This paper proposes an improved critical chain method with fuzzy approach to schedule projects under uncertainty. Utilizing fuzzy numbers can be wise because of the natural uncertainty in estimation of required work to accomplish the project tasks. To minimize the project duration as an objective taking into account resource limitations and other constrains, a resource-constrained project scheduling problem (RCPSP) model is adopted. The output of this model determines the project schedule. Then, a new buffer sizing approach based on the square root of the sum of the squares (SSQ) method is introduced which is modified by coefficients to overcome the schedule risk more efficiently. As a result, a project buffer is added to the end of the project schedule and is used if necessary. The methodology can be used as a control procedure.