Corrosion-induced damage is a major source of deterioration in infrastructure and industrial systems such as bridges, offshore and onshore structures, and underground oil and gas pipelines. The uncertainty is pervasive in the parameters affecting the evolution of corrosion process. Risk assessment and management of these systems require a suitable dynamic description of the corrosion process that sufficiently accounts for the underlying uncertainty and subsequently propagates it into the lifecycle reliability assessment of these systems. In this paper, we present a novel approach for reliability-based life cycle management of buried pipelines subjected to corrosion damage. We view our main contributions as twofold. First, a probabilistic model for time evolution of corrosion growth is constructed from available data using polynomial chaos formalism. The model is used to systematically propagate the underlying uncertainty into the limit state functions and the lifecycle reliability analysis. Second, we propose a computationally efficient and accurate optimization strategy using polynomial surrogate in order to solve the stochastic optimization associated with the lifecycle management of buried pipelines. The proposed method facilitates the optimization of maintenance scheduling to achieve minimum expected lifecycle cost by allowing implementation of a gradient-based optimization scheme. This is generally a challenging task due to the discontinuous nature of the objective function with respect to design variables. We examine the sensitivity of the optimum maintenance scheduling with respect to the different measures related to failure probability representing different risk strategies. The proposed development provides an uncertainty-aware decision support tool for making more informed decision regarding the lifecycle management of corroding pipelines. This formalism can also be adapted to other deterioration mechanisms that result in damage-induced structural failure over the lifetime.