Why Would Anyone Add Another Warehouse?
Direct lanes, consolidation, cross-docks, reserve storage and network design.
Part 9 of 18 · Before the Shelf
The extra building looks like waste
Maya has books for six destinations. Six direct loads look simple: one origin, one destination, no intermediate stop. A planner proposes sending the books through a redistribution point instead. The plan adds a building and another handling step, yet claims to save money.
The proposal deserves a route-by-route comparison. The new node might combine small flows into fuller onward movements. It might also turn moving inventory into waiting inventory. Its value depends on the departures it makes possible and the work it introduces.
Name the transformation at each node
A cross-dock transfers inbound goods toward outbound movements with limited storage as its operating intent. It does not promise literally zero waiting. A reserve warehouse holds bulk inventory for later replenishment. A distribution centre can receive, store, consolidate or redistribute depending on its role. A fulfillment facility additionally supports customer-order work. Some buildings combine functions.
A transload changes the transport equipment carrying the goods, often involving unloading and reloading. Break-bulk divides a larger movement into smaller ones. Neither automatically means that the product has changed owner. A network diagram should label the activity instead of treating every warehouse icon as interchangeable.
FHWA's supply-chain modeling framework is a public foundation for considering distribution channels, shipment sizes, modes and transfers together. Our network and costs are teaching examples, not a reported operator design.
Compare two complete routes
Assume six direct loads cost $300 each: $1,800. The alternative is one consolidated inbound load at $600, handling at $200, and onward transport at $600: $1,400 before delay and other differences. The apparent saving is $400.
If the intermediate stop adds an expected $500 of service and inventory consequences, the alternative becomes $1,900 and loses. If it adds $100, it becomes $1,500 and wins. The extra node's value depends on more than its handling charge.
Keep the cost boundary explicit. Are the trucks already committed? Is unused capacity available? Does the new lane make another lane less dense? A network change can move costs elsewhere. An allocated average charge is not always the cost avoided by removing one movement.
A connection is a place and a time
Maya's load arrives at the cross-dock at 14:00. The onward trailer closes at 14:30. Receiving and sorting take an hour. The geographic connection exists; the scheduled connection does not. A plan needs feasible processing time before the departure cutoff.
The map should therefore show lanes, departure frequency, capacity, eligible products and recovery options. A once-a-day lane behaves differently from an hourly lane even if the driving time is identical. Missing one departure may cost a full day.
Space can fill with waiting
For a stable process with a consistently defined boundary, Little's Law relates average inventory in the process to throughput and average time in it. If 100 pallets a day flow through and spend two days on average inside, average work-in-process is 200 pallets. This is a boundary-specific relationship, not a recommended storage target or a guarantee that the system can handle peaks.
A nominal cross-dock can become an unintended storage operation when departures fail. The extra dwell consumes space, creates handling work and delays incoming goods. Looking only at daily receipts misses this accumulation.
Make the extra node earn its place
The initial $400 saving is a budget for everything the new route changes. Waiting, damage exposure, extra inventory and missed connections can consume it. Better load utilization and reliable connections can preserve it. The decision needs the complete route and a departure schedule, not just a lower handling quote.
Ask the planner to explain the 14:00 arrival and 14:30 trailer close. If receiving and sorting take an hour, those books cannot make that connection under the stated assumptions. The next departure determines the delay. That small timing check may matter more than a polished network diagram.
Once the node is operating, follow dwell as well as receipts. A cross-dock that routinely stores two days of work has acquired a storage burden, whatever its label. The right facility is the one whose actual transformation improves the usable-stock promise enough to pay for its cost and complexity.
Public sources and evidence boundary
- FHWA: Supply Chain Modeling Needs and Concepts, modes and transfer
- MIT OpenCourseWare: Lean Thinking Part II (2012)
Public material checked on 18 October 2026. Sources support the named definitions or frameworks; the scenarios, diagrams, calculations and proposed decisions are original teaching examples. No private company systems or performance data are described.