Страница публикации

A Module for Industrial Safety Inspection Planning Based on Self-organization

Авторы: Berman A., Nikolaychuk O., Pavlov A., Yurin A.

Журнал: Lecture Notes in Computer Science: 19th Russian Conf. on Artificial Intelligence (RCAI 2021, Taganrog 11-16 October 2021)

Том: 12948

Номер:

Год: 2021

Отчётный год: 2021

Издательство:

Местоположение издательства:

URL:

Аннотация: The paper considers the application of the principles of self-organization for the automation of the business process of industrial safety inspection (ISI). Self-organization refers to the process of ordering elements (components) in an information system due to internal factors, without specific external influence. Self-organization is parametric and structural in nature. The architecture of the system as a whole and the main aspects of the main module, namely, scheduler, which provides self-organization of an open multicomponent information system of the ISI, are described. An ontological model of the object of expertise, the task, the method, and the operation that implements it, as the basis of self-organization, is proposed. The model and algorithms for implementing the scheduler are detailed. In particular, the algorithms of the main operations are described: the formation of the task description, the methodology for solving the problem, and the “intelligent” task execution. It also lists the principles of formation and describes the main local rules of the knowledge base of the scheduler, which are responsible for describing the methodology, the decision process, the coordination of expert opinions, and self-learning. A conceptual description of the implementation of the scheduler based on the components of the software platform is given. Some results of application of the considered approach for the problem of technical diagnostics of ISI are given: fragments of the generated software, forms of the user interface demonstrating the description of the object of expertise, local rules, methods and results of self-organization.

Индексируется WOS: 1

Индексируется Scopus: 1

Индексируется РИНЦ: 1

Публикация в печати: 0

Добавил в систему: