Supply chain network design
Supply chain network design determines the physical structure of a supply chain: how many plants and distribution centers to operate, where to put them, which markets each serves, and how product flows between them. It weighs production, freight, inventory, and duty costs against service targets, usually with optimization models run across multi-year demand and tariff scenarios.
Examples
DC consolidation study: An industrial parts maker serves North America from five distribution centers. The model shows three DCs would cut $2.1 million a year in fixed cost but push 9% of orders past the two-day delivery promise. They close one, keep four, and bank $1.2 million without breaking the promise.
Tariff scenario: A 25% duty scenario on imported subassemblies raises Asian landed cost from $38.00 to $46.80 a unit, against $44.00 from a candidate Mexican line. A comfortable gap inverts, and the nearshore option moves from study to board proposal.
Definition
Network design trades four costs against service. Production favors fewer, bigger plants for scale; freight favors more nodes and shorter lanes; inventory punishes every added stocking location with its own safety stock; and duties can flip an entire sourcing geography. Models load demand by market, candidate sites, lane rates, and tariff scenarios, then optimize for total landed cost at a required service level.
The cadence has changed. A network study every 3 to 5 years, triggered by a merger or an expiring lease, made sense when the inputs moved slowly. Tariff swings and supply shocks pushed many companies to refresh annually or keep a living model of the network (a digital twin) so a question like a nearshoring move can be evaluated in days instead of quarters.
The common failure is optimizing for cost alone. A network that is 4% cheaper but concentrates 80% of output in one flood-prone region is a worse network; resilience constraints such as dual regions and maximum single-site share belong in the model, not the appendix.
Frequently asked questions
What is supply chain network design?
Supply chain network design determines the physical structure of a supply chain: how many plants and distribution centers to operate, where to put them, which markets each serves, and how product flows between them. It typically uses optimization models run across multi-year demand and tariff scenarios.
What costs does network design trade off?
Four cost families pull in different directions. Production favors fewer, bigger plants for scale, freight favors more nodes and shorter lanes, inventory punishes every added stocking location with its own safety stock, and duties can flip an entire sourcing geography. Models optimize total landed cost at a required service level.
How often should companies redesign their supply chain network?
A network study every 3 to 5 years made sense when inputs moved slowly. Tariff swings and supply shocks have pushed many companies to refresh annually or to keep a living model of the network, sometimes called a digital twin, so questions like a nearshoring move can be evaluated in days instead of quarters.
What is the most common network design mistake?
Optimizing for cost alone. A network that is 4 percent cheaper but concentrates 80 percent of output in one flood-prone region is a worse network. Resilience constraints, such as requiring dual regions or capping any single site's share of output, belong inside the model rather than in an appendix.