Open AccessMethodological scheme for ranking interval expert estimates of the territories hydrocarbon potential
Author(s) -
Mykhailo Popov,
Олександр Зайцев,
Iryna Piestova
Publication year - 2019
Publication title -
ukraïnsʹkij žurnal distancìjnogo zonduvannâ zemlì
Language(s) - English
Resource type - Journals
ISSN - 2313-2132
DOI - 10.36023/ujrs.2019.22.156
Subject(s) - ranking (information retrieval) , interval (graph theory) , statistics , expert system , computer science , geologist , maxima and minima , mathematics , data mining , geology , artificial intelligence , paleontology , mathematical analysis , combinatorics
The problem of priorities establishing for expert interval-valued estimations when experts hold the opposite opinion is considered. The whole group of expert estimates is subdivided into subgroups, first of which provides the probability of the deposit presence, and the second one provides the probability of deposit missing. A ranking methodology for interval expert estimates of the territories’ hydrocarbon potential, consisting of two stages, is proposed. At the first stage, an estimates formed by two subgroups of experts are separately aggregated by optimization. Two aggregated interval estimates of the corresponding hypotheses probabilities are obtained as a result. In the second stage, a priority estimate is determined by comparing the results. A numerical example of the test territory evaluating for a hydrocarbon deposit presence was calculated. Interval-valued estimates by five experts were used in this example for the hypotheses of hydrocarbons presence/missing. Various metrics of the distance between interval values were used to calculate persistent minima of aggregating estimates. The results of the calculations indicate the hypothesis’ priority of a hydrocarbon deposit presence within the study area. The proposed methodology for ranking interval-valued expert estimates can be used in the “Geologist’s Computer Assistant” software system.The problem of priorities establishing for expert interval-valued estimations when experts hold the opposite opinion is considered. The whole group of expert estimates is subdivided into subgroups, first of which provides the probability of the deposit presence, and the second one provides the probability of deposit missing. A ranking methodology for interval expert estimates of the territories’ hydrocarbon potential, consisting of two stages, is proposed. At the first stage, an estimates formed by two subgroups of experts are separately aggregated by optimization. Two aggregated interval estimates of the corresponding hypotheses probabilities are obtained as a result. In the second stage, a priority estimate is determined by comparing the results. A numerical example of the test territory evaluating for a hydrocarbon deposit presence was calculated. Interval-valued estimates by five experts were used in this example for the hypotheses of hydrocarbons presence/missing. Various metrics of the distance between interval values were used to calculate persistent minima of aggregating estimates. The results of the calculations indicate the hypothesis’ priority of a hydrocarbon deposit presence within the study area. The proposed methodology for ranking interval-valued expert estimates can be used in the “Geologist’s Computer Assistant” software system.