Civil infrastructure is subjected to multiple deterioration processes (e.g., gradual deterioration and shock dete-rioration) caused by environmental exposure and extreme events during its lifetime. To maintain performance and functionality, maintenance actions should be performed and the life-cycle cost may be affected. There is a need to explore the effect of maintenance actions and various uncertainties on the life-cycle performance of the engineering systems. This study proposes a probabilistic life-cycle analysis framework for civil infrastructure based on performance indicators, e.g., reliability and maintenance cost. Stochastic uncertainties resulting from multiple dependent deterioration processes, system reliability, intervention actions, and maintenance cost are considered. In particular, the dependence between the maintenance interval and cost is highlighted. Previous studies generally assume they are independent. Such an assumption can be misleading and lead to inappropriate cost estimation. To address this concern, a copula-based multivariate renewal model is proposed to assess the life-cycle maintenance cost analytically and numerically. In addition to the expected cost, statistical moments (e. g., standard deviation, skewness, and kurtosis) are calculated to quantify uncertainties from higher-order mo-ments. Two illustrative examples show that the dependence and uncertainties can have a large impact on the life -cycle cost, and decisions can be altered by considering statistical moments of the cost.