Journal of Japan Society for Safety Engineering

When a product involving a substantial hazard to consumers or property becomes apparent, a business operator needs to prevent sequential accidents. Many business operators might decide to voluntarily recall the product in the market. To assist the decision, a quantitative method based on the use of R-Map has been proposed by the Ministry of Economy, Trade and Industry in Japan. The R-Map, a risk assessment method, has two criteria. One is the possibility of a hazardous accident; the other is the severity of health/safety damage. A method of modifying the R-Map decision by combining these obvious criteria with some qualitative circumstances has been developed. This hybrid assessment procedure is often called a risk bias method, which aims to realize more appropriate decisions in line with an acceptable level in society. Several risk biases have already been pointed out by previous research. In this research, we investigated several trends by analyzing recall cases in Japan. We successfully found some tendencies to encourage voluntary recalls. One of these appeared in accordance with legislative changes requiring business operators to endeavor to conduct recalls in 2007. Moreover, some types of product were recalled due to quality problems rather than safety problems. Then, we indicate that these tendencies are risk bias factors.

View full abstract

Show abstractHide abstract

Gas explosion in confined spaces generates a sudden rise of internal pressure and can result in extensive damage such as destruction of the structure, generation of blast waves, fragment scattering, and heat release. Predicting pressure behavior is necessary for risk assessment and safety management at places, such as chemical plants, where combustible materials are used. In the past, many explosion experiments have been performed and the data such as pressure histories and K_G values have been obtained. However, in those experiments, less than tens of liters vessels were used and their volume were very small compared to real cases. It will cause underestimation of the risk to full-scale explosions because the effect of flame instabilities is neglected in small scale experiments. In fact, some studies showed that the flame accelerates in large scale because of instabilities of the flame front. In order to prevent the underestimation of the risk for gas explosions, the effect of flame front instabilities has to be considered. In this study, we considered the effect of the flame front instabilities by regarding the flame front as a fractal structure and proposed a new method which could predict the pressure behavior properly even in a large scale. We verified the accuracy of the new method by experiment. In addition, the correction of the K_G value in a large vessel was proposed.

View full abstract

Download PDF (1371K)

Register with J-STAGE for free!