Search academic texts

Information × Registration Number 0212U007615, 0111U005719 , R & D reports Title Duration estimate of software projects based on non-gaussian stochastic models popup.stage_title Head Prikhodko Sergii Borisovich, Доктор технічних наук Registration Date 03-07-2012 Organization Departament Software for Automated Sysrems National University of Shipbuilding named after admiral Makarov popup.description2 Object of research -project duration estimation process in time management in software projects. A research objective - improving the reliability and accuracy of duration estimation in time management in software projects. Research methods - methods of probability theory, the mathematical statistics, mathematical modeling, the interval analysis. Scientific novelty of the research results: non-gaussian probability model of duration of software projects based on of Johnson's normalizing transformation has had the further development that enables to consider real distribution of duration of software projects and to estimate confidential intervals of the statistical moments of duration; the method for constructing nonlinear regression equations based on normalizing transformations has had the further development that enables to get the equation of nonlinear regression and to estimate confidence intervals of nonlinear regression; non-linear regression model of duration of software projects depending on efforts based on application of Johnson's normalizing transformation has had the further development that enables to estimate confidential intervals of non-linear regression. Non-linear dependencies of these confidential intervals have been calculated. Product Description popup.authors Пухалевич А. popup.nrat_date 2020-04-02 Close