Weight_Subjective
DANP: DEMATEL-based ANP
Tzeng, G.-H., Huang, J.-J. · 2010
Overview
Hybrid: DEMATEL total-relation + ANP supermatrix. Output typically weight (higher value = preferred).
Strengths
- •Method-specific: Hybrid: DEMATEL total-relation + ANP supermatrix
Limitations
- •Assumes: Domain experts available
- •Assumes: Experts can express consistent comparisons
Method assistant
Grounded explanations: it explains the method, it does not compute.
Assumptions to verify
- •Domain experts available
- •Experts can express consistent comparisons
When not to use
- •No experts available → use objective weighting
- •High inconsistency → discard and re-elicit
Edge cases
- •if cyclic.
Common pitfalls
- •Hatalı: 'DANP bu varsayımı kontrol etmeden uygulamak'. Doğrusu: Domain experts available
- •Hatalı: 'DANP bu varsayımı kontrol etmeden uygulamak'. Doğrusu: Experts can express consistent comparisons
- •Hatalı: DANP'yi 'No experts available → use objective weighting' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.
- •Hatalı: DANP'yi 'High inconsistency → discard and re-elicit' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.
Worked example
- 1.Adım 1 (F1): Step 1: Direct-influence matrix A from experts (criteria pairwise influence on 0-4 scale). Formül: A = [a_{ij}]_{n\times n},\ a_{ij}\in\{0,1,2,3,4\} Anchor: Tzeng-Huang 2011, Ch.7 Eq.(7.1)
- 2.Adım 2 (F2): Step 2: Normalised direct relation X = A / max(row-sum, col-sum). Formül: X = \dfrac{A}{\max\big(\max_{i}\sum_{j} a_{ij},\ \max_{j}\sum_{i} a_{ij}\big)} Anchor: Tzeng-Huang 2011, p.140 Eq.(7.2)
- 3.Adım 3 (F3): Step 3: Total relation matrix T = X(I − X)^{−1} (DEMATEL output). Formül: T = X(I - X)^{-1} Anchor: Tzeng-Huang 2011, p.140 Eq.(7.3)
- 4.Adım 4 (F4): Step 4: Normalise T column-wise to form unweighted supermatrix W. Formül: w_{ij} = \dfrac{t_{ij}}{\sum_{i} t_{ij}}\ \text{(column-stochastic supermatrix)} Anchor: Tzeng-Huang 2011, p.141 Eq.(7.4)
- 5.Adım 5 (F5): Step 5: Limit supermatrix W^∞ = lim_{k→∞} W^k (powers converge to steady state). Use Cesaro sum if cyclic. Formül: W^{\infty} = \lim_{k\to\infty} W^{k}\ \text{or}\ \lim_{N\to\infty} \dfrac{1}{N}\sum_{r=1}^{N} W^{r} Anchor: Tzeng-Huang 2011, p.141 Eq.(7.5)
- 6.Adım 6 (F6): Step 6: Final criterion weights from any column of W^∞ (DEMATEL-derived dependency baked in). Formül: \mathbf{w} = W^{\infty}_{:,k}\ \text{(any column)} Anchor: Tzeng-Huang 2011, p.142 Sec.7.5
Commonly paired with
- •DANP + TOPSIS (high)
- •DANP + VIKOR (high)
- •DANP + EDAS (high)
- •DANP + PROMETHEE (high)
- •DANP + ELECTRE-III (high)
How to cite
Tzeng, G.-H.; Huang, J.-J. (2010). Multiple Attribute Decision Making: Methods and Applications. Chapman & Hall/CRC.