Abstract
This study analyzed the relationships between repair costs and the period based on actual data from a pumping station. The actual data were collected by surveys of the model site and many other areas. This paper shows a method that predicts future repair costs based on survey results. When managing many pumping stations, it is desirable to predict the cost from the same sites. As a result of the calculations, the precision was high, and approximately one-tenth of the predicted value by the Ackerman curve. Further studies are necessary to confirm the proposal method using a large amount of data.
