The more complex a production process is, the more difficult it is to estimate accurately the production tasks times. The Project Evaluation and Review Technique (PERT) is one of the most frequently used methods for evaluating the project lead times (also sets of production tasks). Various variants of the method are used depending on the estimation accuracy degree and available computing power. The basic method for determination of the critical path duration is called the classical PERT method. One of its best advantages is a relatively low-level calculation complexity. The disadvantage, on the other hand, is a relatively low estimation accuracy resulting from many generalizations. The literature on the subject matter describes two group of algorithms used to improve the critical path duration estimates relative to the classical PERT method. The first group refers to the estimation improvement of variance and standard deviation of random variables that define the tasks times. The other group focuses on the fit improvement of the probability distribution of random variable. Each method has its advantages resulting from the better accuracy in comparison with the classical PERT, but the calculation complexity grows along with the increasing estimation accuracy. The paper aims at developing the estimation method for the critical path duration based on a modified PERT method. The presented method of determining the cumulative execution time of the sets of tasks accounts for skewness of empirical distributions which frequently occurs in the processes. The studies verifying the method’s soundness have been made. The results have been compared with the values obtained with the use of the classical method. Then, the lead times have been determined with the use of the developed method for a real production facility.Keywords: PERT and beta-PERT distribution, critical path
© This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.