High Speed Trainset Market 2025-2035
The high-speed train market in 2025 is simultaneously the most mature and most disrupted segment of the global rolling stock industry.

- The global high-speed train market is projected to grow from $79.8 billion in 2025 to $135 billion by 2035 — a CAGR of 5.4% driven overwhelmingly by China’s network expansion and European fleet renewal: The two demand engines are structurally different. China’s engine is infrastructure-led: the People’s Republic plans to expand its high-speed rail network from 48,000 km at end-2024 to 60,000 km by 2030, requiring continuous procurement of CR400 and CR450 Fuxing series trainsets at a rate of approximately 150–200 train sets per year from CRRC’s Qingdao Sifang plant — the single largest volume of high-speed train production anywhere in the world. Europe’s engine is fleet renewal: the average age of TGV, ICE, Frecciarossa, and AVE fleets across France, Germany, Italy, and Spain is approaching the 30-year service life threshold simultaneously, triggering a wave of procurement orders (Alstom Avelia Horizon, Siemens Velaro MS, Hitachi Zefiro) that will dominate European rolling stock procurement through 2035. The two engines together explain why the high-speed train market is growing at 5.4% — fast, but not as fast as the hydrogen train market (28.2%), because this is fleet replacement rather than a new technology category being born.
- CRRC holds approximately 70% of the global high-speed trainset market by units — but is effectively excluded from Europe and North America by procurement restrictions: CRRC Qingdao Sifang’s market share in high-speed EMUs within China is nearly 50%, and within China’s total deliveries it holds over 70% of the high-end passenger coach market. But CRRC’s export position in high-speed trains is almost zero: the European Commission launched an anti-subsidy investigation in February 2024, the United States classified CRRC as a “Chinese military company” in October 2022 effectively barring it from federal-funded procurement, and no major European operator has placed a high-speed train order with CRRC. The global market statistics that show CRRC’s dominance therefore reflect China’s enormous domestic market in isolation. In the competitive market — where operators choose freely between suppliers — Siemens, Alstom, Hitachi, CAF, and Talgo compete for a market from which the world’s largest manufacturer is structurally excluded.
- The aerodynamic drag reduction between 300 km/h and 400 km/h requires a train to overcome 3.6× more air resistance — making the last 100 km/h of speed the most expensive increment to achieve: Aerodynamic drag force scales with the square of velocity: F_drag = ½ × ρ × v² × C_d × A. For a high-speed train at 300 km/h (83.3 m/s) versus 400 km/h (111.1 m/s): the drag force ratio is (111.1/83.3)² = 1.78², and the power required to overcome drag scales with v³, giving a power ratio of (111.1/83.3)³ = 2.37× more power for 33% more speed. But the structural challenges go beyond power: at 400 km/h, the pantograph-catenary interaction requires entirely redesigned current collection to prevent contact wire uplift exceeding EN 50367 limits; noise (which scales with v⁶ for aerodynamic sources) requires active noise cancellation systems in tunnels; and the stopping distance from 400 km/h to zero at 1.0 m/s² deceleration is 2,469 m — nearly 2.5 km — requiring ETCS Level 2 or equivalent continuous cab signalling with no fixed lineside signals in the braking corridor. The CR450’s engineering achievement is not merely “going faster” — it is building a complete operational system that keeps all of these parameters within safety and regulatory limits at 400 km/h commercial speed.
- Alstom’s La Rochelle factory is doubling its high-speed body production capacity in 2025–2026 — the most significant European rail manufacturing investment in a decade: Alstom’s April 2025 announcement of a €150 million investment programme across its French sites is a direct response to the Avelia Horizon order book: SNCF Voyageurs’ initial 100-train order (2018), plus 15 additional sets (August 2022), plus 15 more (January 2026), plus Proxima/Velvet’s 12 trains (October 2024), plus ONCF Morocco’s 18 trains (March 2025), plus Eurostar Celestia’s 30 trains (October 2025) — a total pipeline of approximately 190 Avelia Horizon trainsets representing over €6 billion in manufacturing value. The La Rochelle bodyshell and assembly capacity is being doubled; a new 250-metre building at Belfort can accommodate a complete Avelia Horizon trainset for commissioning; component manufacturing at Petit-Quevilly, Ornans, Tarbes, and Le Creusot is being upgraded with robotics and advanced welding. At least 1,000 new employees are being hired in France in 2025 alone.
- The first high-speed train compliant with Federal Railroad Administration (FRA) standards entered service in the United States in August 2025 — after a three-year delay: Alstom’s Avelia Liberty (marketed by Amtrak as the NextGen Acela), designed to operate on the Northeast Corridor between Boston and Washington DC, entered revenue service on 28 August 2025 — three years after the originally planned 2022 entry into service. The delays were caused by the unique complexity of the US Northeast Corridor: the Avelia Liberty shares tracks with commuter rail, freight trains, and Amtrak regional services, requiring FRA Tier III crashworthiness standards far more demanding than European TSI LOC&PAS requirements, and the track infrastructure dates back in some sections to the 1830s, producing wheel-track interaction profiles incompatible with the train’s initial suspension calibration. With 28 trainsets ordered at a contract value of approximately $2.4 billion, the Avelia Liberty is the most expensive high-speed train procurement per trainset in history — reflecting the combination of bespoke certification costs, US domestic assembly requirements (at Hornell and Rochester, New York), and the limited economies of scale from a 28-unit order.
When two CR450 prototype trains passed each other at high speed on the Meizhou Bay cross-sea bridge in Fujian Province on a June morning in 2023, the combined closing speed of the two trainsets registered 891 km/h on the monitoring equipment — faster than the cruising speed of a commercial jet aircraft, and a number that acquired an almost surreal quality when reported in the international railway press. The world record for the fastest closing speed between two trains passing each other had been broken. Each individual train was travelling at approximately 453 km/h — itself a record for a conventional wheel-on-rail train, surpassing the previous high-speed record set by a modified TGV Duplex on the LGV Est in France on 3 April 2007, when a specially lightened train with outsized power equipment reached 574.8 km/h in a single record run. The 2007 TGV record was a demonstration of absolute speed capability using a purpose-built test configuration that bore little relationship to any production train. The CR450’s 453 km/h was achieved by a prototype of a train designed to carry passengers in commercial service at 400 km/h — not a one-off modified vehicle, but the first production-representative unit of the next generation of Fuxing trains. The difference between the two records encapsulates the change in the global high-speed train market over the intervening 18 years: in 2007, pushing the boundary of what was physically possible was a European exercise, executed on European infrastructure, by a European manufacturer. In 2023, the most significant high-speed rail technological achievement anywhere in the world was Chinese — engineered in Qingdao, tested on Chinese high-speed lines, destined for the world’s largest high-speed rail network. Understanding the high-speed trainset market in 2025 requires understanding both of these facts simultaneously: Europe still leads in accumulated operational experience, safety certification sophistication, and the ability to sell commercially viable high-speed trains to the global open market. China leads in production scale, speed of development, and the sheer volume of high-speed rail that its network is adding and will add through 2035. The $135 billion market that emerges from these two dynamics by 2035 will be unlike any previous rolling stock market in its combination of established technology, next-generation ambition, and geopolitical complexity.
What Is a High-Speed Trainset — and Where Does the 250 km/h Threshold Come From?
A high-speed train (HST) or high-speed trainset is a railway passenger vehicle designed and certified for operation at speeds exceeding 250 km/h on purpose-built high-speed rail infrastructure, or above 200 km/h on upgraded conventional rail. The 250 km/h threshold is defined in European Union Directive 96/48/EC (the Technical Specification for Interoperability for High-Speed Rail — now consolidated within EU Regulation 1302/2014 as TSI LOC&PAS and TSI INF for high-speed infrastructure), which sets the minimum operational speed of a high-speed train in Class 1 service (new dedicated high-speed lines) at 250 km/h. In Japan, the Shinkansen Act (1964) defines Shinkansen as a railway with a maximum speed of at least 200 km/h, with all commercial Shinkansen services currently operating between 260 km/h (Hokkaido, due to tunnel clearance constraints) and 320 km/h on standard Shinkansen lines. In China, the regulatory standard is set by China Railway (CR), which classifies EMUs for high-speed service in two tiers: 350 km/h operating speed (Class C350) and 250 km/h operating speed (Class C250).
The market encompasses three distinct product categories: very high-speed trains (operating above 300 km/h — TGV, ICE, Shinkansen, CRH/Fuxing, Frecciarossa); standard high-speed trains (200–300 km/h — used on upgraded conventional lines, including many Spanish AVE services and regional high-speed services); and next-generation ultra-high-speed trains (above 350 km/h commercial speed — currently represented only by the CR450, with European programmes at design stage). Maglev trains (the SCMaglev L0 Series tested at 603 km/h in 2015, the Chinese CRRC 600 km/h prototype unveiled July 2025) are a related but distinct product category relying on magnetic levitation rather than wheel-rail contact — addressed separately in the Maglev Train Market article.
The Physics of High Speed: Why Each 50 km/h Increment Costs Exponentially More
Drag force: F_drag = ½ × ρ × v² × C_d × A
Representative high-speed train parameters:
Air density (ρ): 1.225 kg/m³ (sea level, 15°C)
Drag coefficient (C_d): 0.20 (modern HST with nose fairing, closed bogies)
Frontal area (A): 10.0 m² (standard 3.0 m wide, 3.7 m tall HST cross-section)
Drag force at key speeds:
200 km/h (55.6 m/s): F = ½ × 1.225 × 55.6² × 0.20 × 10 = 37,856 N = 37.9 kN
250 km/h (69.4 m/s): F = ½ × 1.225 × 69.4² × 0.20 × 10 = 59,024 N = 59.0 kN
300 km/h (83.3 m/s): F = ½ × 1.225 × 83.3² × 0.20 × 10 = 84,980 N = 85.0 kN
350 km/h (97.2 m/s): F = ½ × 1.225 × 97.2² × 0.20 × 10 = 115,644 N = 115.6 kN
400 km/h (111.1 m/s): F = ½ × 1.225 × 111.1² × 0.20 × 10 = 151,159 N = 151.2 kN
Power required to overcome drag (P = F × v):
200 km/h: 37,856 × 55.6 = 2.1 MW
300 km/h: 84,980 × 83.3 = 7.1 MW
350 km/h: 115,644 × 97.2 = 11.2 MW
400 km/h: 151,159 × 111.1 = 16.8 MW
Power scaling ratio (relative to 200 km/h baseline):
300 km/h: 7.1 / 2.1 = 3.4× more power
350 km/h: 11.2 / 2.1 = 5.3× more power
400 km/h: 16.8 / 2.1 = 8.0× more power
(Note: drag power ∝ v³ — doubling speed requires 8× the power just for drag)
Stopping distance from maximum speed at a = 1.0 m/s²:
300 km/h: d = v²/(2a) = 83.3²/(2×1.0) = 3,469 m = 3.47 km
350 km/h: d = 97.2²/(2×1.0) = 4,724 m = 4.72 km
400 km/h: d = 111.1²/(2×1.0) = 6,171 m = 6.17 km
→ At 400 km/h, a train needs 6.17 km to stop at 1.0 m/s² deceleration
→ This requires continuous cab signalling (ETCS L2/L3) on the full braking corridor;
no lineside signals are visible at approach speeds this high.
The Tunnel Pressure Wave Problem
When a high-speed train enters a tunnel at speed, it generates a compression wave that travels through the tunnel at the speed of sound (approximately 340 m/s) and, when it reaches the far end, causes a pressure impulse — the “sonic boom” or “tunnel boom effect” — that can be heard and felt at distances of up to 2 km from the tunnel exit. The intensity of the pressure wave scales approximately with v² × (blocked area / tunnel cross-section), meaning that at 350 km/h the pressure wave is approximately 36% more intense than at 300 km/h, and at 400 km/h it is 78% more intense than at 300 km/h. Japanese Shinkansen tunnel design for the planned 360 km/h upgrade (for the E10 Series introduction) requires “hood” extensions at tunnel portals — gradually widening entrance structures that slow the compression wave build-up rate — and specifically shaped train noses (the E5 Series’ 15-metre long “aero stream” nose, the N700S’s “aerodynamic wing nose”) designed to reduce the initial pressure disturbance. The CR450’s aerodynamic development involved over 1,000 computational fluid dynamics iterations to achieve a nose profile that passes through tunnels at 400 km/h without exceeding the pressure differential limits specified in UIC 779-11 (applied by reference in Chinese standards).
Market Size 2025–2035: A $135 Billion Decade
| Metric | 2024 | 2025 | 2030 (proj.) | 2035 (proj.) |
|---|---|---|---|---|
| Global HST market size | $75.7 billion | $79.8 billion | ~$104 billion | $135 billion |
| CAGR (2025–2035) | 5.4% per year | |||
| Asia-Pacific share (China + Japan) | ~55% | ~56% | ~58% | ~60% — growing fastest |
| Europe share | ~34% (largest at $34B) | ~33% | ~30% | ~28% (mature, stable) |
| North America share | ~7% | ~8% | ~12% | ~17% (Brightline, California HSR, NEC upgrades) |
| Middle East / Africa | <3% | <3% | ~4% | ~5% (Egypt Velaro, Saudi, Morocco) |
| Dominant speed segment | 300–400 km/h: 57% market share — optimal balance of speed and economics | |||
| Leading application | Passenger train: ~73.8% share; high-speed freight: emerging | |||
Three Structural Growth Drivers
The first driver is China’s network expansion mandate. China’s 14th Five-Year Plan (2021–2025) and the announced 15th Five-Year Plan (2026–2030) target a national high-speed rail network of 60,000 km by 2030, up from 48,000 km at the end of 2024. Each 1,000 km of new high-speed line requires approximately 20–40 trainsets depending on service frequency — implying procurement of approximately 240–480 trainsets for the 12,000 km to be added. This China-internal procurement alone accounts for approximately 40% of the global market’s growth over the decade.
The second driver is European fleet renewal. The first-generation TGV Duplex fleet (France), the ICE 1 and ICE 2 (Germany), the Eurostar Class 373, and the AVE Class 100 (Spain) were all introduced between 1981 and 2001. A 30–35 year service life means these fleets are at or beyond economic life simultaneously — creating a concentrated replacement demand in 2025–2035 that has no equivalent since the original high-speed rail build-out of the 1990s.
The third driver is market expansion into new geographies. Egypt’s 2,000 km high-speed rail network (the first in Africa), Saudi Arabia’s expansion of the Haramain line, Morocco’s Al-Boraq extension to Marrakech, India’s Mumbai–Ahmedabad HSR corridor (using Japanese Shinkansen N700S technology), and North America’s Brightline West (Las Vegas–Los Angeles) and California High-Speed Rail projects represent an entirely new category of high-speed rail procurement in markets that did not exist a decade ago. These projects collectively represent approximately $20–30 billion in rolling stock procurement over 2025–2035.
The World’s Major High-Speed Trainset Models: Technical Specifications
| Model | Manufacturer | Max Speed | Configuration | Capacity | Traction Power | Service |
|---|---|---|---|---|---|---|
| CR400AF/BF (Fuxing) | CRRC Qingdao Sifang / Changchun | 420 km/h (test); 350 km/h commercial | 8-car (CR400AF) or 8-car (CR400BF); 16-car or 17-car versions | 1,193 (16-car) to 1,283 (17-car) | 10,400 kW (8-car) | China: Beijing–Shanghai, Beijing–Guangzhou; export: Jakarta–Bandung (KCIC400AF) |
| CR450AF (Fuxing next-gen) | CRRC Qingdao Sifang | 450 km/h (test); 400 km/h commercial (from 2026–27) | 8-car; aluminium + carbon fibre; PMSM motors | ~540 (8-car) | >10,400 kW (PMSM, 3% more efficient) | Prototype rolled out November 2024; 453 km/h record June 2023; commercial service 2026–27 on Chengdu–Chongqing HSR |
| Siemens Velaro MS (ICE 3neo / Class 408) | Siemens Mobility (Krefeld, Germany) | 320 km/h (AC); 200 km/h (DC) | 8-car; 200 m; aluminium body; multi-system 15kV/25kV/3kV/1.5kV DC | 439 seats | 8,000 kW | DB Germany from December 2022; international to Belgium and Netherlands from June 2024; 90 units ordered (total €3.1B) |
| Siemens Velaro EGY | Siemens Mobility (Krefeld) | 230 km/h (desert-rated for Egypt) | 8-car; reinforced against sand and heat; air filtration for desert conditions | ~600 | 8,000 kW | 41 trainsets for Egyptian National Railways (2,000 km turnkey network); deliveries from 2024; unveiled at InnoTrans September 2024 |
| Siemens American Pioneer 220 (Brightline West) | Siemens Mobility (new Horseheads, NY factory) | 354 km/h (220 mph — first US 200+ mph train) | 7-car; FRA Tier III crashworthiness; US domestic assembly | ~450 | TBD (based on Velaro Novo platform) | 10 trainsets for Brightline West Las Vegas–Rancho Cucamonga; production at new Horseheads, NY facility; planned service 2029 |
| Alstom Avelia Horizon (TGV INOUI / Eurostar Celestia) | Alstom (La Rochelle bodies; Belfort power cars; Valenciennes assembly from 2026) | 350 km/h (max); 300+ km/h commercial | Push-pull; 2 power cars + 7–9 bi-level coaches; articulated Jacobs bogies; quad-current variants (1.5kV/3kV DC, 15kV/25kV AC) | 740 (9-car standard); 540 (Eurostar Celestia 200m) | 9,280 kW (2 × 4,640 kW power cars) | SNCF Voyageurs (130 sets ordered; TGV INOUI service from 2026); Eurostar Celestia (30 sets, €2B, service 2031); Proxima Velvet (12 sets, €850M, 2028); Morocco ONCF (18 sets, March 2025). Total pipeline: ~190 sets |
| Alstom Avelia Liberty (NextGen Acela) | Alstom (Hornell + Rochester, NY — US assembly; parts from 29 states) | 300 km/h (186 mph) capable; operational speed limited to ~177 km/h on NEC | 9-car articulated; tilting; FRA Tier III; 378 seats + 8 wheelchair (386 total per set) | 386 passengers; 25% more than Acela Express it replaces | 11,000 kW | 28 sets for Amtrak ($2.4B contract); revenue service from 28 August 2025 (3 years delayed from original 2022 target) |
| JR Central N700S “Supreme” Shinkansen | Nippon Sharyo + Hitachi (SiC components); Kawasaki excluded (technology transfer dispute) | 300 km/h Tokaido; 285 km/h Sanyo curves | 16-car (J/H sets) or 6-car (Y sets/Kyushu); aluminium hollow extrusion; SiC-MOSFET VFD (first in Shinkansen) | 1,323 (16-car standard); LTO battery self-propulsion system for earthquake evacuation | 17,080 kW (16-car) | JR Central from July 2020; JR West from March 2021; ~50 sets in service by early 2025; price ~¥6 billion (~$40M) per 16-car set; target: Texas Central Railway advisory role |
| JR East E5 / H5 Series Shinkansen | Hitachi (Kasado works, Yamaguchi) + Kawasaki Heavy Industries (Hyogo) | 320 km/h (Tohoku); 260 km/h (Hokkaido, H5) | 10-car; 1.5° active tilt; 4,000 m minimum curve radius at 320 km/h; aluminium alloy | 731 (658 standard + 55 Green + 18 Gran Class) | 9,960 kW | JR East from March 2011 (63 sets total E5+H5); successor E10 Series announced March 2025, test runs 2027, service by 2030 to replace E2 and E5 |
| Hitachi / Trenitalia ETR 1000 (Frecciarossa 1000) | Hitachi Rail (Pistoia, Italy) + AnsaldoBreda (now Hitachi) | 400 km/h max design; 300 km/h commercial Italy; up to 360 km/h certified | 8-car articulated; distributed traction; multi-voltage (3kV DC, 25kV AC, 15kV AC, 1.5kV DC) | 457 (standard; 4 classes) | 9,800 kW | Trenitalia Italy from 2015; licensed for UK service 2023 (Evolyn); Spain (Ouigo); 50 sets + options; Hitachi Pistoia factory key production hub |
| CAF Oaris / Talgo 350 (AVRIL) | CAF (Beasain, Spain) / Talgo (Las Matas, Spain) | Oaris: 350+ km/h; Talgo 350: 330 km/h; AVRIL: 380 km/h max | CAF Oaris: 8-car EMU; Talgo 350: articulated passive tilting; AVRIL: variable gauge, passive tilt | Oaris: ~600; Talgo 350: 318; AVRIL: up to 581 | Oaris: 8,000 kW; AVRIL: 12,000 kW | Talgo 350 RENFE Spain; Talgo 250 (Ouigo España, open access); CAF Oaris ordered by RENFE 2021; AVRIL (Alvia high-speed) Spain; Talgo pursuing open access UK, Europe |
Manufacturing Facilities: Where High-Speed Trains Are Built
| Manufacturer / Site | Location | Area / Scale | HST Models Produced | Annual Capacity / Output |
|---|---|---|---|---|
| CRRC Qingdao Sifang | Jihongtan, Chengyang District, Qingdao, Shandong, China | Large-scale national HSR production base; ~50% market share in Chinese HSR EMUs; 6 national-level R&D platforms | CR400AF/AF-A/AF-AS (Fuxing); CR450AF (next-gen); KCIC400AF (Indonesia); metro and intercity EMUs; 600 km/h maglev prototype (2021) | ~150–200 HSR trainsets/year for domestic Chinese market (dominant share); ~70%+ of China’s high-end passenger coaches; export position constrained by EU anti-subsidy probe (2024) and US national security classification (2022) |
| CRRC Changchun Railway Vehicles | Changchun, Jilin, China | Joint developer of CR400BF (with CRRC Qingdao Sifang); primarily metro and regional EMU production | CR400BF/BF-B (Fuxing); CR450BF (next-gen, co-development); metro EMUs | CR400BF production alongside CRRC Qingdao Sifang for domestic network expansion |
| Siemens Mobility — Krefeld-Uerdingen | Krefeld, North Rhine-Westphalia, Germany (founded 1898 as Waggon-Fabrik AG Uerdingen; 125th anniversary 2023) | Production area: 74,000 m²; logistics area: 64,200 m²; “largest train order in company history” — Egypt 2022 (41 Velaro + 94 Desiro HC) | Velaro MS (ICE 3neo / Class 408); Velaro EGY; ICE 4; Mireo (regional); Desiro HC; Mireo Plus B/H (hybrid) | 90 ICE 3neo for DB (€3.1B total; deliveries to 2029); 41 Velaro EGY for Egypt; 230 German supplier companies involved in ICE 3neo alone; ~thousands of rail vehicles per year across all types |
| Siemens Mobility — Horseheads, New York (new) | Horseheads, Chemung County, New York, USA (new facility, groundbreaking 2024) | New dedicated facility for North American HST market; created specifically for American Pioneer 220 (Brightline West) production | American Pioneer 220 (Brightline West; 7-car, 354 km/h, FRA standards) | 10 trainsets for Brightline West (2024 order); facility expandable for future US HST orders; planned delivery 2027–2028 for 2029 Las Vegas service launch |
| Alstom — La Rochelle (Aytré) | La Rochelle (Aytré), Charente-Maritime, France | Historic TGV bodyshell and assembly plant; capacity being doubled as part of €150M investment programme (2025–2026); “TrainLab” digital test facility | Avelia Horizon (TGV INOUI / TGV M) bodyshell and passenger car assembly; all TGV Duplex-family derivatives | Pipeline of ~190 Avelia Horizon trainsets (SNCF 130, Eurostar 30, Proxima 12, ONCF 18); La Rochelle doubling assembly lines; first TGV INOUI service expected 2026 |
| Alstom — Belfort | Belfort, Territoire de Belfort, France | Historical centre for TGV power car production; new 250-metre commissioning building under construction (part of €150M investment); can accommodate a complete Avelia Horizon trainset | Avelia Horizon power cars; traction systems; historical TGV power car production since the original TGV PSE (1981) | Power car production capacity increased to match doubled La Rochelle body output; target: Avelia Horizon deliveries beginning 2026 for SNCF |
| Alstom — Valenciennes (Petite-Forêt) — NEW | Valenciennes, Nord, France | New Avelia assembly line (20% of the €150M investment — “first” for this northern site); robotisation and lean manufacturing techniques | Avelia Horizon assembly (new capacity addition); existing regional Citadis/Regiolis production continues | New Avelia line supplements La Rochelle; estimated additional capacity of 6–10 trainsets/year when operational |
| Alstom — Hornell + Rochester, New York | Hornell (Steuben County) and Rochester (Monroe County), New York, USA | US assembly facility for FRA-compliant rolling stock; Hornell: car body manufacturing; Rochester: components; over 180 US supplier companies in 29 states | Avelia Liberty (NextGen Acela); initial Avelia Liberty car body manufacturing began October 2017 | 28 Avelia Liberty trainsets ($2.4B Amtrak contract); service began 28 August 2025 (3 years delayed); highest per-trainset cost of any HST programme globally (~$86M per trainset) |
| Nippon Sharyo — Toyokawa | Toyokawa, Aichi, Japan | Primary Shinkansen car manufacturer; partner in N700S development (sole builder alongside Hitachi, Kawasaki excluded from N700S) | N700S (all generations: J/H/Y sets); earlier N700/N700A production; Tokaido/Sanyo Shinkansen focus | N700S production rate: JR Central ordering 19 additional sets (announced May 2022, ¥114B / $897M); 7 sets per year 2024–2025; cost ~¥6B (~$40M) per 16-car set |
| Hitachi Rail — Kasado Works | Kudamatsu, Yamaguchi, Japan | Primary Japanese Shinkansen builder (Hitachi); also produces for UK, Italian, and US rail markets | E5 Series (cars 1–5 per set); N700 Series (head/tail cars); AT-300 (UK Class 800/802); Frecciarossa ETR 1000 (with AnsaldoBreda); ZEFIRO express (UK) | JR East E10 Series (successor to E5) in development 2025; test runs planned 2027; Hitachi Pistoia (Italy) produces ETR 1000 Frecciarossa 1000 for Trenitalia and export |
| Kawasaki Railcar Manufacturing — Kobe (Hyogo) | Kobe, Hyogo, Japan | Shinkansen manufacturing since the original 0 Series (1964); excluded from N700S development by JR Central (technology transfer dispute); continues E5 and other JR East/JR West production | E5 Series (cars 6–10 per set); H5 Series; earlier N700, 700, 500 Shinkansen; Acela Express (joint venture with Bombardier for Amtrak, 1998–2001); US transit cars (Springfield, MA) | JR East E10 Series procurement: Kawasaki’s participation status under JR East evaluation; exclusion from JR Central N700S limits Tokaido/Sanyo volume significantly |
The Top 10 Market Players: Strategic Positions and Competitive Advantages
| # | Company | Key Platforms | Manufacturing Base | Strategic Position |
|---|---|---|---|---|
| 1 | CRRC (China) | CR400AF/BF (Fuxing); CR450AF/BF (400 km/h, 2026+); CRRC 600 km/h maglev (unveiled July 2025) | Qingdao Sifang; Changchun; Tangshan; 183,000+ employees | World’s largest rolling stock manufacturer by revenue; dominant in China’s 48,000 km network; CR450 entering commercial service 2026–27 will make it operator of world’s fastest wheel-rail commercial train; effectively excluded from EU and US by anti-subsidy investigations and national security classifications; export position limited to non-Western markets (Indonesia KCIC400AF, Middle East, Latin America) |
| 2 | Siemens Mobility (Germany) | Velaro MS (ICE 3neo); Velaro EGY; American Pioneer 220 (Brightline West); ICE 4; Velaro Novo (next-gen, R&D) | Krefeld (primary, 74,000 m²); Erlangen (electronics); new Horseheads NY factory | European HST market leader; largest single order in history — Egypt 2022 (41 Velaro EGY + 94 Desiro HC); first US market HST entry through Brightline West; ICE 3neo development delivered in record 12 months; Digital Twin integration for predictive maintenance; AI-driven maintenance partnership announced September 2025 |
| 3 | Alstom (France) | Avelia Horizon (TGV INOUI, Eurostar Celestia, Proxima Velvet, ONCF Morocco); Avelia Liberty (Amtrak NextGen Acela); Avelia Stream (next international) | La Rochelle (bodies, doubling capacity); Belfort (power cars, new 250m building); Valenciennes (new assembly line); Hornell + Rochester NY (US) | ~190 Avelia Horizon trainsets in order pipeline (>€6B manufacturing value); La Rochelle capacity doubling and €150M investment (April 2025); 40+ years of TGV commercial service experience; Eurostar Celestia contract (€2B, Oct 2025); strongest French-market position; 84,700 employees in 64 countries |
| 4 | Hitachi Rail (Japan / Italy / UK) | ETR 1000 (Frecciarossa 1000); AT-300 (Class 800/802 UK); E5/H5 Shinkansen; Zefiro 300/380 (intercity HS); N700S components | Kasado Works, Yamaguchi (Shinkansen); Pistoia, Italy (ETR 1000); Newton Aycliffe, UK (AT-300) | Unique tri-continent manufacturing presence (Japan/Europe/UK); Frecciarossa 1000 certified to 400 km/h (highest speed certification of any commercial HST); AT-300 bi-mode/tri-mode platform for UK/Italian mixed infrastructure; JR East E10 Series development in progress (to replace E5 by 2030) |
| 5 | JR Central / Japan (N700S export ambition) | N700S Supreme (Tokaido/Sanyo); technical advisor to Texas Central Railway, India Mumbai–Ahmedabad HSR | Nippon Sharyo (Toyokawa); Hitachi (Kasado); N700S: ~$40M/16-car set | N700S is Japan’s primary HST export vehicle; Mumbai–Ahmedabad HSR (508 km) adopting N700S technology with local production (Japan financing ¥5 trillion); Texas Central Railway (consultancy); world record safety: zero passenger fatalities in 60+ years of Shinkansen; LTO battery self-propulsion system for earthquake safety |
| 6 | Talgo (Spain) | Talgo 350; AVRIL (Alvia); Talgo 230 (Poland); variable gauge technology; Talgo 106 (RENFE) | Las Matas (Madrid); Rivabellosa (Álava); Talgo US (Nevada — proposed) | Unique passive variable-gauge technology (automatic wheelset adjustment from 1,668 mm Iberian to 1,435 mm standard gauge) enabling direct through-running France–Spain without gauge change stop; AVRIL design speed 380 km/h; strategic US market ambitions (Brightline, Amtrak corridor candidates); strong position in open-access European market (Ouigo España) |
| 7 | CAF (Spain) | Oaris (350+ km/h); CIVITY HS; interoperable high-speed for multiple European markets | Beasain, Gipuzkoa (Basque Country, HQ and primary factory); Zaragoza; Solares | Oaris selected by RENFE for Spanish domestic services (2021); multi-voltage, multi-gauge interoperability for pan-European operations; competitive pricing against Big Three; growing presence in HS market as third major European HST OEM; strong in Latin American and Middle Eastern regional HS markets |
| 8 | Kawasaki Heavy Industries (Japan) | E5/H5 Shinkansen (co-manufacturer); earlier 500 Series, 700 Series, N700 Series; US transit cars (Springfield, MA) | Kobe, Hyogo (primary rail); Springfield, Massachusetts (US assembly) | Excluded from N700S by JR Central (technology transfer to China dispute); maintains significant JR East production share (E5); Springfield MA factory positioned for US transit market; E10 Series selection under JR East evaluation as potential co-manufacturer; US Acela Express original manufacturer (1998–2001 joint venture with Bombardier) |
| 9 | Hyundai Rotem (South Korea) | KTX-III; EMU-250 (Hanwha high-speed); EMU-320 (next-gen, 320 km/h); HEMU-430X prototype (430 km/h) | Changwon, South Gyeongsang Province (primary HST factory); Incheon | Domestic Korean KTX market supplier; HEMU-430X prototype tested at 421 km/h (2013 Korean record); growing export ambitions in Middle East (Saudi Arabia) and Southeast Asia; competitive pricing; benefit from Korean government’s high-speed rail export promotion programme; KTX-III planned for Korean routes exceeding current 300 km/h KTX speeds |
| 10 | Nippon Sharyo (Japan) | N700S (primary Tokaido Shinkansen manufacturer alongside Hitachi post-Kawasaki exclusion); long history of Shinkansen production from 0 Series | Toyokawa, Aichi, Japan (primary); owned by JR Central (majority stake) — making it the only major rolling stock manufacturer that is also majority-owned by its primary customer | JR Central’s ownership creates a vertically integrated supply chain for Tokaido Shinkansen production; N700S annual delivery rate of 7 sets per year (2024–2025) is consistent and stable; India Mumbai–Ahmedabad project will require localisation — Nippon Sharyo positioned as technology transfer partner for Indian domestic manufacturing under the Make in India programme |
European HST Market: TGV vs ICE vs Shinkansen Technology Comparison
| Parameter | Alstom Avelia Horizon (TGV) | Siemens Velaro MS (ICE 3neo) | Hitachi ETR 1000 (Frecciarossa 1000) | JR Central N700S (Shinkansen) |
|---|---|---|---|---|
| Architecture | Push-pull: 2 power cars + articulated coaches | Distributed power: all axles powered (motorised bogies throughout) | Distributed power: all cars self-powered | Distributed power: SiC-MOSFET VFD on all cars |
| Maximum commercial speed | 350 km/h (300+ in service) | 320 km/h | 360 km/h certified (300 km/h in service) | 300 km/h (Tokaido); 320 km/h (Sanyo) |
| Deck configuration | Double-deck (bi-level coaches) | Single-deck | Single-deck | Single-deck |
| Capacity (standard configuration) | 740 passengers (9-car double-deck) | 439 (8-car single) | 457 | 1,323 (16-car) |
| Total traction power | 9,280 kW (2 power cars) | 8,000 kW (distributed) | 9,800 kW (distributed) | 17,080 kW (16-car distributed) |
| Tilt capability | No (passive articulation) | No | No | Yes: 1° active tilt (allows 285 km/h on 2,500m radius curves) |
| Regenerative braking energy | 20% energy consumption reduction vs predecessors | 10% energy reduction vs ICE 3 | Significant regen; 4 voltage systems | 7% energy reduction vs N700A (SiC advantage) |
| Carbon footprint reduction vs predecessor | 37% lower carbon footprint | 14 g CO₂/passenger-km | Significant vs air travel | 7% lower vs N700A |
| First service | 2026 (SNCF TGV INOUI) | December 2022 (DB ICE 3neo) | 2015 (Trenitalia) | July 2020 (JR Central) |
| Unit price (approx.) | €70M per trainset (from ONCF/Proxima orders) | ~€34M per trainset (90 × €3.1B) | ~€30–40M per trainset | ~¥6B = ~$40M per 16-car set |
The high-speed train market in 2025 is simultaneously the most mature and most disrupted segment of the global rolling stock industry. It is mature in that the technical parameters of what a high-speed train must achieve — 300 km/h operating speed, distributed or power-car traction, aerodynamic profiling, energy regeneration — have been commercially proven for over 40 years, from the TGV PSE’s inaugural Paris–Lyon service in September 1981 to the CR400’s 350 km/h Beijing–Shanghai service today. It is disrupted in that the geopolitical architecture of the market has changed more fundamentally in the past five years than in the previous 40: CRRC — the manufacturer of approximately 70% of the world’s high-speed trainsets by volume — is now effectively absent from the competitive market for political rather than technical reasons, leaving the European and North American markets to a competition between Siemens, Alstom, Hitachi, and smaller specialists that CRRC’s pricing and scale could otherwise have reshaped. The two facts interact in a way that creates significant distortion: the competitive market’s prices (€34M for a Velaro MS, €70M+ per Avelia Horizon, $86M for an Avelia Liberty) are set without competitive pressure from the world’s highest-volume HST manufacturer, and they reflect that absence. European and North American rail operators are paying a substantial “geopolitical premium” for trains that would be cheaper if CRRC could bid. Whether that premium is worth the security and supply chain resilience benefits of procuring from non-Chinese manufacturers is a political question, not an engineering one. What is clear is that Siemens’s new Horseheads factory and Alstom’s La Rochelle capacity doubling are investments predicated on a CRRC-excluded market remaining excluded — a strategic bet that the geopolitical conditions of 2024–2025 will persist through the 2030 delivery window of these investments. That bet may prove correct. But it carries significant policy risk if relations between China and Western economies shift in ways that change the procurement calculus before the 190-trainset Avelia Horizon pipeline is fully delivered.
— Railway News Editorial
Frequently Asked Questions
1. What makes the CR450 technically different from the CR400 — and why does 400 km/h commercial service require so much more than simply increasing the motor power?
The CR450 differs from the CR400 in six engineering domains simultaneously, not just in traction power. First, the permanent magnet synchronous motors (PMSM) replace induction motors, providing approximately 3% higher energy conversion efficiency and a 20% reduction in motor weight — important because unsprung mass (mass below the suspension) must be minimised at 400 km/h to prevent excessive wheel-rail force. Second, carbon fibre and advanced aluminium composites replace conventional aluminium alloy throughout the body structure, achieving a 12% weight reduction versus the CR400. Third, the aerodynamic nose profile was developed through over 1,000 CFD (Computational Fluid Dynamics) iterations to manage tunnel compression waves and crosswind stability at 400 km/h — the nose shape is not the same as the CR400’s. Fourth, the pantograph-catenary system is redesigned: at 400 km/h, the contact wire is moving past the pantograph at 111 m/s, and the catenary system must maintain contact force within EN 50367 limits (10–300 N) throughout this interaction — requiring a new pantograph head design and a catenary tensioning system that keeps wave propagation speed in the wire above the train speed. Fifth, the braking system uses carbon-ceramic brake discs (a CR450 innovation) that maintain braking effectiveness at very high temperatures, combined with eddy current track brakes as the primary deceleration device above 300 km/h — the same principle used on French TGVs for high-speed emergency braking. Sixth, the stopping distance from 450 km/h (test speed) at 1.0 m/s² deceleration is 6,171 m — requiring dedicated signal-free braking corridors of over 6 km before any fixed obstacle. The CR450 is not a “faster CR400” — it is a comprehensively redesigned system.
2. Why did the Avelia Liberty take three years longer than planned to enter service — and is the US Northeast Corridor a unique challenge for high-speed train manufacturers?
The Avelia Liberty’s delay from 2022 to August 2025 reflects the unique complexity of the US Northeast Corridor (NEC), which is genuinely the most challenging operating environment for a high-speed train anywhere in the world — not because of speed requirements (the Avelia Liberty is physically capable of 300 km/h but operates at 150–177 km/h on the NEC due to track conditions) but because of the infrastructure’s age and complexity. The NEC was built between the 1830s and 1970s by at least seven different railroad companies with different standards, is shared by Amtrak long-distance trains, seven commuter rail operators, Amtrak Northeast Regional, and occasional freight movements, and has curves, switches, interlockings, and catenary systems of diverse vintages simultaneously in use. The Avelia Liberty had to be certified to FRA Tier III crashworthiness standards — designed for collision scenarios with freight trains and road vehicles at crossings, far more demanding than European TSI requirements — and then had to demonstrate safe operation over the specific wheel-rail interface profiles of the NEC, which include curves as tight as 1,200 m radius at locations where the track is nearly a century old. The specific reported issues included spontaneously shattering window glass (caused by stress concentrations from the NEC’s vertical alignment variations) and wheel-rail interaction anomalies requiring recalibration of the active suspension system. These are not failures of the train’s basic design — they are the predictable consequences of integrating a state-of-the-art high-speed train with aging infrastructure that was never designed for it. The NEC will need approximately $150 billion in infrastructure investment over 20 years (per Amtrak’s Gateway Program and NEC Commission assessments) to achieve speeds anywhere near the Avelia Liberty’s actual capability.
3. Why was Kawasaki excluded from the N700S Shinkansen programme — and what were the consequences for both JR Central and Kawasaki?
JR Central’s decision to exclude Kawasaki Heavy Industries from the N700S Shinkansen procurement was a direct consequence of the technology transfer dispute over the CRH2A series in China. Kawasaki had partnered with CNR Sifang to supply 60 CRH2A trainsets to China in 2004–2007, based on the E2 Series Shinkansen design. JR Central had explicitly requested that Kawasaki not transfer Shinkansen technology to China, citing concerns that the technology would be reverse-engineered for Chinese domestic production — a concern that proved prescient when China produced the CRH380A (based on adapted Shinkansen concepts) and filed patents on components derived from transferred technology. JR Central’s response was to remove Kawasaki from the N700S development consortium, retaining only Nippon Sharyo (majority-owned by JR Central itself) and Hitachi as manufacturing partners. The consequences were significant for both parties. For Kawasaki: the loss of Tokaido Shinkansen manufacturing contracts represents a substantial reduction in its highest-value rail manufacturing volume, partially compensated by continued JR East E5 Series production. Kawasaki has attempted to develop an independent position in the US transit market (Springfield, MA facility) but the cancellation of SEPTA’s bi-level car contract in April 2024 — citing delays and build quality issues — has left its North American position uncertain. For JR Central: the manufacturing duopoly of Nippon Sharyo and Hitachi creates a more concentrated supply chain with reduced competitive tension in procurement; prices for N700S sets (approximately ¥6 billion = $40 million per 16-car set) have not fallen materially despite accumulated learning curve, as would be expected if a third manufacturer competed.
4. What is the Brightline West project — and why does the Siemens American Pioneer 220 represent a potential turning point for high-speed rail in the United States?
Brightline West is a privately funded high-speed rail project between the Rancho Cucamonga transit hub (near Los Angeles) and Las Vegas, Nevada — approximately 320 km — targeting operations in time for the Los Angeles 2028 Olympic Games. In May 2024, Siemens Mobility was selected to supply 10 seven-car trainsets based on the Velaro Novo platform, branded “American Pioneer 220” — a name reflecting the 220 mph (354 km/h) designed maximum speed. The project represents a potential turning point for several reasons. It is the first privately financed, fully dedicated high-speed rail project in US history — the entire corridor is purpose-built for high-speed, not shared with commuter or freight traffic, eliminating the NEC-style compatibility challenges that delayed the Avelia Liberty. Its design speed (354 km/h) would make it the fastest wheel-rail train in regular commercial service outside China and Japan when launched. Siemens is building a new production facility at Horseheads, New York specifically for this programme — establishing US domestic manufacturing capability for HST production for the first time in modern history, potentially qualifying Siemens for future federally funded US HST procurement under Buy America provisions. If Brightline West opens on schedule for the 2028 LA Olympics — an ambitious target given the programme’s complexity — it will be the most visible demonstration of high-speed rail’s commercial viability in North America since Amtrak’s original Acela service in 2000, and will almost certainly trigger additional privately financed HST corridors in Texas (Dallas–Houston), the Pacific Northwest, and the Florida Atlantic coast.
5. What is the aerodynamic significance of Shinkansen nose design — and why do Japanese bullet trains have such long, dramatic nose shapes?
The elongated nose shapes of Japanese Shinkansen trains — from the E5 Series’ 15-metre “Aero Stream” nose to the N700S’s “Dual Supreme Wing” nose — are not aesthetic choices. They are engineering solutions to a specific problem unique to Japan’s Shinkansen network: the ratio of train cross-section to tunnel cross-section. Japanese Shinkansen tunnels were designed in the 1960s with a clearance gauge optimised for the train speeds of that era (210 km/h for the original 0 Series). As speeds increased to 270, 285, and then 320 km/h, the same tunnels produced progressively more severe micro-pressure waves at tunnel exits — the “sonic boom” effect. The Shinkansen has a particularly high “blockage ratio” (train cross-section / tunnel cross-section) of approximately 0.14–0.20, compared to European high-speed tunnels designed for modern trains at approximately 0.10–0.14. A higher blockage ratio means the air column compressed ahead of the train has less room to flow past the train in the tunnel annulus, producing a larger pressure wave. The solution Japanese engineers developed — an elongated, gradually tapering nose that compresses the air ahead of the train slowly rather than suddenly — reduces the initial pressure gradient of the compression wave by extending the compression phase over time. The E5 Series’ 15 m nose (compared to approximately 6–8 m for a comparable Velaro or TGV) was specifically optimised through testing at the RTRI (Railway Technical Research Institute) to reduce micro-pressure wave intensity by approximately 50% compared to the preceding E4 Series design. Each Shinkansen generation refines this geometry: the N700S’s “Dual Supreme Wing” nose generates not only a gentler compression wave but also slightly lower aerodynamic drag than its predecessors, contributing (alongside the SiC-MOSFET traction systems) to the 7% energy consumption reduction versus the N700A.






