This article was originally published on LinkedIn.
(This article is based on an article I published 20 years ago. But that article was completely focused on an industrial setting. This present article takes those concepts and applies them to contested sustainment. Waller, M.A., R. Cassady, J. Ozment. “Impact of Cross Docking on Inventory in a Decentralized Retail Supply Chain,” Transportation Research Part E: Logistics and Transportation Review, 42, (2006), 359-382.)
A logistics support area in a theater exercise. Days to build. Fuel bladders staged. Ammunition stacked. Repair parts sorted and binned.
It is also stationary, concentrated, and radiating signature. In the scenario, it lasts long enough to be found and then it’s neutralized. The supplies survived the trip forward. They didn’t survive sitting there.
That’s the contested sustainment paradox: we work hard to move materiel forward and then we make it easy to target once it arrives.
It is a flow problem. Not a storage problem.
The Storage Default
For decades, the Army’s sustainment architecture has relied on intermediate nodes (distribution centers, supply support activities, brigade/division support areas) as places to accumulate inventory in anticipation of demand. Those nodes buffer uncertainty with mass.
In permissive environments, that makes sense. The cost of excess inventory is money and space.
In a contested environment, the cost structure changes. A static stockpile isn’t just “inventory.” It’s vulnerability: a signature, a protection burden, and a single point of failure that can erase combat power in minutes.
So the question shifts from “How much should we store?” to “What should be stored at all and where?”
A clue comes from an idea the commercial world has been refining for years: cross‑docking.
What Cross‑Docking Actually Means
Cross‑docking is often treated as a warehouse trick. It’s bigger than that.
The core idea is simple: a node can be a coordination point instead of a storage point. Inbound shipments arrive, get broken down, reconfigured, and pushed outbound quickly (hours, not days). Inventory flows through rather than sits.
Years ago, I coauthored research modeling what happens when a distribution center stops holding stock and becomes a flow‑through node. The findings were commercial. But the structure maps to contested logistics, if we adapt it correctly.
The Hidden Cost: Lead Time and Forward Stockage
Cross‑docking removes inventory from the intermediate node. That’s the appeal: less concentration, less “big, juicy target.”
But it also changes the effective lead time seen by the customer.
If the support area no longer holds inventory, a forward unit isn’t waiting on “delivery from nearby.” It’s waiting on the chain to sense demand, source the item, move it to a coordination point, sort/configure it, and push it forward.
Longer lead times require more safety stock to maintain the same service level. If replenishment time increases, you either carry more forward or you accept more stockouts.
That’s not an argument against cross‑docking. It’s a design parameter.
Why Contested Environments Can Favor Cross‑Docking
Combat demand is often correlated. Adjacent units make contact. A brigade shifts axes. Tempo spikes. Consumption rises together.
In retail, that correlation drives the bullwhip effect: small changes at the edge become large oscillations upstream. In a storage‑based architecture, intermediate nodes must carry safety stock to absorb that volatility.
In a cross‑dock architecture, intermediate nodes hold little to no inventory. Buffering shifts upstream (where pooling is stronger) and allocation can happen later (when you know more). The same volatility that makes forward stockpiles fragile can make flow‑through coordination more attractive (because it reduces what you must park, protect, and risk in one place).
But there’s a catch.
The Real Danger: Cross‑Docking Without Decision Alignment
Cross‑docking succeeds or fails on decision architecture.
Cross‑docking performs best when allocation decisions are coordinated (when the system can pool information and make late, informed choices about who gets what).
In a contested fight, communications degrade and units will make replenishment decisions based on what they can see. That is the most dangerous configuration:
Decentralized ordering + limited visibility + longer lead times.
If every unit inflates orders “just in case,” you amplify volatility, overload lift, and recreate signature through emergency movements. You may even increase total inventory (just moved forward and spread across more locations).
So if the Army wants “flow, not storage,” it has to pair cross‑docking with aligned rules, even under degraded networks:
- Priority tiers and guardrails (what auto‑fills, what requires confirmation)
- Late‑binding allocation (delay committing inbound loads to a destination until the last responsible moment)
- Preplanned disconnected operations (demand bands, minimums/maximums)
- Reconciliation when comms return (rebalance, learn, adjust)
This is the difference between a mobile coordination node that reduces risk and a mobile coordination node that becomes a chaos amplifier.
Speed Through the Node Is the Whole Game
There’s one variable that dominates military cross‑docking: time.
Every hour materiel sits at a “flow‑through” point waiting for sortation, load configuration, route clearance, or a movement window undermines the architecture:
- It extends lead time (driving up forward safety stock)
- It increases signature (the node becomes easier to find)
- It recreates the very targetability cross‑docking was meant to reduce
This is where autonomy and pre‑configured loads become strategically meaningful (not because they replace labor, but because they compress time at the point of greatest vulnerability).
The Bottom Line
The contested logistics problem is not fundamentally a storage problem. It is a flow problem.
A sustainment architecture built around “accumulate and hold” will keep producing large, stationary targets no matter how many autonomous vehicles we add.
Cross‑docking (turning intermediate points into coordination points) offers a framework for shifting from stockpiles to flow. But it only works with clear‑eyed design:
- Forward stockage sized for the new lead time reality
- Decision alignment even under degraded networks
- Relentless compression of time through every coordination node
In a contested fight, supplies that are sitting are supplies that are dying.
The goal is to keep them flowing and to design the decisions that make that flow possible.
