Research and Development of Railway safety

Katsuji AKITA
General Manager, Technological Development Dept., RTRI


  By expanding and improving the function of safety equipment, the number of accidents occurring on Japanese railways has decreased to 1/4 in the last 20 years, but currently there are still more than 1000 accidents occurring a year. Approximately 50% of those are accidents that happened at crossings. Since its inception in 1997, the Railway Technical Research Institute has been engaged in research and development of safety controls for trains, analysis of train movement characteristics and safety, technology of safety measures for crossings, the human factors in train operation, construction and maintenance work, and technology for prevention of natural calamities.
  With regard to safety controls for railways trains, fail-safe technology for computer systems, an ATP (Automatic Train Protection) system for high-density operation lines, which transmits digital information between ground and railway vehicle using track circuits, and other measures have been developed. Furthermore, we have developed the CARAT (Computer And Radio Aided Train control) system which, using wireless transmission, allows a train to self-control the running speed and the distance from the train ahead of. For safety systems, achieving quality in software is mandatory. Based on the Formal method, we are also tackling highly reliable design technology. In addition, a safety technical guidance has been established, which aims at ensuring safety throughout the life cycle.
  With regard to technology of safety measures for crossings, we are committed to improvement of visibility of crossing equipment taking the human factors into consideration. We are also working on obstacle detection technology for motor vehicles that enter a crossing while the warning is sounding, and evaluation of the danger level for each crossing based on accident data. Along with higher train speeds, lighter weights, and more intersections with roads, the risk at crossings has increased. For intersections with roads, we are developing an integrated crossing safety system operating in conjunction with intelligent transport systems under development for road traffic. Moreover, we are also engaged in formulating a design guidance for train interior fixtures that factor in passengers' safety through analysis of their behavior during a train collision, and research and development of train structure with superior crashworthiness.
  Japan is a country with numerous earthquakes. The UrEDAS (Urgent Earthquake Detection & Alarm System) which can detect weak earthquake tremors, estimate the scope of the earthquake, and perform an emergency stop of trains prior to arrival of strong oscillations, has been developed and employed on major railway routes. Moreover, using the opportunity provided by the 1995 southern Hyogo earthquake, which caused major damage to railway equipment, seismic design standards for railway structures were re-formulated. In addition, diagnostic technology for the soundness of structures and repair methods have been developed. Progress has been made in research and development of technology for shortening the braking distance of train emergency brakes, analysis of behavior of moving trains during an earthquake, and quantitative evaluation of safety. In order to evaluate the impact of strong winds on trains moving over bridges and banks, we have conducted wind tunnel tests and analytical study. We are also engaged in quantitative policies for train operation control, forecasting technique of slope accidents caused by rain, and evaluation of the effect of slope protection work. Along with constructing a safety database, the development of methods of forecasting safety based on probability theory is also an issue.
  With regard to the human factors, we are engaged in research into the occurrence mechanism of human error, and a psychological and physiological evaluation technique of driver workload, by constructing a database of incidents and accidents, and conducting analysis of the chain of events leading to an accident, and using train operation simulators.
  With our goals for the railways of the twenty-first century, we are promoting these system changes by introducing state-of-the-art technology. Among these trends, safety technology requires challenging areas unknown simply by extending former empirical engineering. The Railway Technical Research Institute plans to promote research and development as a total system including boundary issues, aiming for a system of railway safety technology and a method of evaluation.