Performance Evaluation of Multi-Stage Equipment Allocation Chains Based on Interval Cloud Model and Network DEA

Abstract

: Performance evaluation of equipment allocation chains is a critical closed loop linking strategy optimization and resilience design. To address the limitations of traditional evaluation methods—namely, the fragmentation of multi-stage transmission mechanisms, the neglect of expert cognitive uncertainty, and the lack of robustness verification—this paper constructs an integrated framework of “cloud model quantification, entropy-weight dimensionality reduction, network data envelopment analysis (DEA) evaluation, and Bootstrap robust ranking”. Specifically, an interval cloud model is employed to transform experts’ fuzzy linguistic evaluations into cognitive consensus boundaries; a stage-wise dimensionality reduction model based on fused entropy weights and game-theoretic combination weighting is designed to output subjective–objective coordinated composite intervals; a polarity-corrected interval serial network DEA (ISN-DEA) model with intermediate linking constraints is constructed to achieve four-stage efficiency decoupling and bottleneck identification from demand planning to performance feedback; and a possibility degree matrix is introduced to produce a complete ranking, whose robustness and convergence validity are jointly verified via Bootstrap resampling, multi-baseline comparison, and sensitivity experiments. Case studies demonstrate that the proposed method can effectively characterize the performance degradation patterns of the chain under different disruption scenarios, differentially identify typical performance states such as the “critical state” and the “system failure state”, precisely locate stage bottlenecks, and output statistically significant strategy rankings, thereby providing quantitative decision support for equipment prepositioning strategies and chain reconfiguration.