PiF-SAW: Picture extension of SAW

Overview

Picture outranking/ranking: Picture Fuzzy Set (PiFS: μ, η, ν; μ+η+ν ≤ 1). Output typically utility (higher value = preferred).

Strengths

• •Method-specific: Picture outranking/ranking: Picture Fuzzy Set (PiFS: μ, η, ν; μ+η+ν ≤ 1)
• •Preserves picture uncertainty through the pipeline rather than premature crispification at elicitation
• •Native group-decision support (multi-DM aggregation built into the pipeline)

Limitations

• •Assumes: Decision matrix entries are valid Picture Fuzzy numbers/tuples
• •Assumes: Underlying crisp method's compensation assumption holds in uncertain space
• •Assumes: All decision-maker(s) and experts use the same linguistic/uncertainty scale

Assumptions to verify

• •Decision matrix entries are valid Picture Fuzzy numbers/tuples
• •Underlying crisp method's compensation assumption holds in uncertain space
• •All decision-maker(s) and experts use the same linguistic/uncertainty scale

When not to use

• •Crisp data sufficient: use base SAW directly (avoid unnecessary uncertainty layer)
• •Aggregation operator (PFWA/PFOWA/etc.) not specified: output ambiguous

Edge cases

• •Ties broken by accuracy H(α) = μ + η + ν (Garg 2017 Eq.2).

Common pitfalls

• •Hatalı: 'PIF-SAW bu varsayımı kontrol etmeden uygulamak'. Doğrusu: Decision matrix entries are valid Picture Fuzzy numbers/tuples
• •Hatalı: 'PIF-SAW bu varsayımı kontrol etmeden uygulamak'. Doğrusu: Underlying crisp method's compensation assumption holds in uncertain space
• •Hatalı: 'PIF-SAW bu varsayımı kontrol etmeden uygulamak'. Doğrusu: All decision-maker(s) and experts use the same linguistic/uncertainty scale
• •Hatalı: PIF-SAW'yi 'Crisp data sufficient' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.
• •Hatalı: PIF-SAW'yi 'Aggregation operator (PFWA/PFOWA/etc.) not specified' durumunda kullanmak: recommendation_metadata.not_recommended_when alternatif öneriyor.

Worked example

1. 1.Adım 1 (F1): Step 1: Construct picture fuzzy decision matrix D = (α_ij)_{m×n} where α_ij = ⟨μ_ij, η_ij, ν_ij⟩ with μ_ij + η_ij + ν_ij ≤ 1. Formül: D = (α_ij)_{m×n}; α_ij = ⟨μ_ij, η_ij, ν_ij⟩; μ_ij + η_ij + ν_ij ≤ 1 Anchor: Cuong 2013 Def.1; Garg 2017 Sec.4 Step 1
2. 2.Adım 2 (F2): Step 2: Cost-complement normalization: for cost (min) criteria r_ij = α_ij^c = ⟨ν_ij, η_ij, μ_ij⟩; for benefit (max) criteria r_ij = α_ij. Converts all criteria to benefit-type without altering μ+η+ν ≤ 1. Formül: r_ij = α_ij^c = ⟨ν_ij, η_ij, μ_ij⟩ if c_j ∈ cost; r_ij = α_ij = ⟨μ_ij, η_ij, ν_ij⟩ if c_j ∈ benefit Anchor: Garg 2017 Eq.8 (Sec.4 Step 2)
3. 3.Adım 3 (F3): Step 3: PFWA aggregation per alternative with weight vector w: r_i = PFWA_w(r_i1, ..., r_in). μ_i = 1 - ∏(1-μ_ij)^w_j; η_i = ∏(η_ij)^w_j; ν_i = ∏(ν_ij)^w_j. Formül: r_i = PFWA_w(r_i1, ..., r_in) = ⟨ 1 - ∏_{j=1}^n (1-μ_ij)^{w_j}, ∏_{j=1}^n (η_ij)^{w_j}, ∏_{j=1}^n (ν_ij)^{w_j} ⟩ Anchor: Garg 2017 Eq.4 (PFWA), Sec.4 Step 3
4. 4.Adım 4 (F4): Step 4: Defuzzify r_i using Garg score function S(α) = μ - η - ν. Formül: s_i = S(r_i) = μ_i - η_i - ν_i, s_i ∈ [-1, 1] Anchor: Garg 2017 Eq.1 (Sec.4 Step 4)
5. 5.Adım 5 (F5): Step 5: Rank alternatives in descending order of s_i. Ties broken by accuracy H(α) = μ + η + ν (Garg 2017 Eq.2). Formül: ranking = argsort_desc(s_i); on tie use H(α) = μ + η + ν Anchor: Garg 2017 Sec.4 Step 5

Commonly paired with

• •n_a + PIF-SAW (common)

How to cite