دانلود رایگان مقاله انگلیسی روش تصمیم گیری چند منظوره طبقه بندی شده - الزویر 2018

Abstract:

Multiple Criteria Decision Making (MCDM) methods generally require the decision maker to evaluate alternatives with respect to decision criteria and also to assign importance weightings to the criteria. Then, based on the assigned weightings, the best alternative can be selected. However, after a decision is made it often happens that the decision maker becomes doubtful whether the right weightings have been assigned to the criteria given that a variety of eventualities may occur in the near future. The main aim of this paper is to address this concern and improve the application of MCDM methods by addressing possible fluctuations in the criteria weightings. The recently proposed concept of stratification (CST) is used in conjunction with MCDM methods to stratify the decision environment. The method is then applied to a supplier selection problem. The stratified MCDM (SMCDM) approach is in its early stages only and requires further research to reach its maturity.

6. Concluding remarks

CST is a recent and an innovative approach in problem solving which considers a number of states, each with its own inputs and outputs. The main purpose of this study was to take into account the dynamicity of an environment in which multiple criteria decisions are made. MCDM methods are capable of considering several criteria when making a decision. A typical challenge while applying MCDM methods is the possibility that the conditions under which a decision is being made may change. Such a change affects the weightings of criteria and hence may result in a different selection of alternatives. This paper applied the SMCDM approach and showed how the impact of such incidents can be taken into account so that smarter decisions can be made. The combined method reduces the doubt about making the best decision. This is because the method is able to cover various situations that have an impact on a decision. In other words, the environment in which a decision is made is considered under different conditions with respective probabilities. Then, the weightings of the criteria are computed with respect to each likely condition. These computed weightings of the criteria contribute to the final weightings, taking into account the likeliness of each condition happening. This approach was developed following an examination of the process of decision-making in the human brain and it is expected that it would be employed to in future studies improve decision-making in artificial intelligence. Future studies could also apply CST in combination with a variety of well-known MCDM methods and develop SMCDM methods, such as SAHP, SANP, SBWM and similar, to address various multi-criteria decision-making problems.