Premium
Distance‐Based Measures of Specialization and Balance in Multi‐Criteria: A DEA‐Integrated Method
Author(s) -
Dyckhoff Harald,
Mbock Eleazar,
Gutgesell Sebastian
Publication year - 2014
Publication title -
journal of multi‐criteria decision analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.462
H-Index - 47
eISSN - 1099-1360
pISSN - 1057-9214
DOI - 10.1002/mcda.1532
Subject(s) - data envelopment analysis , complement (music) , measure (data warehouse) , balance (ability) , relation (database) , regular polygon , simple (philosophy) , computer science , point (geometry) , cone (formal languages) , mathematical optimization , operations research , econometrics , mathematics , data mining , algorithm , medicine , biochemistry , chemistry , philosophy , geometry , epistemology , complementation , physical medicine and rehabilitation , gene , phenotype
Decision makers pursuing several distinct goals may differ in their behaviour. Some try to achieve all goals simultaneously, whereas others concentrate on only certain ones and are therefore specialized. We present a new method of measuring the specialization degree with respect to a given standard balanced relation of the goals. The measurement is integrated into data envelopment analysis (DEA) methodology, uses a non‐compensative distance measure, permits optional convex polyhedral cones as predefined balance sets and guarantees feasible solutions. Geometrically, it is based on the angle between the balance cone and the ray through the point of investigation. The proposed models can be used to support and complement the usual effectiveness and efficiency analysis in multi‐criteria decision analysis and DEA. They are motivated and illustrated by a simple numerical example, compared with an alternative method recently published and applied to real data from the literature. Copyright © 2014 John Wiley & Sons, Ltd.