Abstract:

In the context of project portfolio management，a critical chain project is faced with brittleness risk from outside the project because of the interaction between multiple projects. Compared with the risks faced by various activities within the project， a more serious impact of brittleness risk on the performance of project buffer management， project duration， and project cost may be caused. Namely， the project duration delay and cost overrun still can’t be avoided especially due to the brittleness risk.However，in order to solve the above problems and improve the efficiency of buffer monitoring and the overall project， the existing buffer monitoring methods mainly focus on the common risks from within the project.The brittleness risk from outside the project is neglected. In addition，most of the approaches put the buffer at the end of the activity before monitoring the buffer， which will lead to the Domino effect and can’t ensure the rationality of buffer allocation.Therefore， the research problem in this paper is that considering and analyzing the brittleness risk in critical chain project portfolio scenario to build a quantitative model to make the buffer allocation more reasonable and buffer monitoring more efficient.According to the research problem， a new method is proposed to monitor the buffer dynamically.Firstly，resource， technology， and market of the project are considered to measure the brittleness risk by using the niche theory. After determining the buffer allocation ratio， the project buffer is allocated reasonably and the buffer positions are designed for project-level brittleness risk and activity-level risk. Secondly，the trigger points of buffer monitoring are set according to the buffer allocation. Then，a dynamic buffer monitoring strategy of cumulative buffer and reallocation is constructed， which ensures continuous buffer update and avoids buffer waste. Finally， the dynamic buffer monitoring strategy is implemented to monitor and control buffer consumption in real time.The simulation results of Monte Carlo show that the new method has better performance in indicators such as warning frequency， project duration， and project duration， thus the proposed buffer monitoring method can effectively improve the comprehensive performance of the project through reasonable buffer allocation and monitoring.Based on the research results of this paper， our research will help relevant scholars better to understand the brittleness risk in critical chain project portfolio scenario and solve the dynamic buffer monitoring problem.In practice， a new perspective is also provided for the project decision-makers or managers to manage the critical chain in the complicated environment.

Key words: critical chain project management, portfolio scenario, brittleness risk, buffer allocation, dynamic buffer monitoring