New fuzzy GERT method for research projects scheduling

The aim of This work is to present a new applicable and suitable technique for research projects scheduling. Research projects have special characteristics that separate them from other projects. These include uncertainty in definition, sequence and duration time of activities; uncertain results; having loops and cycles; and no repetitive activities, so there is a need for special tools and techniques for scheduling these projects. Conventional scheduling techniques are not able to model research projects or have a lot of errors. Fuzzy GERT networks using capabilities of GERT and fuzzy, is powerful tools for scheduling, Iran research projects. This work presents a new method for solving fuzzy GERT networks. In this method, activity duration time and loops trapezoidal fuzzy numbers states repetition number, and output activities from nodes of network belong to a fuzzy set. In this method, the First time GERT networks computations are done based on evaluating nodes. Process outputs are scheduled network and project fuzzy completion time. These outputs are fuzzy numbers and can be analyzed by /spl alpha/ cuts. This method is more applicable and needs less computation than existing fuzzy and probability GERT methods. Besides, this method can get computerized and up to date easily. This method can be also applied for cost estimations in research projects. For checking validation, we compared this method with existence fuzzy and probability GERT methods, the results were very near to each other. We also applied it in an electronic institute. The results were more realistic than the results of their pervious method (CPM).