Article ID Journal Published Year Pages File Type
4942777 Engineering Applications of Artificial Intelligence 2017 13 Pages PDF
Abstract
By considering target values for attributes in addition to beneficial and non-beneficial attributes, a traditional MADM technique is converted to a comprehensive form. In many machine selection problems, some attributes have given target values. The target value regarding a machine attribute can be reported as a range of data. Some target-based decision-making methods have recently been developed; however, a research gap exists in the area. For example, fuzzy axiomatic design approach presents a target-based decision-making supported on common area of membership functions of alternative ratings and target values of attributes. However, it has detects on finding a complete ranking because of probable infinite values of assessment index. Two target-based VIKOR models with interval data exist in the literature; however, the target values of attributes or ratings of alternatives on attributes are crisp numbers in the models and their formulations may have some limitations. The present paper tries to fill the gap by developing the VIKOR method with both interval target values of attributes and interval ratings of alternatives on attributes. Moreover, we attempt to utilize the power of interval computations to minimize degeneration of uncertain information. In this regard, we employ interval arithmetic and introduce a new normalization technique based on interval distance of interval numbers. We use a preference matrix to determine extremum and rank interval numbers. Two machine selection problems concerning punching equipment and continuous fluid bed tea dryer are solved employing the proposed method. Preference-degree-based ranking lists are formed by calculating the relative degrees of preference for the arranged assessment values of the candidate machines. The resultant rankings for the problems are compared with the results of fuzzy axiomatic design approach and the interval target-based MULTIMOORA method and its subordinate parts.
Related Topics
Physical Sciences and Engineering Computer Science Artificial Intelligence
Authors
, ,