Interface Standardisation in China's Urban Rail Transit — Lessons for Vietnam

To understand why China could dictate terms to the world's leading technology groups, one must first grasp the market size it commands. This is not a handful of lines — it is the largest urban rail market on the planet, and that scale is the source of every piece of leverage that followed.

▸ A near-vertical growth curve

Beijing's first subway opened in 1969, but the network did not genuinely explode until the early 2000s. The pace of expansion over the following fifteen years has almost no precedent in urban infrastructure history.

At end-2024 alone, a further 44 cities were actively constructing 5,833 km¹ — meaning China's pipeline alone is larger than the entire urban rail network of any other country. This continuous, large-scale repetition is the key enabling condition: a market large enough and sustained enough to nurture domestic manufacturing capacity and to negotiate with foreign suppliers from a position of strength.

▸ The central government's coordinating hand

A point critical for the sections that follow: although urban rail investment is funded locally, approval of metro construction plans rests with the National Development and Reform Commission (NDRC) at the central level. Beijing sets approval thresholds (on population, GRDP, urban density, local fiscal capacity) and therefore holds a coordinating position of real weight. It is this "gatekeeper" role — not the role of financier — that later became the instrument for driving industry-wide standardisation.

Twenty years ago, China stood exactly where Vietnam stands today: core signalling, rolling stock and control technology all came from abroad, products were dominated by a handful of large groups, and even training curricula drew directly on imported systems' manuals.⁷ The story of how China broke out of that position follows a clear logic — and this is the section Vietnam should read most carefully.

▸ "Market for technology"

The flagship policy was 市场换技术 — "market for technology": to access China's enormous market, multinationals had to transfer technology to domestic enterprises.⁸

The clearest example is high-speed rail. In 2004, the Ministry of Railways made technology transfer a condition of procurement. Four leading technology suppliers — Alstom (France), Siemens (Germany), Bombardier (Canada) and Kawasaki (Japan) — signed transfer agreements with two domestic manufacturing consortia (later merged into CRRC).⁸ China did not "steal" the technology; it purchased it with mandatory transfer terms, relying on the weight of contracts no supplier wanted to forgo.

▸ Nurturing national champions

In parallel with transfer, the state deliberately built "national champions" in each subsystem: CRRC for rolling stock, CRSC (China Railway Signal & Communication) for signalling, and companies rising from academia such as Traffic Control Technology (TCT) — spun out of Beijing Jiaotong University.

The symbolic milestone is 30 December 2010: Beijing's Yizhuang Line entered service with a communications-based train control (CBTC) system developed entirely by TCT, making China the fourth country worldwide to master and operate core CBTC technology, breaking the long-standing foreign monopoly.⁷ Within a single decade, China went from full import dependence to having domestic suppliers standing alongside Siemens, Alstom, Thales and Bombardier.

▸ Sequential localisation by subsystem

The methodological point worth noting: China localised sequentially by subsystem, prioritising those with the highest repetition volume and a manageable entry barrier (rolling stock, then signalling) before advancing to more complex integration challenges. They did not attempt to be self-sufficient in everything at once — a sequencing lesson Vietnam can apply with far more limited resources.

The entire roadmap was formalised as an official policy formula: 引进 – 消化 – 吸收 – 再创新 ("import – digest – absorb – re-innovate"). The emphasis lies not on importing but on the three subsequent steps: domestic firms must digest technical documentation, absorb design capability, and then re-innovate to produce their own products. The difference from ordinary "turnkey" procurement is that the state sets endogenous capability as the goal, treating equipment as the means rather than the end.

This is the core of the article, and the most transferable lesson for Vietnam. Technology localisation (Section 2) requires scale and time that Vietnam does not yet have — but interface standardisation requires neither, and it solves exactly the problems Vietnam faces: synchronisation, spare-parts compatibility, and vendor flexibility.

▸ The technology fragmentation problem

When each line is built by a different supplier using its own proprietary protocols, the result is a network of technology islands: one line's CBTC cannot communicate with the next; trains cannot run through across lines; equipment is not interchangeable; each line needs its own spare-parts inventory, its own maintenance team, its own training programme. This is precisely the vendor lock-in problem that inflates whole-life cost and eliminates competitive pressure.

▸ The China Association of Metros (CAMET)

To standardise urban rail development nationally, the China Association of Metros (CAMET) was established in 2011. CAMET brings together 252 experts across nine specialist committees (design, operations, construction, equipment…), is authorised by the National Bureau of Statistics to operate the industry statistical system, and publishes an annual report.⁴ More importantly, CAMET issues industry standards (T/CAMET series) — and it is this standards body that became the instrument for defining interface requirements.

▸ The Chongqing demonstration project: field evidence

In 2015, NDRC formally approved the "Chongqing Urban Rail CBTC Interoperability Demonstration Project" — China's first interoperable metro deployment.⁹ Its organisation is a template worth studying:

• Four signalling integrators (including TCT, UniTTEC and others) each built a separate line — Line 4, Line 5, Line 10 and the Ring Line — one integrator per line.⁹
• Three independent safety assessors and two LTE providers participated, forcing open interfaces and shared standards development.⁹
• Standards T/CAMET 04010–04013 defined interoperability architecture, interfaces, testing and technical deployment. Every signalling supplier was mandated to comply with this architecture and standard set.^10,11

The result: trains fitted with different suppliers' signalling equipment can share and cross between lines without stopping or degrading operating mode.^9,11 In other words, China did not force all lines to use one supplier — it forced all suppliers to comply with a common interface standard. Competition between suppliers was preserved while interoperability and asset sharing were guaranteed.

▸ Scaling the standard: from one project to national practice

What turned Chongqing from a pilot into an institution was the diffusion mechanism through tender documents. After Chongqing, in tender documents for urban rail projects in Qingdao, Changsha, Hohhot, Beijing and many other cities, suppliers were required to integrate interoperable data interfaces following exactly the Chongqing template.¹¹ The standard did not sit on paper — it was embedded directly into bid eligibility conditions, where it had binding force on every supplier that wished to participate.

▸ The M&E integration tier: open architecture instead of hard lock-in

The same logic applies to the Integrated Supervisory Control System (ISCS) tier — which aggregates multiple M&E subsystems into the Operations Control Centre (OCC): power SCADA (PSCADA), building automation (BAS), security (CCTV), fire alarm (FAS), platform screen doors, passenger information… This is the shared integration tier for all station and line M&E systems.

China's modern ISCS architecture uses a Front-End Processor (FEP) as a conversion layer: the FEP translates different hardware interfaces and software protocols, while isolating data flows between the ISCS and each subsystem.¹² The practical significance is large: if the employer requires an open architecture using a documented-protocol FEP, the integrator cannot hard-lock the system — because every subsystem communicates through a standardised, publicly documented interface layer.

▸ A parallel pillar: certification equivalency recognition

Interface standards solve "can systems talk to each other." But there is a second type of lock-in, especially important for mechanical components, materials and equipment: lock-in at the certification tier. A technically equivalent product bearing the "wrong" certification mark — say, meeting China's GB standard but facing a project requirement for UL/FM — is legally excluded even if technically perfectly usable.

This is a non-technical barrier: a documentation recognition problem, not a quality problem. The solution is a mutual recognition / equivalency mechanism: the regulatory authority accepts major certification schemes (UL ≈ FM ≈ CE ≈ GB/CCC ≈ JIS) as evidence of meeting the same performance level, together with a cross-reference table between classification systems. When this is in place, high-quality equipment from multiple supply sources qualifies for tender — broadening competition while maintaining technical consistency.

▸ An easy-to-miss detail: shared testing platforms

Chongqing did not only issue standards — it also built an interoperability testing platform and an on-board adaptation system to verify that multi-vendor equipment was genuinely compatible before operations began.¹¹ The lesson: an interface standard is only valuable if there is an independent verification mechanism for compliance. A standard without an accompanying certification laboratory is merely a declaration on paper.

When telling the story of China's urban rail achievement, attention usually goes to hardware — trains, track, signalling. But across all three processes analysed above (investment, localisation, interface standardisation) there is a quiet actor serving as the connective tissue: consulting firms and design institutes. They are simultaneously the project designers, the institutions that absorb and retain technological knowledge, and the drafters of the standards themselves. This is the "soft infrastructure" without which all three other processes cannot function.

▸ In-house design institutes: keeping knowledge domestic

A typical Chinese model is the design institute owned by the local metro operator. The Guangzhou Metro Design and Research Institute (GMDI), established in 1993 as a subsidiary of Guangzhou Metro Group, employs approximately 1,200 engineers, has contributed to the design of more than 1,000 km of urban rail, achieved revenue of around CNY 2.4 billion in 2021, and was the first A-share listed urban rail design firm in China.¹⁴

The strategic significance of this model: the employer develops consulting capability within its own organisation. Technical knowledge accumulated line by line stays in the country, rather than leaving with the foreign contractor when each project closes. This is precisely where the "absorptive capacity" emphasised in Section 2 is made real — without a sufficiently capable domestic consulting team, technology transferred on paper cannot permeate into practice.

▸ Standard drafters, not merely standard appliers

The second tier comprises national survey and design institutes, most of which evolved from former Ministry of Railways institutes. China Railway Design Corporation (CRDC, Tianjin, formerly the No. 3 Railway Survey and Design Institute, founded 1953) employs more than 4,800 engineers, has surveyed and designed more than 44,000 km of railway (about one-third of China's total network) and operates the National Engineering Laboratory for Digital Construction and Evaluation of Urban Rail Systems.¹⁵

The most instructive point is this: consulting institutes do not merely apply standards — they write them. China Railway Engineering Consulting Group (CEC) has participated in drafting 17 national standards and 104 industry standards and contributed approximately 80% of the railway sector's standardised drawings.¹⁶ This is the link connecting "interface standards" in Section 3 with actual technical capability: the 252 experts in CAMET's nine committees are drawn largely from these institutes. The T/CAMET interoperability standards were not a product of administrative desk work — they are a crystallisation of the design institutes' accumulated project experience.

▸ Full-lifecycle consulting and independent verification

The third tier is turnkey consulting over the full project lifecycle: survey, master planning and feasibility, design, construction supervision, EPC contracting, commissioning and handover.^14,15 A single general consultant spanning the whole lifecycle preserves technical consistency across subsystems and accumulates lessons across lines — especially important when the goal is whole-network synchronisation.

Finally comes the role of Independent Safety Assessor (ISA). The Chongqing interoperability demonstration project involved three independent safety certifiers simultaneously.⁹ The independent consultant is the party that verifies that suppliers actually comply with the interface standard and meet the required safety level — ensuring that the standard is not only stated on paper but enforced and objectively confirmed.

The implication for Vietnam is clear: to localise and standardise, the country must simultaneously develop domestic consulting capability — nurturing design institutes attached to employers, involving consultants in standard-drafting, and using independent consultants as the parties that verify equivalency and confirm compliance.

Vietnam is at the start of a major metro investment cycle in Hanoi and Ho Chi Minh City. The question is not "which standards system is safest" — because all mature systems (GB, EN/BS, NFPA, Japanese) ensure safety when correctly applied. The real question is how to achieve synchronisation, spare-parts compatibility, vendor flexibility and maintainability over 30 years of operations. China's experience offers a very concrete framework for answering it.

▸ Two layers: what to copy and what not to

The easy mistake is to try to copy the entire Chinese story. In fact, it has two clearly separate layers with very different feasibility for Vietnam:

The central lesson: China achieved its sharing and interoperability objectives primarily through Layer 2, not Layer 1. Even when equipment comes from multiple suppliers across different heritage systems, a shared interface standard still allows cross-fitting, through-running, and shared spares. Vietnam can skip Layer 1 and still capture most of the benefit — if it executes Layer 2 well.

▸ Distinguishing the two OCC/SCADA tiers to avoid overstating the problem

A common confusion is to bundle two very different tiers together:

Most of China's "defeating monopoly" story relates to the signalling tier. But for M&E systems generally, the relevant tier is ISCS — which is far more tractable if an open FEP architecture is specified from the outset. Transplanting the signalling tier's difficulty onto the M&E tier is an unnecessary over-statement of the problem.

▸ Vietnam metro employers' actual levers

A Vietnamese metro employer often feels weak because it can only propose applying a standards system, not legislate a new national code. But China's experience shows that the strongest lever sits at a different tier entirely — one the employer already controls:

In other words, to achieve synchronisation and vendor flexibility, the employer does not need to wait for a new national standard. They need to write the right clauses into the tender document and contract: interoperability interface standard, open FEP architecture, certification equivalency recognition (UL/FM/CE/GB/EN), and full handover of licences and configuration tools.

▸ Recommended actions for employers

Synthesising the Chinese experience, the following specific actions are available to Vietnamese metro employers immediately:

On the consulting role: the substitution of "equivalent" equipment to maintain vendor flexibility must be defined on a form-fit-function basis and confirmed by the design consultant, not merely "equivalent by catalogue comparison." An item that appears equivalent on paper but deviates on a critical technical parameter may break the calculations or logic of the entire system.

China's urban rail achievement is usually told through enormous numbers. But for a country at the investment stage like Vietnam, the most valuable lesson lies in the institutional dimension that rarely receives attention: how the state used approval authority and the industry association used standards to compel the market to follow a common interface framework — achieving synchronisation and interoperability while preserving competition between suppliers.

Vietnam cannot copy China's scale or self-manufacturing capacity, and does not need to. What needs copying is the mechanism: define interoperability interface standards, require open architecture and accept equivalent certifications, embed all of this in tender documents as mandatory conditions — and establish an authoritative body to maintain those standards over time. The four objectives every metro employer pursues — safety, synchronisation, vendor flexibility, manageable whole-life cost — are not in conflict. They are solved simultaneously by a single instrument: interface standardisation led by the state and enforced through employer authority.