Oveření využití nelineárních vicekriteriálních dekompozičních metod rozhodování v oblasti managementu podniku

Abstract

Doctoral dissertation is focused on the application of methods of multiple criteria decision-making management methods and techniques from the perspective of the scientific discipline. Particular attention is paid to the nonlinear decomposition multiple criteria decision making methods, which can be used to determine the preference allocation. The goal of the doctoral dissertation is to propose and verify application of advanced non-linear decomposition multiple criteria decision making methods on the business management level in the area of performance evaluation. The dissertation is divided into three main parts. The first part deals with description and explanation of decision making and decision making process principles in the managerial tasks with focus on multiple criteria methods. The second part of this thesis consists of applied methods and approaches description. This part deals with a literature review of selected weighting methods, multiple criteria decision making methods and their application on different managerial tasks. The outcome of the second part is an own concept of selection of appropriate MADM methods on particular managerial tasks and a generalized approach to their application. The third part of this thesis is dedicated to the verification of the created approach and selected MADM methods on performance evaluation of a manufacturing company. Firstly, an analysis of criteria structure is performed using DEMATEL method. Then based on the existing criteria structure and novel network structure based on results from DEMATEL subjective weighting is performed by a group of experts. Objective weighting using entropy and CRITIC methods. Then performance evaluation is performed using AHP, ANP, WINGS, TOPSIS, VIKOR and COPRAS-G using actual data from the company and expert based preferences. In the end results for performed evaluations are compared and discussed. The outcome of this doctoral thesis is the verification of a proposed approach of MADM application using advanced methods and as well as their hybrid versions. Doctoral thesis follows the outcomes of Student grant competition SP2013/173 and SP2014/126 projects.