In information system projects (ISPs), successful project completion under time and budget constraints significantly impacts project performance and feasibility evaluation. Traditional analytical methods, such as the critical path method, often relies on deterministic durations or restrictive probabilistic assumptions, making them impractical under uncertainty. These methods require complex boundary derivations once stochastic durations are considered—such as beta distributions for activity durations—and suffer from combinatorial explosion due to the probabilistic nature of activity durations. This leads to NP-hard problems, significantly increasing computational complexity, reducing modeling flexibility, and limiting applicability in highly uncertain practical environments. To address these limitations, this study employs stochastic project networks for modeling and analysis. The project network adopts an activity-on-arc (AOA) structure, wherein each activity has multiple possible execution states characterized by specific durations, costs, and occurrence probabilities. The project reliability is proposed to indicate the performance of the project network, where reliability is defined as the probability that a project meets both time and budget constraints. To overcome the computational intractability and modeling rigidity of traditional analytical methods in stochastic project environments, a Monte Carlo simulation approach is proposed. Within the simulation framework, activity durations are randomly sampled according to their probability distributions, and costs are calculated based on selected states. Validation through simulation of an ISP shows that the proposed approach accurately estimates project reliability without complex boundary derivations. The results indicate strong computational efficiency, improved scalability, and practicality, making it a valuable decision-support tool for feasibility and performance evaluation in uncertainty-driven project environments.