Development of the method of distances for processing expert estimates in information systems
Journal Title: Восточно-Европейский журнал передовых технологий - Year 2019, Vol 2, Issue 3
Abstract
<p>Current research focuses on expert estimation of indicators by employing a limited number of qualified experts, which involves considerable time and cost. The advent of modern information technology allows rapid and efficient assessment of indicators that characterize performance of enterprises. The main tool for solving the issue related to the credibility of such an estimation is the development of new methods for processing its results.</p><p>We have proposed a method of distances to treat the results from expert estimations of indicators and examined the results of its work.</p><p>To this end, theoretical substantiation of the method has been performed, based on the concept of proximity (distances) among estimates relative to the average value, coefficients of experts' competence, and normalization of the point scale of assessment.</p><p>We have explored three variants of expert point- and verbal-based estimates of independent and dependent indicators in line with the method of distances, which showed that the convergence rate of iteration process for all cases varies from 1 to 4 orders of magnitude at each iteration. That makes it possible to draw a conclusion about a very small number of computations by the information system and sufficient speed of processing expert estimates.</p>A comparative analysis of the method of distances with a similar method of square deviations has revealed almost the same rate of convergence up to 3‒5 iterative steps, but the proposed method yields an estimate close to the average estimate of each indicator. It also enables the processing of results in information systems by 99 % faster, from dozens, even hundreds of times, more estimates than that in the methods of expert selection, which is important under current competitive environment. Average efficiency of the method compared to the method of expert selection is 5.8 %
Authors and Affiliations
Tetyana Otradska, Natalia Shybaieva, Oleksandr Shyrshkov, Mykola Rudnichenko, Ashot Halustian
Keywords
Related Articles
- EP ID EP666292
- DOI 10.15587/1729-4061.2019.163964
- Views 80
- Downloads 0
Development of dimensionally stable structures of multilayer pipelines and cylindrical pressure vessels from carbon fiber reinforced plastic
<p>In the framework of the momentless theory of cylindrical thin shells, the elastic deformation of multilayer pipes and pressure vessels is investigated. It is assumed that the pipes and pressure vessels are made by two...
Development of a model for the estimation of financial processes in logistic systems at industrial enterprises
<p>The model and the method for assessment of the effectiveness of management of financial processes in logistic systems of industrial enterprises were substantiated. The model takes into consideration the parameters of...
Development of techniques to predict and prevent both the effect of xenobiotics and their migration into pig-derived products
<p>This paper reports research into the influence of heavy metals, specifically cadmium and plumbum, on the body of young pigs, and the effectiveness of the preparation to prevent intoxication with heavy metals. Heavy me...
Development of the algorithm of video image adaptation to spectral power distribution of illuminants
Proposals for further progress of video technologies, issues that need to be resolved to implement this progress and possible ways to implement them in real devices of special and general application are made. It is prop...
Optimized smoothing of discrete models of the implicitly defined geometrical objects' surfaces
<p>When using many modern methods of automatic generation of surface meshes of implicitly defined geometric objects, the accuracy of approximation in the vicinity of surface singularities (holes, breaks, etc.) is lost. T...