Departmental pre-defense

Over the last three years, my life has been closely tied to research aimed at solving one of the most pressing problems in the digitalization of agriculture: processing and analyzing information under conditions of high uncertainty. On June 24, 2025, I successfully completed the pre-defense of my Candidate of Sciences dissertation at the department, and now I would like to share the main results of this work, talk about the path I took, and express my sincere gratitude to everyone who supported me.

* What is my dissertation about?

The title of my dissertation is “Development of Algorithms for Information Analysis and Processing in the Design of Platform Systems Under High Uncertainty.” The work was completed in specialty 2.3.1-systems analysis, control, and information processing.

The goal of the research was to develop adaptive information-processing algorithms that can be integrated into digital platforms for the analysis of data coming from agro-industrial information systems. Under conditions where the volume of information is growing and its structure is becoming increasingly heterogeneous, it is especially important to present data in a form suitable for making timely and correct management decisions. That is why the study proposed the bitwise fuzzification method and the SFI (Stukalin Fuzzy Index)-a universal way of representing fuzzy attributes of objects as a binary vector. This makes it possible to simplify analysis, identify patterns, and generate generalized assessments understandable both to a specialist and to a system.

The proposed algorithms can be applied not only in the agro-industrial sector, but in any industry that has to deal with incomplete, contradictory, or weakly structured data.

* How does it work?

One of the central elements of the research was the method of dynamic fuzzification-the transformation of numerical data into qualitative categories adapted to context: region, crop, characteristics of the object. For example, a milk yield of 12.5 liters in northern regions may be a high indicator, while in central Russia it may be low. Using adaptive membership functions, the system itself “understands” what should be considered the norm in each specific case.

The algorithms I developed were tested on livestock breeding data and demonstrated high efficiency in assessing the breeding value of cattle. A digital solution architecture was created that can scale and integrate with existing agricultural information systems such as FGIAS PR and EFGIS ZSN.

* A three-year journey

Work on the dissertation proceeded in parallel with my work as an information systems architect, which allowed me not merely to model processes, but also to test the algorithms in real projects.

I published 14 scientific papers and prepared analytical publications on digital agriculture, digital learning, and digital profiling of agricultural enterprises.

* Words of gratitude

I would like to express special gratitude to my academic advisor, Candidate of Political Sciences and Professor Oleg Alekseyevich Motorin. His support, his constant prodding, and his understanding of the essence of agro-industrial problems motivated me and pushed me to work throughout all three years of study.

I thank all the faculty members of the department for their attention to detail, for their criticism, for their motivation, and for always being ready to help. It was a pleasure to hear the department head say that she considers my work interesting and unusual. I value the comments and recommendations that were made-they have already formed the basis for revising the text before the defense.