Reliability analysis of software for automatic monitoring of population movement databases

The approach and results of the reliability analysis of the program for collecting data on population movement in the form of city formation are presented. A brief description of this program, which operates with monitoring of open databases of Ukrainian Wikipedia and the OpenStreetMap software application, is provided. The program was developed by use of the Python programming language as well as the Requests, BeautifulSoup, Pandas libraries, which provides effective work with HTML content of web pages, geodata processing, and structured information storage. The development, testing, and visualization of the process were performed using Jupyter Notebook. The data on the detection of the number of defects obtained during the development of the software tool, depending on the testing time interval, were considered. The Jelinski-Moranda model was used for computational analysis, and its mathematical description was provided. The model parameters were determined by solving a system of nonlinear equations by the iterative method. The obtained calculated dependences on the testing time of the failure intensity and the reliability function of the data collection program are presented. The obtained data of the preliminary computational modeling of reliability were compared with the data obtained during testing of the program under consideration. The time for ensuring the required reliability of the software tool was determined. It was shown that the use of the described approach and the Jelinski-Moranda model provides a satisfactory degree of description of the behavior of the software tool during testing, which allows determining the time for ensuring the required reliability of its operation. It was concluded that the use of the presented approach and the results obtained at the same time allow us to consider its application possible in the design of similar software tools dedicated to the analysis of database fields.