This research investigates the development of expert systems (ES) based on large language models (LLMs) enhanced with augmented generation techniques. The study focuses on integrating LLMs into ES architectures to enhance decision-making processes. The growing influence of LLMs in AI has opened new possibilities for expert systems. Traditional ES require extensive development of knowledge bases and inference algorithms, while LLMs offer advanced dialogue capabilities and efficient data processing. However, their reliability in specialized domains remains a challenge. The research proposes an approach combining LLMs with augmented generation, where the model utilizes external knowledge bases for specialized responses. The ES architecture is based on LLM agents implementing production rules and uncertainty handling through confidence coefficients. A specialized prompt manages system-user interaction and knowledge processing. The architecture includes agents for situation analysis, knowledge management, and decision-making, implementing multi-step inference chains. Experimental validation using YandexGPT 5 Pro demonstrates the system’s capability to perform core ES functions: user interaction, rule application, and decision generation. Combining LLMs with structured knowledge representation enhances ES performance significantly. The findings contribute to creating more efficient ES by leveraging LLM capabilities with formalized knowledge management and decision-making algorithms.

Keywords: large language model, expert system, artificial intelligence, decision support, knowledge representation, prompt engineering, uncertainty handling, decision-making algorithms, knowledge management

Oil spills require timely measures to eliminate the causes and neutralize the consequences. The use of a case-based reasoning is promising to develop specific technological solutions in order to eliminate oil spills. It becomes important to structure the description of possible situations and the formation of a representation of solutions. In this paper, the results of these tasks are presented. A structure is proposed for representing situations in oil product spills based on a situation tree, a description of the algorithm for situational decision-making using this structure is given, parameters for describing situations in oil product spills and presenting solutions are proposed. The situation tree allows you to form a representation of situations based on the analysis of various source information. This approach makes it possible to quickly clarify the parameters and select similar situations from the knowledge base, the solutions of which can be used in the current undesirable situation.

Keywords: case-based reasoning; decision making; oil spill, oil spill response, decision support, situation tree

The article presents the creation of a software platform for automatic configuration of ground facilities systems using ontological models, intelligent algorithms for selecting objects and generating options. The mathematical formulation of the configuration problem, the architecture of the application and the user interface for entering design data and obtaining the result of selecting objects with a high degree of detail are given. In the future, the developed prototype can become a tool to support the work of system engineers and technologists at the stage of conceptual design during the variable study of the configuration of the oil and gas system, taking into account the requirements.

Keywords: conceptual design, field development, systems engineering, ontological engineering, ontology of the oil and gas system, function-oriented ontology, autoconfiguration of the oil and gas system, formalization of the system configuration procedure

The possibility of using neural networks to classify the states of complex objects is considered. A software implementation of a neural network classifier is made and the results are compared with a selection algorithm based on K-nearest neighbors.

Keywords: classification, neural networks, decision making, case

The paper considers the implementation of the Case Base Reasoning method when managing a city infrastructure complex technological object. A method of formalizing the situation at a complex technological object using a matrix of states is presented. A method is proposed for determining the proximity of situations in the state space, taking into account the proximity of the states of each element of a complex technological object. The implementation of the proposed method when selecting a situation from the knowledge base allows avoiding collisions when two critically different situations with different solutions are selected.

Keywords: Case Based Reasoning, CBR, case retrieve, knowledge base, facility management, city infrastructure, Euclidean distance