The distributed systems of engineering supervision of permanent structures

Abstract

This article explores the benefits of engineering analysis distributed systems for the purposes of analysis of depreciation and restoration cost of permanent buildings and structures. We propose the methodology for the study of complex automated systems of industrial and residential facilities life cycle control which is based on an integrated approach. We also propose the methodology for using the benefits of decomposition and proposed approach to analyze the events occurring during the production process using machine learning. The brief review of methods of calculation of depreciation of construction objects is presented. The choice of task examination of technical condition of a building or structure is indicated. Investigated the importance of the comprehensive analysis of geometrical parameters, instrumental studies, the determination of the actual characteristics of materials of the basic bearing structures and their elements, the measurement of the operational environment and operating loads, etc. Considered the models implementing machine learning for the classification of life cycle events. Descriptors of the internal production network data were used for training and testing of applied models. k-Nearest Neighbors and Random forest methods were used to illustrate and analyze proposed solution. The proposed methods allow to analyze causes of defects and damages in structures. The result of this work showed the possibility of successful application of the statistical approach for data analysis in construction, demonstrate the effectiveness of the implementation of such systems. The result makes possible to detect problems at an early stage and simplify the task of life cycle management of buildings and structures.