A new objective weighting method based on robustness of ranking with standard deviation and correlation: The ROCOSD method

Yükleniyor...
Küçük Resim

Tarih

2023

Dergi Başlığı

Dergi ISSN

Cilt Başlığı

Yayıncı

Elsevier Inc.

Erişim Hakkı

info:eu-repo/semantics/closedAccess

Özet

In this study, we proposed a new objective criteria weighting method simultaneously addressing standard deviations, correlation coefficients and robustness of ranking. The Robustness, Correlation, and Standard Deviation (ROCOSD) method assigns weight values according to three objectives. The first objectives are intended to minimize the overall maximum deviation from the ratio that the criteria deserve and are based on calculated standard deviations and correlation coefficients. The final one maximizes the minimum change in any criteria weight which leads to a rank reversal on alternatives. We focused on four cases and examined the proposed method, and later created a comparative analysis based on the application of five well-known methods. The average rank correlation consistency and weighting accuracy indicated that ROCOSD achieved better statistics in various ranges (0.0365% to 37.3393%), (4.6292% to 28.4925%) respectively. Compared to other well-known techniques, the average results for robustness, the primary tenet of ROCOSD, demonstrated that the proposed approach yielded far better results (ranging from 1.6407 to 6.4706 times). With the aim of determining the performance of the ROCOSD method in real-life applications, we examined e-government development and evaluated the effectiveness and success of the ROCOSD method.

Açıklama

WOS:000982263100001

Anahtar Kelimeler

Correlation, Criteria Weighting, E-Government Development, Multiple Criteria Analysis, Robustness, Standard Deviation

Kaynak

WoS Q Değeri

N/A

Scopus Q Değeri

Q1

Cilt

636

Sayı

Künye

Pala, O. (2023). A new objective weighting method based on robustness of ranking with standard deviation and correlation: The ROCOSD method. Information Sciences, 636 doi:10.1016/j.ins.2023.04.009