8. Rail & intermodal

Picture an ocean container that just discharged in Long Beach, headed for a DC outside Chicago. From the port to the DC is roughly 2,000 miles. Two ways to cover it:

By truck. A driver pulls the container from the LA terminal, drives it to Chicago in about 2 days. Cost: roughly $4,500 all-in. Reliability: high. Driver is on hours-of-service rules and the schedule is predictable.
By rail. The container gets drayed from the LA terminal to the BNSF Hobart rail yard, loaded onto a double-stack train, railed 2,000 miles to a Chicago-area ramp (Joliet or Cicero), then drayed to the DC. Cost: roughly $3,000 all-in. Reliability: lower. Transit: about 5–6 days.

The truck is faster and more reliable. The rail saves you $1,500 a box. Across 200 boxes a year on this lane, that’s $300k. Why don’t more shippers move to rail?

Mostly inertia. Sometimes ramp positioning. Sometimes service reliability concerns. Rail is structurally underused on most lanes where it’s economically viable, and once you understand the operational constraints, you can identify the lanes where shifting to rail is the highest-ROI mode change in inbound.

Starting cold? Rail is what carries your ocean container the long inland leg in the US (West Coast port to Chicago, etc.). Cheaper and greener than truck on long hauls but has its own reliability and ramp-positioning issues.

The rail product set

Three distinct product families. They share track but little else operationally.

Intermodal (containerized)

Imagine an ocean container loaded onto a railcar, riding 2,000 miles, then drayed to a DC. The container never gets unloaded; it just changes vehicles. That’s intermodal.

Standardized ISO containers (20’, 40’, 40’ HC , 45’) or domestic containers (53’ in North America; 45’ EMU-container in the EU) move on flatcars or double-stack wells.
The same container can ride truck, rail, and ship without unloading. That’s the “intermodal” promise.
Two flavors in inbound:
- International intermodal: an ocean container lands at a port, gets drayed to a rail ramp, rides to an inland ramp near the destination, and is drayed final-mile. " IPI " (Interior Point Intermodal) is the carrier-routed version; " MLB " (Mini Land Bridge) is the all-rail port-to-port-coast-to-coast move.
- Domestic intermodal: 53’ North American container is loaded at origin (e.g., Chicago DC), moves via rail to destination ramp (e.g., LA), delivered by drayage.
Carriers in North America: BNSF, Union Pacific, CSX, Norfolk Southern, Canadian National, CPKC. In the EU: DB Cargo, SNCF Fret, Lineas, CargoNet, Hupac, Kombiverkehr.

Carload

Imagine 80 tons of paper rolls leaving a Wisconsin paper mill. They’re not in a container — they’re loaded directly into a boxcar. That’s carload.

Individual railcars (boxcar, gondola, hopper, tanker, flatcar, centerbeam, etc.), each holding 50–200 tonnes, routed individually across the network.
Priced per car per origin-destination. Used for bulk commodities (paper, lumber, steel, chemicals, machinery) where containerization doesn’t make economic sense.
Switching dominates transit time. A carload spends most of its trip being assembled and disassembled at classification yards.

Unit train

Now imagine 110 hopper cars of grain leaving a North Dakota elevator, all bound for a single Gulf Coast export terminal. The cars stay together as one train, dedicated to one shipper, one origin, one destination. That’s a unit train.

An entire train (50–200+ cars) dedicated to one commodity from one origin to one destination. Coal, grain, crude oil, ethanol, frac sand, aggregates, vehicles.
Most cost-efficient rail product per tonne-km. Requires large volumes, loading/unloading facilities at both ends with unit-train capacity (usually loop tracks), and typically long-term contracts.
Practically invisible to non-bulk inbound planners but the backbone of commodity inbound for industries that live on it.

Where rail wins economically

Rail’s sweet spot is a very specific geometry:

Long haul. Advantage widens above ~800 km (500 mi). Under that, truck wins on total time and often on cost once drayage at both ends is loaded.
High-density, non-urgent freight. The lower the value density and the longer the acceptable transit window, the better rail looks.
Reliable volume. Rail networks run on scheduled trains; single-shipment optionality is low.
Port-to-inland. Ocean to rail ramp is the dominant US inbound intermodal pattern. Roughly half of international containers moving inland from US West Coast ports ride rail.

Rough framework: on a 2,000-mi lane with 20+ FTL-equivalent containers per week, intermodal rail typically saves 20–35% on freight cost against truck, at the price of 2–5 more transit days and slightly higher CoV.

Where rail loses

Rail loses (sometimes badly) when:

Network reliability tanks. North American rail service had a brutal 2022–23 marked by crew shortages, congestion, and the East Palestine derailment’s regulatory fallout. Average on-time performance dropped meaningfully on some corridors. Shippers migrated freight back to truck.
Ramp positioning is wrong. If the closest inbound ramp to your DC is 120+ miles away, drayage cost and time eat the rail advantage.
Volume is sub-scale. Rail ramps charge gate fees and minimum-count surcharges that punish single-container moves.
Service disruption propagates. A washout, derailment, or strike on one corridor can reroute trains thousands of miles, and there is no parallel network to swing to. Truck networks recover in hours; rail in days.
Cross-gauge breaks (relevant outside North America). Russia / CIS use 1520 mm gauge; most of the rest of Eurasia uses 1435 mm. A cross-gauge shipment requires bogie changes or transshipment at border stations, each of which adds 12–36 hours.

International intermodal: IPI, MLB, merchant haulage

Some North-American-specific vocabulary that trips up global shippers:

IPI (Interior Point Intermodal). Ocean carrier’s through-bill service where the ocean BoL covers port-to-inland-ramp-to-your-door. Single contract, single liability. Simpler operationally; usually pricier than the alternative.
Merchant haulage (also called store-door or CY-to-door). You take delivery at the ocean carrier’s inland ramp and arrange your own drayage. Lower ocean-carrier rate, more operational work, more flexibility.
MLB (Mini Land Bridge). Coast-to-coast all-rail routing for ocean cargo, e.g., an Asian cargo discharged at LA/LB then railed to a NY/NJ ramp and cleared there. Cheaper than transshipping through Panama on a TPWB vessel, slower than direct vessel to East Coast, useful for certain inventory positioning plays.
4CB (Four Corners Bridge) and AWB (All-Water Bridge). Carrier-specific product variations; most practitioners won’t encounter these outside major retailer networks.

When an ocean carrier IPIs your inland move and it goes wrong, your contractual remedy is against the ocean BoL, not against the rail carrier. That risk-allocation difference can matter on high-value shipments; merchant haulage gives you a direct line to the rail provider.

EU rail freight: why it’s structurally smaller

Europe moves a smaller share of freight by rail than North America, despite higher population density that would seem to favor rail. Why:

Fragmented infrastructure ownership. Different national operators, gauge and electrification differences, and legacy passenger-priority dispatching.
Shorter average hauls. EU lanes rarely stretch past 1,500 km; the intermodal breakpoint barely kicks in.
No double-stacking on most corridors. Halves the cost advantage.
Regulatory momentum. The EU’s 2021–2030 rail freight doubling target is real policy; rail freight corridors, ETCS signaling harmonization, and TEN-T investment aim at this. The effect in inbound planning remains modest so far.

Asian & intra-Asia rail

China Railway is a scale-in-itself operator. The notable inbound flows:

China-Europe Express (CR Express / Yuxinou / Zhengou etc.). Containerized rail service from inland China (Chongqing, Chengdu, Xi’an, Zhengzhou) to Europe (Duisburg, Hamburg, Łódź, Madrid). Transit: 16–22 days, ~2–3× ocean cost, geopolitically exposed (routing through Russia or detouring via Caspian/Middle Corridor post-2022).
Trans-Asian corridors. BRI-era network extensions into Central Asia and South Asia. Uneven execution; useful for specific origin-destination pairs.

Rail-specific operational hazards

Terminal dwell. Containers sit at origin and destination rail ramps longer than most shippers expect. 1–3 days on each end is common.
Car supply / flatcar shortage. Domestic intermodal capacity is constrained by well-car availability on some lanes during peak.
Bypass freight. Pre-blocked containers on a train may be offloaded and held for a later departure if the train is re-blocked mid-route. Low-frequency issue, high-surprise impact.
Weather-sensitive routings. Midwest winter storms, Sierra snow closures, PNW flooding. Lane-specific.
Labor. US Class I rail labor negotiations are national and generate shutdown risk. The 2022 near-strike cost shippers an estimated $2B in pre-positioning and mode-shift expense. Performance data is published by the US STB [STB Metrics].

Putting it together

Rail’s biggest problem in modern inbound is inertia. Planners who grew up running truck lanes often treat rail as an adjunct for bulk commodities and miss the intermodal mode migration opportunity on their own network. Meanwhile, sustainability reporting is quietly making rail’s carbon advantage visible to CFOs in a way it wasn’t five years ago.

How to think about rail on your own network

Five decisions worth revisiting:

When you have long-haul lanes (> 800 mi) on truck: compute current intermodal share and a 3-year target. Many shippers are 10–20 points under the optimum on qualifying lanes.

When you’re greenfielding a DC location: weigh proximity to a Class I ramp as a 10–15-year cost driver. Drayage distance from DC to nearest ramp dominates total cost on shorter intermodal lanes.

When rail service reliability degrades on a corridor: be ready to shift volume back to truck. Rail performance is cyclical; track STB performance metrics (US) and EU rail freight KPIs.

When you’re bidding RFPs: don’t treat intermodal and OTR as separate exercises. Strong carriers price both and offer meaningful discounts on blended commitments.

Where lower intermodal share is right: lanes with thin volume, time-sensitive commodities (pharma, perishables), DCs sited far from any Class I ramp, or industries with rail-incompatible packaging (oversize machinery, wide-load construction). Don’t force rail onto a network it doesn’t fit.