Information and Analytical System for Urban Public Transport Using Artificial Intelligence Tools
Students Name: Hrechanyi Viacheslav
Qualification Level: magister
Speciality: Information Control Systems and Technologies
Institute: Institute of Computer Science and Information Technologies
Mode of Study: full
Academic Year: 2025-2026 н.р.
Language of Defence: англійська
Abstract: Relevance. Modern cities face increasing transportation demand, complex transport infrastructures, and growing requirements for sustainability, safety, and efficiency. Traditional transport planning approaches are often static and unable to adapt to dynamic urban mobility patterns. Therefore, the application of artificial intelligence (AI) methods for analyzing, forecasting, and optimizing urban transport systems is a highly relevant research direction within intelligent information systems and smart city solutions. Object of research – processes of operation and planning of urban transport systems. Subject of research – models, algorithms, and software tools for decision support in urban transport planning using machine learning and deep learning methods. Purpose of research – to develop a prototype decision support information system for planning and optimizing an urban transport system based on artificial intelligence methods. Structure of the thesis. The qualification thesis consists of an introduction, four chapters, conclusions, a list of references, and appendices. The first chapter presents the characteristics of the research object, a literature review, system analysis of the urban transport system, and the conceptual model of the system. The second chapter describes the system architecture, algorithmic and mathematical support, and data structures. The third chapter focuses on software development, including the description of program modules, used libraries, and implementation of the system prototype. The fourth chapter presents experimental studies, testing results, performance evaluation, and analysis of obtained outcomes. Methods and tools. The research applies time-series analysis, graph-based transport network models, recurrent neural networks (LSTM), route optimization algorithms, and deep learning frameworks such as TensorFlow and PyTorch. The system prototype is implemented in Python using NumPy, Pandas, NetworkX, and Matplotlib libraries. Results and practical value. A conceptual and software model of a decision support system for urban transport planning has been developed. A functional prototype capable of forecasting traffic flows, dynamically adjusting transport network weights, and generating recommended public transport routes has been implemented. Experimental results confirm the effectiveness of applying artificial intelligence methods to improve urban transport planning. The obtained results can be used by municipal authorities, transport planners, and for further scientific research. The total volume of the thesis is 104 pages, including 92 pages of the main text, 24 figures, and 1 table. The list of references contains 35 items. Keywords: urban transport system, artificial intelligence, transport planning, decision support system, machine learning, deep learning, Python.