СLASSIFICATION AND FEATURES OF DATA COLLECTION IN ADMINISTRATIVE DECISION SUPPORT SYSTEMS

Abstract

Decision support systems are the development of computer management decision support systems (DDS) using the principle of proactive computing, i.e. transferring human actions to a higher level of management and using algorithms of intelligent data processing and machine learning. Taking preventive decisions can lead to cost savings in the management of processes, in particular, the management of transport infrastructure based on the analysis of large volumes of heterogeneous data. In the concept of proactive systems, there is a scheme: detect, predict, make decisions, act, and the principles of building an EMS, taking into account the processing of events. When implementing transport infrastructure management, a critical task is the task of efficient processing of various data received from various sources. The quality and timeliness of data affects the efficiency and effectiveness of decision-making. In the modern discussion there is the issue of developing approaches to effective data collection and processing, finding new ways of storage and preliminary analysis. The exponential growth of the volume of data and the increase of the network bandwidth provided for data transmission opens up new opportunities for managing the transport infrastructure, but at the same time there are problems of efficient data processing. The statistics provide a classification of data based on various criteria. We have seen the following types of data that are analyzed in proactive systems to support decision making: sensors data, log data, visual data – images or videos, textual data, social data alnikh merezh (social data) geospatial data. We have analyzed the features and problems of collecting and collecting various data from proactive systems to support the adoption of management decisions. The following features were observed: the need for special data formats, the need to ensure data consistency; the need to minimize human participation in the process of collecting diverse data; ensuring the collection of data in real time.