Double-hierarchy hesitant fuzzy linguistic information-based framework for green supplier selection with partial weight information
- Publisher:
- Springer
- Publication Type:
- Journal Article
- Citation:
- Neural Computing and Applications, 2021, 33, (21), pp. 14837-14859
- Issue Date:
- 2021-11-01
Closed Access
| Filename | Description | Size | |||
|---|---|---|---|---|---|
| Krishankumar2021_Article_Double-hierarchyHesitantFuzzyL.pdf | 1.57 MB |
Copyright Clearance Process
- Recently Added
- In Progress
- Closed Access
This item is closed access and not available.
Green supplier selection (GSS) is a crucial issue in green supply chain management. CAPS indicate that industries spend yearly USD 25 million per procurement, which is a huge amount that necessitates a systematic GSS to avoid financial catastrophes. Literatures on GSS reveal that researchers have not addressed the issue of missing preferences, consistency of decision matrices, and repairing inconsistencies. Moreover, handling of complex linguistic expressions is another open challenge in GSS. Motivated by these research gaps, a two-stage decision framework was proposed. From the analysis of different linguistic models, it is clear that a double-hierarchy linguistic model is flexible for handling complex expressions. In the preprocessing stage, preferences were imputed using the case-based method. Later, consistency of matrices was determined using the Cronbach’s coefficient, and inconsistent matrices were repaired using the iterative method. In the next stage, new mathematical models were formulated to calculate weights of experts and criteria. Preferences were sensibly aggregated by using the Maclaurin symmetric mean operator, which captures the criteria interrelationship. Green suppliers were prioritized by using the TODIM method. Finally, the practicality, strengths, and weaknesses of the proposed framework were realized by demonstrating a case study of GSS and comparison with other methods. Results infer that the proposed framework (i) is consistent with the existing models, and the values are 0.60, 0.86, and 0.75, respectively; (ii) is robust with 100% rank-order stability even after adequate weight alterations; and (iii) finally can better discriminate suppliers with a broad deviation range of 0.34–0.35.
Please use this identifier to cite or link to this item: