Abstract:

Buffer management is a crucial component in critical chain project management （CCPM） as it helps address the risks and uncertainties inherent in project execution. Traditional buffer management methods primarily focus on risk characteristics and attribute factors to allocate and control the project buffer in a static manner. These approaches often fail to account for the dynamic nature of the project environment， leading to inefficient buffer configuration and excessive buffer consumption. As a result， projects may suffer from suboptimal resource allocation， increased costs， and delays due to poor responsiveness to real-time project variations. To address these challenges， a dynamic buffer allocation and monitoring method is proposed for critical chain project based on investment portfolio theory. Unlike traditional methods， the proposed approach reconceptualizes buffer allocation as an optimization problem aimed at maximizing expected buffer returns， subject to the constraint that buffer returns second-order stochastically dominate the baseline returns. To achieve this， a second-order stochastic dominance-based optimal buffer allocation model is established， explicitly incorporating buffer control costs to enhance resource utilization efficiency. Furthermore， the proposed method integrates uncertainties in project activity durations， as well as dynamic changes in activity link execution and buffer consumption patterns， to construct a buffer action matrix for adaptive buffer allocation and monitoring. By continuously updating buffer distribution based on project progress and risk fluctuations， the model ensures that buffer resources are optimally allocated where they are most needed， minimizing unnecessary buffer waste while maintaining high project completion reliability. Simulation results demonstrate that， compared with the traditional static allocation method， the proposed method offers significant improvements in several key performance indicators， including the probability of on-time project completion， average buffer consumption ratio， actual project duration， and effective project output. This enhancement in buffer efficiency not only boosts the adaptability of critical chain project scheduling but also improves project stability and risk absorption capability. The findings of this study provide valuable theoretical insights and practical strategies for advancing buffer management practices， offering meaningful contributions to the management of complex projects characterized by uncertainty and variability， with direct implications for improving project performance and ensuring timely delivery.

Key words: critical chain project management, buffer monitoring, portfolio management, schedule control, uncertainties