Stratified Best Worst Method
Also known as: Stratified BWM
Stratified BWM is an extension of the Best Worst Method that applies the BWM logic recursively across multiple hierarchical layers. Instead of weighting criteria at a single level, it identifies the best and worst criterion within each level of a hierarchy, then aggregates weights across levels. This enables more realistic modeling of complex decision problems with natural hierarchical structures.
Key highlights
- Respects hierarchical problem structure; weights are derived consistently within and across levels
- Reduces cognitive burden per level; best-worst comparisons are easier when criteria at a level are fewer and more homogeneous
- Enables distributed decision-making; different teams can apply BWM to their respective hierarchical level
- Captures dependencies; weights at lower levels are conditioned on higher-level structure
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use Stratified BWM for hierarchical decision problems with 3+ levels. It is especially valuable when different stakeholders are responsible for different levels (e.g., executives set strategic criteria, managers set tactical ones). Prefer it over flat BWM when the hierarchy reflects the problem structure and reduces cognitive burden significantly.
Strengths & limitations
- Respects hierarchical problem structure; weights are derived consistently within and across levels
- Reduces cognitive burden per level; best-worst comparisons are easier when criteria at a level are fewer and more homogeneous
- Enables distributed decision-making; different teams can apply BWM to their respective hierarchical level
- Captures dependencies; weights at lower levels are conditioned on higher-level structure
- Requires explicit hierarchical structure; ambiguous or non-hierarchical problems are difficult to model
- Aggregation across levels is model-dependent; different aggregation rules (multiplication, weighted sum) yield different results
- Sensitive to hierarchy design; redefining the hierarchy can significantly change final weights
- Computational complexity grows with hierarchy depth; deep hierarchies may require many pairwise comparisons
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How deep should my hierarchy be?
Generally, 3-4 levels is optimal. Deeper hierarchies increase cognitive burden and computational complexity without proportional gains in accuracy. If you need deeper levels, consider clustering criteria at lower levels to reduce branching factor.
How do I aggregate weights across levels?
Multiplication is most common (global weight = level-1 weight × level-2 weight × ...). Weighted sum is an alternative if levels represent independent aspects. Always justify your choice and test sensitivity to aggregation method.
Can different teams work on different levels simultaneously?
Yes, that is one key advantage of Stratified BWM. Teams can work in parallel on their respective levels, then combine results. However, ensure that the hierarchy is clearly defined upfront so teams align on structure.
Sources
- 1.Rezaei, J. (2015). Best-worst multi-criteria decision-making method: Some properties and a linear model. Journal of Cleaner Production, 229, 976-985.
- 2.Asadzadeh, A., Rezaei, J., & Tavasszy, L. (2017). Sustainability assessment of supplier selection criteria using the BWM method in hierarchical procurement. Supply Chain Management, 22(6), 551-563.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 3). Stratified Best Worst Method. ScholarGate. https://scholargate.app/decision-making/stratified-bwm