Ranking method · Zavadskas et al., 2012

WASPAS

Weighted Aggregated Sum Product ASsessment

A robustness trick: blend the weighted-sum and weighted-product models into one score.

How it works

1. Normalize by linear ratio: divide by the column best (benefit) or divide the column best by the value (cost).
2. Compute the weighted sum score (WSM) and the weighted product score (WPM).
3. Blend: Q = λ·WSM + (1−λ)·WPM. λ = 0.5 treats both models equally.

Use it when

• You want a simple, transparent score but more stability than a plain weighted sum.
• Compensation between criteria is acceptable (a strength can offset a weakness).

Watch out for

• Fully compensatory one great criterion can mask a serious weakness.
• Zero values break the product model (they are floored internally).

Parameters

λ ∈ [0, 1] the sum-versus-product blend (default 0.5).

Cite the method

Zavadskas, E. K., Turskis, Z., Antucheviciene, J., & Zakarevicius, A. (2012). Optimization of weighted aggregated sum product assessment. Elektronika ir Elektrotechnika, 122(6), 3-6.

@article{zavadskas2012optimization,
  author  = {Zavadskas, Edmundas Kazimieras and Turskis, Zenonas and Antucheviciene, Jurgita and Zakarevicius, Algimantas},
  title   = {Optimization of weighted aggregated sum product assessment},
  journal = {Elektronika ir Elektrotechnika},
  volume  = {122},
  number  = {6},
  pages   = {36},
  year    = {2012},
  doi     = {10.5755/j01.eee.122.6.1810}
}