Published on: 2026-05-19
Source: Saint Petersburg State Polytechnic University of Peter the Great –
An important disclaimer is at the bottom of this article.
The digital platform for multimodal data analysis “Polanis,” developed at the Saint Petersburg Polytechnic University of Peter the Great, was included in the Unified Register of Russian Software of the Ministry of Digital Development, Communications and Mass Communications of the Russian Federation. The platform greatly increases computational and analytical capabilities in the fields of industry and science.
The Political Analytical Information System “Polanis” is a comprehensive digital solution for management and analysis in the most complex technical and social systems. The platform is indispensable for effective coordination in industries where traditional data analysis methods, including digital ones, can no longer cope with the exponentially increasing flow of information: for example, in the oil and gas and chemical industries, energy, transport, healthcare, as well as in interdisciplinary research.
The main feature of our platform and its difference from analogues is the creation of a unified integrated work environment. The user gets an end-to-end toolkit that allows uploading the data almost in any format and receiving analytics in the form of a convenient interactive system. For the most complex calculations, the platform uses one of the most powerful supercomputer centers in the country, “Polytechnic.” In the field of science, this opens a new level of research by operating volumes of information previously inaccessible.Inclusion of the “Polanis” platform in the Russian Ministry of Digital Development’s software registry is recognition that the university creates its own high-tech products of a high standard that can be used on a nationwide scale, noted Yuri Fomin, Vice-Rector for Research at SPbPU.
The use of “Polanis” for optimizing traffic flows provides acceleration of calculations by up to 20 times with a controlled accuracy reduction of no more than 5%. In the task of processing seismic data, the neural network modules of the platform reduce processing time by 70%, and the prediction of geophysical reservoir characteristics achieves an accuracy exceeding 85%. In the scientific field, the “Polanis” toolkit speeds up work on new technologies and materials. Using the platform, it is possible to optimize complex production cycles, manage quality, and predict energy consumption.So, one of the software modules of the platform allowed to create a digital model of the CHP and analyze equipment wear taking into account real, not just planned, data, which allows for proactively changing parts to ensure stable operation.
Neural networks that have already become familiar are not capable of effectively processing colossal volumes of diverse information from industrial systems, nor of carrying out systematization and deep intellectual processing. “Polanis,” on the other hand, combining in its architecture technologies of hybrid artificial intelligence that integrate imitation modeling, machine learning, neural networks, and multi-agent systems, allows industry to reduce costs, improve product quality, accelerate innovation, and make more balanced decisions,” noted the project leader, head of the laboratory.”Industrial Data Stream Processing Systems” of the Advanced Engineering School “Digital Engineering” SPbPU Marina Bolsunovskaya.
The scale of the work done is reflected in the number of registered results of intellectual activity. In 2025, the team registered seven programs for computers, and in January 2026, another 15 new software modules were registered in the Rospatent registry. The total number of registered programs for the project reached 22, each contributing to the architecture and functional capabilities of the platform.
The project team is currently working on the implementation of a platform of large language models, further trained and fine-tuned on domain-specific data, as well as a module for automated machine learning for intelligent selection and hyperparameter tuning of algorithms for the specifics of the data. Scientists also plan to create and patent systems for efficient and predictable work with heterogeneous data, such as geodata, telemetry, and images simultaneously. According to the developers, this will become the foundation for industry standards in the target sectors.
Work is carried out with the support of the “Priority 2030” program of the Ministry of Science and Higher Education of the Russian Federation.
Please note; this information is raw content obtained directly from the source of information. It represents an accurate report of what the source claims and does not necessarily reflect the position of MIL-OSI or its clients.