A Semi-Markov Maintenance Model for Bridges Synthesizing with Four Types of Uncertainties

Abstract

Bridges are one of the most critical infrastructures, and their maintenance management has aroused widespread concern due to the high expenditure and serious consequences after the failure. Most of the current studies on bridge maintenance only explored limited uncertainties, which can hardly match the real dynamic situation. Hence, we formulate a semi-Markov maintenance model for bridges synthesizing with four types of uncertainties. The uncertainties include the uncertainty of the state, the uncertainty of the deterioration rate, the uncertainty of the maintenance effect, and the uncertainty of the inspection interval. These types of uncertainties are all common in actual bridge maintenance. They have effects on the maintenance decision and increase the difficulties of bridge maintenance. Our approach considers these uncertainties and enables a more realistic model for bridge deterioration and maintenance. In the numerical example, we explore the impact of the uncertainty of inspection interval on the optimal policy. The results demonstrate that a large inspection interval may lead to the conservative optimal policy. This approach can be extended to other similar scenarios.