Urban rail safety: Lessons from the US

July 15, 2011

Samuel Zimmerman, Liu Zhi China Daily

The tragic accident that occurred at an escalator of Beijing Metro Line 4 on July 5, 2011, is shocking to many people in China. It has triggered intensive discussion among netizens about the causes of the accident.

Naturally, metro riders express their concerns with metro safety. Their concerns are legitimate. As cities are rapidly building and expanding metro systems, more and more people will use metros regularly. Safety is one of the key factors in their choices for the mode of transportation.

It may take time for the investigation to reach a final conclusion what caused the accident. Thus it is too early to say what specific actions are needed to prevent the same type of accidents from happening again. There are, however, some general lessons from other metro systems around the world that are appropriate to consider.

International experience tells us that the causes of urban rail accidents fall into six general categories: design issues, construction/manufacturing quality, human operating errors, natural events during operations such as earthquakes, snow/ice storms and flooding, unaddressed deterioration of equipment, infrastructure and facilities over time, and institutional issues related to safety.

A fatal accident in Washington D.C. on June 22, 2009, on its extensive metros, the second busiest and one of the newest in the US, provides an excellent case study for explaining the meaning and scope of these categories. This event has been meticulously and thoroughly studied by local officials in Washington and the National Transportation Safety Board, and the root causes were found similar to those for metro accidents elsewhere in Europe and Asia. There are some lessons that can be extracted for Beijing, the rest of China and the entire world.

The accident occurred at the beginning of the afternoon peak hour. A train struck the rear of another which had stopped at an overground station. The front car of the moving train collapsed. Nine people in that car were killed and 52 in both trains were injured. The immediate cause of the accident was the failure of the fully-automated train control system to detect the stationary train and send a signal to the moving train to stop. People were killed because the front of one of the cars was crushed as it (first put into service in 1976) was structurally under-designed.

The National Transportation Safety Board found four basic problems: failure of the authority to effectively maintain and monitor the performance of its automatic train control system; lack of a safety culture at the Washington Metropolitan Area Transit Authority; ineffective safety oversight by the board of directors of the authority; and lack of safety oversight by both the independent local safety body required by the federal government law and the Department of Transportation's Federal Transit Administration.

What are the implications of these findings for other places?

First, the importance of institutions. Simply put, each authority or company building and then operating an urban rail system should be organized and operated to give safety and security its highest priority. This should not only be through creation of a safety directorate reporting directly to the highest management level, but also through formal, ongoing operations and management activities including staff capacity building and performance evaluation.

Even where this is done, there is still a need for second and even third level independent oversight and technical support. Safety is so important to the success of public transport, that a second independent oversight institution should be created at the local or provincial level, funded and empowered to carry out routine safety assurance activities. This institution should conduct routine safety audits and studies as well as respond to any event with implications for the long-term safety of the system.

The third level, or national oversight level (for example, the Federal Transit Administration of the US Department of Transportation) is important for providing technical assistance and capacity building for the planning, design, management and operating personnel of all systems in the country, conducting necessary research to address common needs, providing national benchmarking data and even performing periodic audits.

A second important general point coming out of the Washington accident was the need to pay more attention to keeping the entire system, not just the signal/train control system, in an acceptable, safe state of repair.

In 1986, when the Washington Metro was less than 60 percent its current size, a study was done by the federal government that addressed, for the first time the issue of what financial resources were required to maintain the system (whose first segment had opened only 10 years before) in a state of good, safe repair.

The study found that an average of approximately $200 million in constant 1986 dollars would have to be spent to maintain the metro and bus systems in a state of good repair. Unfortunately, local authorities were not able to follow up with the necessary funding.

In October 2010, the machinery on an escalator at a busy station failed and the escalator full of people went out of control downward, injuring many people, some seriously. Thankfully, this year's capital budget for Washington Metro is $1 billion, compared to an operating budget of $1.5 billion.

The interest of local authorities in building new, often glamorous infrastructure, while paying less attention to its ongoing maintenance, repair and replacement needs, is not unusual. In the case of Washington and elsewhere, it has led to many reliability and safety issues that could have been avoided. It is never too early to begin planning for the ongoing costs of maintaining the complex infrastructure, equipment, systems and machinery that comprise urban rail system.

Despite accidents in Beijing, Washington D.C. and elsewhere, by far the safest ways to travel in any city are on public transport, especially urban rail system. Accident, injury and fatality rates for urban rail are much lower than the equivalents for private cars.

Public transport safety will continue to improve if authorities and public transport companies take lessons learned seriously from the tragic accidents. The World Bank stands ready to assist - by bringing international knowledge and expertise and introducing international best practices.

Samuel Zimmerman is urban transport consultant, and Liu Zhi is lead infrastructure specialist with the World Bank.