Abstract
This study is intended to investigate a way to consider changes in temperature and vehicle weight as environmental and operational factors for long-term bridge health monitoring by applying a Bayesian approach to long-term monitoring data and artificial damage data. The Bayesian approach consists of three steps: step 1 is to identify damage indicators; step 2 is to calculate residuals by means of the Bayesian regression; step 3 is to make a decision utilizing the probability of residuals within a threshold and the Bayesian hypothesis testing. Observations showed that validity of using the data observed at a specified time to reduce the influence of traffic loads can be confirmed. In the Bayesian hypothesis testing utilizing data from the healthy bridge, the probability of the bridge damage was judged as `very small'. The number of data can influence the results of the Bayesian hypothesis testing.
