Herald of Advanced Information Technology 2021; Vol.4 No.3: 268−277 268 Information Technology in Socio-Economic, Organizational and Technical Systems ISSN 2663-0176 (Print) ISSN 2663-7731 (Online) DOI: https://doi.org/10.15276/hait.03.2021.6 UDC 004.891 The multilayer distributed intelligence system model for emergency area scanning Andrey O. Tsariuk 1) ORCID: https://orcid.org/0000-0001-5657-8977; tsariukandreyjs@gmail.com Eugene V. Malakhov 1) ORCID: https://orcid.org/0000-0002-9314-6062; eugene.malakhov@onu.edu.ua. Scopus ID: 56905389000 1) Odessa I.I. Mechnikov National University”, 2, Dvoryanskaya Str. Odessa, 65082, Ukraine ABSTRACT Emergency situations have a huge impact on various important areas of human life. Every year there are many situations, the elimination of which requires a lot of financial and human resources. Therefore, the ability to reduce the impact of the consequences and increase the speed of their elimination is extremely important. In this article, a multi-level model of a system was proposed that provides support for performing operational tasks in emergency situations in open areas. The most important elements, areas of their responsibility, and interconnection were identified and described in architectural style. The idea of the work is to design a system that should use Swarm intelligence under the hood to provide continuous support in emergency situations. The system consists of 4 main parts: Cloud, Swarm, Swarm operator, and Swarm Node. The Cloud (Swarm Wamb) is the main decision-maker that provides ETL data pipelines and operates under strategically tasks. In accordance with the idea, Swarm womb should be a cloud service-like system with the ability to scale over the world. The Swarm is a combined set of multiple Swarm Nodes and only one Swarm Operator. The main task of the Swarm is to provide support in local operational tasks where SN is responsible for the execution and SO is for control. Rescue and search operation after any natural disaster is a target to show the system’s purpose. In practice, the cl oud system (Swarm Wamb) receives requests to perform an operation, calculates resources effort first, and delegates a task to the Swarm. When the swarm reaches the location, it starts executing. Operator with nodes tries to find survivors and collect as much important information as they can. Video, images, recognized objects are continuously sending to the Cloud for additional analysis in real-time. Any information in an emergency situation can help save more humans lives and reduce risks. In this article, the multilayer distributed intelligence system architecture for emergency area scanning was designed and described. The set of terminology was proposed as well. This architecture covers different levels of tactical and operational tasks. Keywords: Expert systems; system architecture; system model; emergency situations; swarm; cloud computing For citation: Tsariuk A. O., Malakhov E. V. The multilayer distributed intelligence system model for emergency area scanning. Herald of Advanced Information Technology. 2021; Vol. 4 No. 3: 268–277. DOI: https://doi.org/10.15276/hait.03.2021.6 INTRODUCTION Every year many natural emergencies happen in the world with unpredictable consequences. According to statistics, in 2019 only, 361 cases of natural origin on a global scale were recorded (Fig. 1), and it is impossible to calculate local incidents of technogenic and biological-social origin. The result of such incidents always leads to large-scale, costly material and physical operations aimed at eliminating the consequences and developing means of preventive action. At the moment, the level of technology development provides many opportunities for the delegation of complex and dangerous work to mechanical means, such as manipulators, robots, and drones. Auto factories have long used industrial manipulators for welding and painting, robots are used to distribute goods in warehouses, and technology using AI approaches has flooded stores. Collaboration between man and machine is no longer new and has established itself as a successful combination in many spheres of life [1]. Due to technological progress, robotics is already becoming available for widespread use. Toy robots and aerial photography drones have already become mainstream commodities and are powerful and cheap to manufacture. Companies such as Amazon have already paid attention to the prospects of using this tool for solving problems in the delivery of goods [2]. Also, many developments are underway in the following directions: remote observation, research of objects and territories, drawing up maps of the area [3]. Cloud technologies and solutions are another powerful and widely used tool now. Many projects and large corporations are using a cloud computing and scaling approach to improve reliability and performance and reduce costs. Due to the distribution and fault tolerance, cloud solutions, for example, Amazon, can guarantee the SLA of their services up to 99.99 % [4] and using combined approaches, that allows creating a system in afault- tolerancemanner. © Tsariuk A., Malakhov E., 2021 This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/deed.uk)