A new buffer sizing approach based on the uncertainty of project activities

摘要

A new buffer sizing method based on the uncertainty of the project is proposed in order to improve the accuracy of buffer management. Assuming that the project activities follow lognormal distribution, the model proposed first analyzes the influence of uncertainty, which is characterized as the main factor to affect the activity's duration. Fuzzy theory is then introduced to calculate the uncertainty and is adjusted using resource tightness and network complexity. Finally, a new buffer sizing model based on the uncertainty and the project attributes is introduced. The effectiveness of the method is verified by employing the Monte Carlo simulation, and the actual duration and cost are obtained on the basis of the mean and variance of the project. The experimental results show that the buffer obtained using this method is smaller than the cut and paste method but larger than the root square error method. These findings indicate that the new model is able to avoid the excessive protection issues of the cut and paste method and overcome the root square error method's problem as regards an insufficient consideration of project attributes. The method proposed fully considers the factors that affect buffer sizing, signifying that it can provide the project with effective protection and an appropriate buffer size.