Ask whether the provider can rebook after hours, replenish refrigerant at a transit point and work a customs hold with the temperature clock running. Mercury's guide to selecting a life science logistics partner has the wider set of questions and a scorecard.
Mercury's part sits in the lanes. Its life sciences logistics service coordinates shipments across CRO, CDMO and research partners through a 24/7 control tower. Its cold chain service plans routes around dry ice and coolant replenishment availability at each transit point and reviews each shipment through an OK-to-Send process before it leaves origin. For recovery, Mercury offers next flight out and onboard courier services around the clock, dry ice replenishment in transit and an in-house customs broker. Second-source qualification and FDA filings stay with your quality and regulatory teams, and each lane's realistic recovery time feeds into the plan they maintain.
Start shipping today!
Start your shipment now — no login required. Fast, secure, and guided by experts.
