MANAGING THE RESTRUCTURING OF INFORMATION AND COMPUTING PLATFORMS OF EVOLVING CYBERPHYSICAL SYSTEMS UNDER CONDITIONS OF UNCERTAINTY

Author:

Eremenko V. T.,Loginov I. V.,Fisun A. P.,Rytov M. Yu.

Abstract

The paper considers the problem of managing the computational platform reconfiguration of evolving cyber-physical systems. The urgency of the task is determined by a significant changing increasing of the cyber-physical systems’ using conditions (traffic management, logistics flows, distributed production), leading to a change in the requirements of the destination. The intensive of change for innovation cyber-physical can be more than 0.5-1.0 by year, that require significant resource for reconfiguration and modernization. A significant factor in managing the structural and functional composition of informational computing modules is a high degree of uncertainty in the indicators of the such modules using at the early stages of the decision-making cycle. Within the framework of the study, an approach to managing the reconfiguration of informational computing platforms based on the mechanism of maximum functional suitability of a set of informational computing modules is considered. The effectiveness of Shortest Job the Next (SJN), Dynamic Priorities (DPR) and Static Priorities (SPR) planning mechanism are evaluated in the case of uncertainty (the error coefficient of variation varies in the range from 0 to 1.75). In the case of zero error – DPR mechanism is more effective than SPR, that is more effective than SJN. Based on the analysis of the effectiveness of a set of methods for managing reconfiguration with variations in the error of estimating the parameters of the upgraded information and computing modules, a context-dependent MAX management model is obtained in the article. MAX management model is a combination of approaches based on static and dynamic priorities: if the value of the coefficient of variation more 0.75 than MAX = SPR else DPR. The use of such a management model makes it possible to increase the integral efficiency of the target cyber-physical system.

Publisher

Izdatel'skii dom Spektr, LLC

Subject

General Materials Science

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3