Abstract : Construction projects play a great role in the economy of any country, also it is a great source of income for investors and stakeholders. Time is related to cost in any project, it is one of most important aspects to be planned and scheduled. The accuracy of project duration was the concern of planners and estimators throughout the project management history, many tools and methods were used to get this accuracy, at first the traditional methods such as critical path and program evaluation and review techniques were used. Then a relatively new method, which is critical chain, was introduced to overcome the pitfalls found in the traditional methods and to increase the accuracy of the planned project duration. This will decrease cost and time overruns on one hand, and will increase the profit of the project on the other hand. The aim of this research is to increase the accuracy in the planning of the project duration by developing a mathematical model for project buffer to create a critical chain that will end with a project duration near to the actual required duration for building construction projects in Bahrain. The gathered actual data and questionnaire survey data were used to evaluate the previous models for project buffers, one was selected as the best model to be used for two story villas in Bahrain. This model was used to develop a new model that will be effective, reduce cost, and increase profit of such projects in Bahrain. The results showed that a merged model of critical chain density and resource density is the best model that has produced a final project duration with a variance of 23.065 days from the actual project duration of 89.94 days obtained from the documented data. This research focused on projects of two story villas in Bahrain, having defined the specifications of these projects, the developed model can be used for projects with similar specifications.