13. Support Method for Extending Inspection Periods Based on Statistical Analysis of Equipment Inspection Records

Assessing equipment condition in detail and with continuous values, it is possible to extend inspection periods without lowering the current management standards. Since a lot of equipment has inspection records that regularly document their condition through daily inspections, we developed a method that utilizes the statistical properties of these inspection records to determine the feasibility of extending inspection periods.

One of the challenges in analyzing inspection records statistically was the use of discrete evaluation scales like ○(normal) and △(failure), and the need to account for individual differences that cannot be explained solely by easily accessible factors from ledgers (such as the number of operations in point machines). To address this, we developed a statistical analysis method that models failure occurrence using a probabilistic approach while also accounting for differences in failure occurrence frequency among individual equipment.

This method enables the quantification of the relationship between service life and failure occurrence rate (risk) for each piece of equipment based on past inspection records. If the maximum risk under the current inspection period is set as the allowable value, the feasibility of extending the inspection period can be evaluated by comparing it with the failure occurrence rate of each piece of equipment after the extension. For example, in the case shown in Figure 1, even if the inspection period is extended from the current 90 days to 120 days, the failure occurrence rate for the majority of equipment remains below the current allowable value, indicating that the extension is feasible. Furthermore, for some equipment whose failure occurrence rate exceeds the allowable value after the extension, it is still possible to consider extending the inspection period by setting conditions such as implementing monitoring measures. This method was applied to 280 point machines on an actual railway line, supporting the railway operator’s decision to extend the inspection period by approximately 1.3 times.

This method was developed as a functional module of an integrated analysis platform for railways, which enables centralized management and analysis of various types of maintenance data, including location information. It can also be applied to various types of equipment where inspection records are stored in a similar format, and can be used to review and optimize inspection periods for them.