A Generalized Analytical Framework for the Nonlinear Best-Worst Method (2508.06048v1)

Abstract: To eliminate the need for optimization software in calculating weights using the nonlinear model of the Best-Worst Method (BWM), Wu et al. proposed an analytical framework for deriving optimal interval-weights. They also introduced a secondary objective function to select the best optimal weight set. However, their framework is only compatible with a single Decision-Maker (DM) and preferences quantified using the Saaty scale. In this research, we generalize their framework to accommodate any number of DMs and any scale. We first derive an analytical expression for optimal interval-weights and select the best optimal weight set. After demonstrating that the values of Consistency Index (CI) for the Saaty scale in the existing literature are not well-defined, we derive n formula for computing CI. We also derive analytical expressions for the Consistency Ratio (CR), enabling its use as an input-based consistency indicator and proving that CR satisfies some key properties, ensuring its reliability as a consistency indicator. Furthermore, we observe that criteria with equal preferences may get different weights when multiple best/worst criteria are present. To address this limitation, we modify the original optimization model for weight computation in such instances, solve it analytically to obtain optimal interval-weights, and select the best optimal weight set. Finally, we demonstrate and validate the proposed approach using numerical examples.