[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

Recent years, "System for Report and Publication of Product Accident Information" is introduced, so it became easier to get the safety accident information of other companies (or other industries). But those information are utilized in the prevention have been extremely rare. In this paper, in order to utilize the published information to prevent safety accident, we have applied each accident mechanism to the network model, and have represented and organized the various characteristic values by using these models. As a result, we have devised a high versatile method of extracting the potential technical issues. In that model, macro analysis of safety accident information that focuses on reaction mechanism length by using the network analysis (mainly adjacency matrix) can visualize the states and state transitions. Manufacturers can be expected that it is possible to clarify the issues by comparing between product accidents of other industries and their own product accidents. They are issues that cannot be extracted only from their own accident information, can be expected to be very effective in prevention.

View full abstract

Railway signalling systems need high level of safety, and are also required to achieve higher availability. In the past, there were some analyses for effective measures from the viewpoint of the minimization of total costs. However, these analyses are the bottom up approach and are not from the view point of policies of measures, which are directions from the view point of frequencies and consequences of failures, required by systems for improving availability. Therefore, a strategic decision method of effective measures for the achievement of targeted availability is proposed. This method can consider failure frequencies per unit time and the duration of downtime of train operation at the same time, and by applying the proposed method to a railway line, it is shown that it can contribute to applying effective measures depending on kinds of failures and characters on the line.

View full abstract