Zhang Fusheng: Insights from the Evolution of International Traffic Signal Standards
Editor‘s Note: On March 6, at the sub-forum of the 14th Intelligent Transportation Market Annual Conference (ITSMRS 2025) held in Suzhou, Zhang Fusheng, a researcher at the Beijing Key Laboratory of Intelligent Traffic Control at North China University of Technology, delivered a keynote report titled Insights from the Evolution of International Traffic Signal Standards. Through a meticulous study of standards from the UK, EU, U.S., and other regions, the report analyzed the historical context, version evolution, systemic relationships, and drafting methodologies of these standards. The insights drawn provide valuable guidance for professionals in the traffic signal industry.
Editor's Note: On March 6, at the sub-forum of the 14th Intelligent Transportation Market Annual Conference (ITSMRS 2025) held in Suzhou, Zhang Fusheng, a researcher at the Beijing Key Laboratory of Intelligent Traffic Control at North China University of Technology, delivered a keynote report titled Insights from the Evolution of International Traffic Signal Standards. Through a meticulous study of standards from the UK, EU, U.S., and other regions, the report analyzed the historical context, version evolution, systemic relationships, and drafting methodologies of these standards. The insights drawn provide valuable guidance for professionals in the traffic signal industry.
01 Why This Topic?
I am a conservative thinker who prefers to approach problems through long-validated common sense and fundamental logic. I maintain a cautious, skeptical, and even apprehensive stance toward many "groundbreaking" or "disruptive" claims, often voicing unpopular critiques.
Before diving into my report, let me share a recent online article I came across. The well-known blogger "Liu Shenleilei" recently published a piece titled Humanity May Have Missed the Window for Self-Rescue. The core argument is that the "window for self-rescue" refers to the race between the spread of common sense and the acceleration of human folly. Truth often struggles to emerge, while biases and fallacies dressed in sensationalist packaging spread like wildfire.
Similar dynamics are unfolding in our industry. Are we, too, missing our "window for self-rescue"? This question warrants deep reflection.
I chose this topic to scrutinize the industry’s current state and seek actionable solutions. With 25 years of experience in traffic signal systems—spanning product development, project implementation, and academic research—I have engaged in drafting, revising, and reviewing standards, including China’s GB/T 20999-2017 (Data Communication Protocol Between Traffic Signal Controllers and Central Systems). My observations are sobering.
Industry Challenges:
Perpetual "Empowerment": New entrants to the traffic sector invariably prioritize "revolutionizing" signal control, touting "disruptive" solutions without understanding foundational principles.
Forced Endorsements: The industry is often pressured to endorse unproven innovations, where vague claims of "signal optimization" mask a lack of rigorous analysis.
Scapegoating: Traffic signals are blamed for every issue—congestion, safety failures, accessibility gaps, transit delays, even air pollution.
Low-Barrier Experimentation: Newcomers rush to develop "revolutionary" controllers, systems, or algorithms without mastering basics like actuated or coordinated control. Terms like "AI" and "adaptive" are thrown around recklessly.
Education-Industry Disconnect: Graduates (undergraduate and postgraduate) struggle to apply classroom knowledge due to fragmented standards. For instance, students learn System A but encounter System B in the field, with incompatible concepts and workflows.
Fragmentation: Siloed practices plague product development, project delivery, and operations. Even within the same company, teams use disparate terminology. Isolated systems, crude standards, and zero interoperability have become chronic ills.
Superficial "Unified Platforms": Recent attempts to create unified control platforms ignore granular details. For example, smooth transition between timing plans involves nuanced methods (e.g., residual-based, cycle-length adjustments), yet no consensus exists on functional definitions. Most systems operate as "black boxes," rendering platform integration a hollow exercise.
Hypercompetition ≠ Healthy Competition: The industry fixates on buzzwords—"big data + signals," "blockchain + signals," "digital twins + signals"—while neglecting implementation specifics. Price undercutting, not technical excellence, dominates bidding wars, squeezing profit margins and degrading service quality.
Lifecycle Failures: Projects often prioritize handover over sustainability. Post-construction, systems decay as innovative features promised during bidding are abandoned. Delivering repeatable success and iterative progress remains elusive.
R&D Dilemmas: Emerging demands (e.g., vehicle-road-cloud integration) and disruptive technologies force reactive development, sidelining reliability and sustainability. Custom code per project and version sprawl (e.g., 80+ intersections with 20+ firmware versions in one city) exemplify this chaos.
Globalization Gaps: As the industry eyes international markets, the lack of globally aligned standards and knowledge systems raises questions: Will we export low-end products or systemic solutions?
A Root Cause: Dysfunctional Tendering
A persistent frustration is the tendering process. Bidders submit 100+ page technical proposals evaluated in mere hours, reflecting users’ unclear requirements and evaluation criteria. This farce leads to adversarial, exhausting negotiations, with projects often fizzling out. Both users and providers crave escape from this cycle.
02 Standards Under Study
I focused on influential European and U.S. standards, particularly those tied to my work on SCOOT-related systems:
UK: Updated traffic signal manuals (e.g., TR0141 → TR2210 → TOPAS2500 series), UTMC protocols.
EU: EN12675 (signal controllers), EN50556 (control systems).
U.S.: NEMA TS2 (hardware), ATC5201 series, NTCIP1202 (v1–v4), MUTCD 11th Edition, and CTI4501 (connected intersections, 2022).
03 Key Observations
Case 1: NTCIP1202 Evolution
1996 (v1): 94 pages, foundational.
2001 (v2): ~200 pages, expanded. Widely referenced in China’s ring-structure controllers.
2019–2023 (v3): Overhauled to address V2X demands, incorporating systems engineering (needs → functions → implementation).
2024 (v4 draft): Transcends pure signal control, detailing architecture, user needs, and future-proofing.
Modern Standards: Beyond Technical Specs
Broader Audience: Standards now target managers, integrators, operators, and developers to align stakeholders.
Rich Content: CTI4501 (2022) exemplifies rigorous connected intersection specs, clarifying subsystem responsibilities and interfaces.
Modularization: NTCIP1202 decoupled V2X aspects into standalone NTCIP1218.
Version Analysis via AI: Large language models aided cross-version comparisons, highlighting changes (e.g., phase parameters in v2 → v4). Revisions reflect evolving scenarios, like pedestrian timing adjustments during plan transitions.
04 Key Takeaways
Systems Engineering Approach:
Standards like NTCIP1202/CTI4501 start by exhaustively mapping user needs (50+ pages in 1202) to product functions. This granular alignment ensures solutions address real-world demands.
Safety by Design:
International standards mandate independent storage/modification locks for critical parameters (e.g., conflict checks, clearance intervals). Some require dedicated hardware (e.g., U.S. CMU/MMU units). In contrast, China’s pursuit of hot-swappable components or remote upgrades often sacrifices safety.
Standards ≠ Rigid Rules:
The U.S. MUTCD emphasizes engineering judgment 120+ times. Professionals must adapt standards to context, as no document covers all scenarios.
Supporting Ecosystem:
U.S. standards are bolstered by handbooks (detection, timing, design, safety) offering case-based guidance.
Future-Proofing:
CTI4501 addresses connected intersections’ interoperability and safety. MUTCD’s Part 5 outlines infrastructure requirements for autonomous vehicles (e.g., signage legibility for machine vision).
Appendices in modern standards (e.g., 1202v3+, CTI4501) flag emerging research areas like peer-to-peer intersection coordination.
Validation Before Innovation:
Many standards mandate rigorous testing before deploying new technologies, curbing reckless experimentation.
05 Closing Thoughts
China’s traffic signal sector has advanced rapidly over 30 years by adopting foreign systems. While domestic alternatives are inevitable and desirable, I fear insularity—discarding global expertise in the name of "localization." To avoid becoming frogs in a well, we must stay humble, open, and globally engaged. Only then can we build systems that are not just "Chinese" but world-class.
Pic:pixabay
