Symonov D. Mathematical methods of integrated supply chains

The dissertation is devoted to the exploration of mathematical methods in integrated supply chains. The primary objective of the dissertation is the theoretical and methodological justification for the applicability of mathematical modeling in processes within integrated supply chains. The relevance of integrated supply chains in today's business environment is paramount due to ongoing transformations in global trade and manufacturing sectors. Optimizing integrated supply chains has become a strategic imperative for organizations, especially in the context of selecting locations for establishing enterprises and service centers. Global market globalization necessitates meticulous consideration of geographical aspects by firms, taking into account various factors such as labor costs, tax policies, resource availability, and the expectations of potential consumers. The study of integrated supply chains transcends the conventional approach to supply chains and becomes a strategic necessity for the successful operation of contemporary business models. The challenge lies in the prevalent use of measures originating predominantly from operations management within the supply chain management domain. These measures typically focus solely on individual companies and overlook external environmental aspects. Consequently, supply chain management measures addressing these concerns are essential. The relevance and insufficient exploration of these issues determined the selection of the dissertation's topic, objectives, and tasks, as well as its theoretical and methodological foundation. The scientific novelty of this work encompasses several theoretical and practical outcomes: a novel mathematical model for the supply of goods or services within a specific supply chain, accounting for inventory levels arising from system inertia; a new algorithm for constructing a fuzzy cognitive map simulating the investigated system; an equivalent solution method for a multi-criteria problem concerning optimal location selection, leveraging pseudo-Boolean transformations for polynomial time solution; a new approach for constructing a morphological control block using machine learning algorithms; and an adapted iterative method for determining optimal supply chain structures considering the capacities and throughputs of individual chain links, providing insights into resource optimization directions; modifications to the "Game of Life" enabled the development of an algorithm for modeling consumer migration risks and the resilience of locations to equilibrium states. Another adapted iterative method was proposed for determining optimal supply chain structures, and an algorithm was introduced for determining the optimal order batch size and formulating an optimal supply plan under stochastic demand conditions. The practical significance of the obtained results lies in their potential efficacy for addressing optimization challenges in supply chain operations requiring the processing of large datasets. This includes tasks related to investment planning for new enterprises, risk assessment for investment returns, logistics flow planning, resource parameter determination, and relationship management programs with consumers and partners within supply chains. The work was conducted within the framework of research topics of V.M. Glushkov Institute of Cybernetics of the National Academy of Sciences (NAS) of Ukraine, specifically under the state registration numbers: 0118U001111 for the years 2020-2022, 0121U112884 for the year 2021, and 0123U102942 for the year 2023.