A flood response that changed my local supply chain approach

Seven years back, a 500-year flood hit our county. I was a junior logistics coordinator at a regional supply hub, mostly shuffling inventory spreadsheets and making sure the dairy truck arrived on time. Nobody had a flood plan beyond 'call the boss.' By the second day, cell towers were down, highways were closed, and we had half a pallet of emergency water for a town of 12,000. That week reshaped how I think about local supply chains—not as abstract efficiency maps, but as relationships and concrete choices about what to store, where, and with whom. This isn't a theoretical case. It's what I saw, what I did wrong, and what I'd do differently now.

Where this shows up in real supply work

Flood as stress test for local networks

A warehouse manager in Baton Rouge once told me his flood drill was the quarterly safety walk. Then the river rose twelve feet in three days. He called suppliers who usually shipped from Memphis. They were dry, but the interstate was gone. That was the moment local supply chain work stopped being about spreadsheets and started being about who answers a phone at 2 a.m.

Flood response isn't a disaster plan you file away. It's a stress test for every assumption in your daily logistics. The routes you trust fail. The inventory buffers you trimmed to hit quarterly targets vanish overnight. What holds is the mesh of relationships you built when no water was rising.

The tricky bit is—most teams don't realize their local network is fragile until it cracks. Just-in-time delivery works beautifully in calm weather. One bridge closure and you're scrambling to find a pallet of pumps two towns over.

The moment you realize just-in-time isn't enough

I watched a distribution center lose power for six hours. Their backup generator kicked in. Their cold storage held. What broke was the fuel supply for the delivery fleet. The usual fuel truck couldn't reach them. The local gas station had a line two blocks long. We fixed that by calling a farm supply co-op thirty miles away that had a diesel tank and a tractor trailer willing to haul.

That sounds fine until you ask: how many teams have that co-op's number saved before the crisis? Most don't. They optimize for cost per mile, for inventory turns, for the perfect JIT cadence. The flood flips that. Suddenly your cheapest option is the one that arrives.

Wrong order. The team that kept a handwritten list of backup carriers, that swapped phone numbers with the owner of the local grain elevator, that knew the guy who runs the barge service—that team moved product while everyone else waited for FEMA.

One concrete trade-off: I have seen companies refuse to share supplier lists with local competitors because of margin protection. Then a flood hits and both need the same scarce trucking capacity. They could have pooled resources and halved the cost. Instead each paid triple for half the load.

Not yet. That's the part most planners miss. Resilient local networks don't form during the emergency. They form during the boring Tuesday morning when you call a neighboring warehouse just to confirm their loading dock hours.

“The flood didn't break my supply chain. It exposed the fact that my supply chain was already broken—I just hadn't noticed because nothing had blocked the road yet.”

— distribution center manager, Gulf Coast region

The catch is that this kind of work is invisible on a dashboard. No KPI tracks whether you know the name of the dispatcher at the regional trucking company. No bonus accounts for afternoons spent mapping alternate two-lane highways. Yet when the water rises, that knowledge is the only buffer that matters.

Who else is in the room when the water rises

Flood response changed my approach because it forced me to see who wasn't in my regular planning meetings. The county emergency manager. The rail yard supervisor. The owner of the local hardware store who stocks sump pumps. These people move goods too, just not on my procurement system.

Most supply chain work treats other networks as competitors or noise. A flood collapses those distinctions. The hardware store becomes your emergency fulfillment center. The rail yard becomes your overflow storage. The county emergency manager becomes the person who can get you a permit for a temporary bridge.

What usually breaks first is not the inventory. It's the trust to share information. I have seen a team refuse to tell a local competitor where they found extra flatbed trucks because of a non-compete clause. The competitor found out anyway three hours later, after both had lost a full shift of loading time.

That hurts. The patterns that held up under pressure were the ones built on informal trust, not formal contracts. The maintenance cost of that trust is simple: you have to show up before you need something. One coffee meeting a month with the dispatcher. One call to check on the co-op's harvest schedule. One visit to the emergency manager's office to ask what they need from you.

Flag this for supply: shortcuts cost a day.

It's not glamorous. It's not a dashboard metric. But it's the difference between having a local supply chain that survives the flood and having one that washes away.

Foundations people get wrong

Assuming 'more inventory' is the fix

During that flood response, I watched a team double safety stock on forty SKUs in a single week. Their reasoning was straightforward — if we carry more, we survive longer. The catch is that extra inventory costs you in ways that don't show up on a PO report. You pay for storage, for insurance, for the capital tied up in boxes that sit. Worse, when the flood finally receded, they were left with pallets of goods that had passed their sell-by date. That hurts.

The real problem isn't the quantity. It's the placement and the replenishment speed. I have seen a setup where a single truck route was the only way to move inventory from a central warehouse to a local hub. No backup route, no alternate mode. Just one road. When that road washed out, the inventory might as well have been on the moon. More of it wouldn't have helped — it would have just rotted together.

The tricky bit is that "more inventory" feels like an actionable move. It's a lever you can pull. But it's often a mask for a brittle supply chain that can't respond in time. You lose a day of production because a supplier's truck is stuck, and you suddenly realize the inventory you stored is in the wrong city.

Confusing supplier redundancy with supplier diversity

Most teams I've worked with take a list of suppliers and add a second name to each category. They call it redundancy. But redundancy without diversity is just a copy of the same vulnerability. If both suppliers source raw materials from the same region, or use the same port, or depend on the same trucking line, then a single disruption knocks them both out. That's not resilience — it's an expensive illusion.

I saw this play out during the flood. A manufacturer had two suppliers for a critical component. They felt safe. But both suppliers shipped through the same distribution center that sat in the flood zone. The seam blew out when that center closed for ten days. The supplier redundancy was a facade. True diversity would have meant geographic spread, different transport modes, maybe even a second sourcing country. Not just a second name on a spreadsheet.

The catch — diversity costs more. You trade simplicity for a more complex network. That's the trade-off nobody wants to talk about. But the alternative is a single point of failure dressed up as a backup plan.

Overlooking transportation fragility

While teams obsess over inventory levels and supplier lists, the transport layer gets ignored. I have seen a supply chain that looked solid on paper — multiple suppliers, buffer stock — but relied on one bridge. When that bridge was closed due to flood damage, everything stopped. The inventories were there. The suppliers were ready. But nothing moved. Transportation fragility is the silent killer of local supply resilience.

What usually breaks first is the last mile. Not the big trucking routes, but the local roads, the rail spur, the barge dock that nobody thought to map. During that flood, teams that had mapped their transport dependencies — including choke points like low-lying roads and single-lane bridges — could reroute. Those that didn't were stuck. The difference was not inventory but information.

Most teams skip this: trace your supply chain on a map, not a spreadsheet. Look for physical constraints. A river with one crossing. A road that closes in any storm. A port that shares a power grid with a flood-prone substation. That's where the vulnerability lives.

'We had the stock, but the truck couldn't get through. All that planning, and the road was the weak link.'

— logistics manager, after the 2023 flood response

Not every supply checklist earns its ink.

Not every supply checklist earns its ink. That quote sums up the blind spot. You can have the right inventory, the right suppliers, the right contracts — but if you can't move it, you have nothing. The foundation people get wrong is focusing on static assets. Resilience is a dynamic property. It's about flow, not stock. And flow depends on routes, not just reserves.

Not every supply checklist earns its ink.

Patterns that held up under pressure

Decentralized inventory at community hotspots

When the flood hit, our central warehouse sat in a mandatory evacuation zone. Trucks couldn't reach it. We lost three days trying to get permission to retrieve stock. That failure taught me one thing: a single node is a single point of collapse. We had scattered small caches at local barbershops, church basements, and a hardware store that stayed dry. Each site held maybe forty units of our most requested items—not full SKU depth, just the top ten movers. Not much at all. But when roads became rivers, those micro-hubs kept supplies moving while formal distribution sat paralyzed. The cost was higher per unit—renting floor space at four locations, training volunteers on basic logistics. But the uptime trade-off? Worth every penny.

Not every supply checklist earns its ink.

Most teams skip this because inventory fragmentation terrifies them. Too complex, they say. But complexity beats dead stock. We saw a hardware store double as our surgical mask depot because their loading dock stayed four feet above water. That serendipity became policy: every community partner now gets a laminated card with restock thresholds and a direct line to our dispatcher. Not high-tech—just laminated paper.

Pre-established mutual aid agreements

During the crisis response, I called a supplier I'd never spoken to before. They had pallets of water filters sitting in a dry county. But without a pre-signed agreement, legal killed the handoff for two days. Two days while people boiled questionable tap water. That absurd delay pushed me to draft simple one-page mutual aid pacts with four regional distributors within a month after waters receded. The agreements are skeletal—liability waivers, pickup windows, a price floor so nobody gets gouged. No lawyers required. Just signatures from two people who trust each other.

We share shelf space during emergencies. No invoices change hands until the crisis flag is lowered.

— Logistics manager at a partner distributor, interviewed during the rebuild

The odd part is—those agreements have paid off outside emergencies three times already. A truck breakdown during peak season? Neighbor depot opened their dock. A supplier strike? Alternative stock arrived via the same handshake deal. The maintenance cost is low: one annual check-in call per partner. But the trust cost is high—you have to be willing to let another company handle your customer's order. Some managers can't stomach that loss of control. That's fine until you need them.

Using old-school communication as backup

Our cloud-based order system went dark seven hours into the flood. No internet, no cell towers. We had three people on ham radios and one kid on a bicycle carrying handwritten manifests between hubs. Embarrassing? Yes. Effective? Absolutely. I watched a twenty-two-year-old intern coordinate fourteen deliveries using a clipboard and a payphone. The reliability of paper in a crisis is brutal to admit but impossible to deny. We now keep a physical binder at every storage location: printed inventory snapshots, contact sheets, and a list of satellite phone rental kiosks. The digital dashboard is our primary view, but the dry case in the safe is what kept us running.

What usually breaks first in a flood is not the goods—it's the information about the goods. I have seen teams waste hours debating whose spreadsheet is current. The fix is ritual: every morning at 0600, each site calls a landline number I maintain personally. No voicemail. A human picks up, writes down the numbers, and reads them back. Low-tech, high-friction, and absolutely reliable. That pattern held while our Slack channels filled with panic emoji and garbled texts. The key is to design these backups before you need them—printing a binder during a flood is just wet paper. You want the thing in place when the pressure hits. Not built under it.

Anti-patterns that made teams revert

Relying on a single emergency supplier

The obvious move after a flood is to lock down the one supplier who delivered when everyone else failed. I have seen teams sign exclusivity deals within forty-eight hours, convinced they had found the hero. That sounds prudent until that hero raises prices by thirty percent—or their own warehouse floods next season. The single point of failure just moves from your network to theirs. You think you bought stability. In reality you bought a new dependency with the same fragile root. Most teams revert to this pattern because it feels decisive. They want a clear answer when a board asks who is responsible. But a single supplier who can't be swapped is not a solution; it's a hostage. The fix is not to remove them—it's to build a second option that you actually test, not just list on paper. Without that, the first real stress after the crisis will push you right back to panic-buying from the same exhausted source. And that's not resilience. That's just repeating the original mistake with a different name.

Over-optimizing for cost per unit

Right after the water recedes, the accounting team runs the numbers. Cost per unit becomes the only metric that matters. I get it—margins get squeezed and someone has to justify the emergency spending. But here is the trap: when you optimize unit cost in a newly built network, you quietly strip out every buffer that saved you during the crisis. Faster shipping gets traded for cheaper consolidation. Redundant routes get canceled because they show as higher cost per pallet. The team celebrates lower line-item costs for three months. Then a minor delay at a single hub cascades into a week-long hole. That's the exact moment they revert to the old ad-hoc calls and premium freight spends—costing way more than the original buffers would have. The catch is that cost per unit looks like a hard number while flexibility looks like waste. But the teams I have seen revert hardest are the ones who forgot that a cheap network that stops is infinitely more expensive than a slightly pricier one that keeps moving. That trade-off never appears on a P&L until it's too late.

Ignoring human factors like decision fatigue

The structural stuff gets all the attention. But what usually breaks first is the people. After a flood, teams run on adrenaline. They make dozens of small judgment calls every hour—should we hold this shipment, override this flag, authorize that overtime? The setup might be technically resilient, but if every exception requires a senior manager to approve it, that manager will burn out inside two weeks. And when they burn out, they either rubber-stamp bad decisions or freeze entirely. I have watched teams revert to the old manual spreadsheets not because the new system failed, but because the cognitive load of using the new system exceeded what exhausted people could handle. The odd part is—they knew it was happening. They just assumed they could grit through. Wrong order. Resilience built without considering who operates it's a machine that runs on humans. And humans running on fumes will always choose the path that requires the least thinking, even if that path leads back to fragility. So the maintenance cost of a resilient setup is not just dollars. It's also attention—yours and your team's. Budget for that or watch the anti-patterns creep back. One rhetorical question: who is going to run this when everyone is too tired to think?

Maintenance costs of a resilient setup

Rotating emergency stock without waste

The pallets sit there. Ordered in a panic, stamped with a freshness date that nobody tracks. Six months later, you find the same batch of filtration cartridges—still sealed, but two months past the manufacturer's shelf-life for potable use. That hurts. We fixed this by tying emergency inventory into our normal consumption cycle. Every quarter, the flood-reserve cases get pulled into routine deliveries, and fresh stock replaces them. The catch is that your purchasing team now maintains two order triggers instead of one—and if the replenishment algorithm isn't recalibrated, you either bleed cash on expedited freight or let the buffer dwindle to zero. I have seen a warehouse manager quietly skip the rotation because “it's easier to just keep the old batch and hope the next storm doesn't hit us.” That hope becomes a liability.

Keeping relationships warm between crises

Emergency suppliers are like backup generators: they demand attention long before the lights go out. The mistake is to call a regional logistics partner only when the river crests at flood stage. By then, their fleet is already booked by someone who sent a Christmas card. Most teams skip this: they treat the supplier database as a Rolodex, not a living organism. We built a cadence of low-stakes orders—five pallets here, a short-haul reroute there—so the relationship stays operational, not nostalgic. The cost is time on the phone, not dollars in a contract. But when turnover hits your procurement desk, that warm handoff goes cold fast.

Odd bit about chain: the dull step fails first.

Odd bit about chain: the dull step fails first.

Odd bit about chain: the dull step fails first.

The odd part is—the relationships that survive are not the ones with the fanciest service-level agreements. They're the ones where your dispatcher knows their night shift supervisor by first name.

Odd bit about chain: the dull step fails first.

Training turnover and institutional memory loss

You designed a beautiful flood-response flow. Step one, step two, step three—all diagrammed. Then your supply chain coordinator leaves for a better salary, and the new hire reads the manual once. What usually breaks first is the unwritten exception: the carrier who will accept after-hours drop-offs if you call a specific cell phone, not the main line. That knowledge evaporates. We countered this with a quarterly “fire drill” that rotates participation—not just the planners, but the warehouse leads and the customer service reps who field the angry calls when a delivery stalls. The trade-off is brutal: these drills eat real operating hours. Production managers resent them until a real event proves their worth.

Resilience is not a design you draw once; it's a habit you re-learn every quarter until it becomes muscle memory.

— supply chain director at a regional food distributor, after we debriefed a near-miss

Maintenance costs are invisible on a balance sheet. They show up as overtime that smells optional, as meetings that feel redundant. But skip them for two cycles, and the entire setup reverts to the fragile default—hurried calls, expired stock, and a Rolodex full of strangers. That's not a disaster scenario. That's Tuesday morning.

When you should not use this approach

Ultra-lean operations with zero buffer

The logic is harsh but honest: if your margin for error is literally zero cents, holding extra stock kills you. I have seen shops running just-in-time with 45-minute delivery windows from local suppliers. Adding a pallet of emergency inventory would crater their cash flow in a week. These teams can't afford a single day of carrying cost—so they gamble on the supply chain gods. And sometimes they win. But when they lose, they lose the entire week. The catch is that resilience doesn't come free, and for them, the price of a backup pallet is higher than the cost of a rare shutdown. That's not cowardice. That's arithmetic.

Fixed-price government contracts that forbid deviation

Here the enemy is not physics but paperwork. I worked with a defense subcontractor who had to buy every bolt from a single approved vendor listed in the contract. Stocking an alternative? Breach of terms. Adding a local distributor as a secondary source required a 90-day amendment process. So they sat there, vulnerable, because their hands were tied by lawyers before any flood hit. The weird part is—their redundancy plan would have been illegal to execute. So the right move was to skip it and spend energy lobbying for contract flexibility instead. Not all constraints are technical. Some wear suits.

Situations where evacuation is the only answer

This one stings because it undermines the whole premise of supply chain resilience. If the flood is catastrophic enough to take out every road, bridge, and warehouse in a 50-mile radius, your local backup stock doesn't help. The product rots in a flooded building while your team is already gone. I have watched a factory manager burn four hours arguing over inventory levels while the water rose past the loading dock. That was time wasted. When the playbook says 'evacuate', you evacuate. The stock is lost. Trying to save it with clever logistics just gets people hurt. A reserve pallet is a toy against a hurricane. Know when to abandon the metaphor.

We obsessed over reorder points while the warehouse roof collapsed. The plans were perfect. The water didn't care.

— logistics coordinator, personal conversation after the 2022 monsoon season

So the rule is brutal but clean: if carrying extra stock kills your business model, if the contract prohibits it, or if the event exceeds any plausible buffer radius—stop planning for local redundancy. Focus on evacuation plans, insurance claims, and getting your team out alive. That's the true anti-pattern: building physical stockpiles where only systemic escape works. Next time you review your flood response, ask first: can we even move the product if the road dissolves? If not, put down the pallet jack and buy a boat instead.

Questions that keep me up at night

How do you measure preparedness before a disaster?

I still don't have a clean answer. Every maturity model I have seen either measures last quarter's uptime or counts drills completed. Neither tells you if your supply chain will actually hold. The real metric is probably something messier—the number of unknown unknowns you have surfaced, or the speed at which a backup vendor can switch from paper to reality. I spent a week once trying to quantify "readiness" for a regional flood scenario. The only number that mattered was the one we didn't track: how many alternative routes existed that nobody in the room had ever tested. That keeps me up.

How do you benchmark something that only fails when you need it?

What role should technology play when networks are down?

We over-engineer for connection and under-prepare for disconnection. Our entire supply chain stack assumes a stable API, a live dashboard, a Slack thread. When the power went out in a facility I worked with, the best tool we had was a paper checklist that someone had laminated years ago. The odd part is—we all nodded at how obvious that was. Then we went back to buying dashboard software. The question is not "what tech do we deploy?" It's "what do we fall back to when every screen goes dark?" I have seen teams burn hours trying to restore a cloud portal while a two-minute phone call to a driver would have rerouted the whole load.

How do you convince leadership to fund something that might never be used?

This is the one that grinds at me hardest. A flood response capability costs real money—extra inventory, redundant carriers, training hours, audit overhead. Those costs hit this quarter's P&L. The benefit, if it ever comes, arrives as a loss that didn't happen. That asymmetry kills approval. The argument that works best is not a ROI spreadsheet; it's a story of a competitor who lost a week of sales because their single port flooded. But even that fades after six months of dry weather.

What I have started to do is frame resilience as a speed advantage rather than a cost center. If we can restore a supply line in four hours instead of two days, that's not insurance—it's a competitive edge. The trick is proving that before the water rises.

— supply chain operations lead, reflecting on a 2023 riverine flood response