ABC analysis is a famous technique for inventory classification. However, this technique on the inventory classification only considering one indicator even though other important factors may affect the classification. To address this issue, researchers have proposed multiple criteria inventory classification (MCIC) solutions based on data envelopment analysis (DEA)-like methods. However, previous models almost evaluate items by different weight sets, and the index system only contains quantitative criteria and output indicators. To avoid these shortcomings, we propose an improved common-weight DEA model for MCIC issue. This model simultaneously considers quantitative and qualitative criteria as well as establishes a comprehensive index system that includes inputs and outputs. Apart from its improved discriminating power and lack of subjectivity, this non-parametric and linear programming model provides the performance scores of all items through a single computation. A case study is performed to validate and compare the performance of this new model with that of traditional ABC analysis, DEA–CCR and DEA–CI. The results show that apart from the highly improved discriminating power and significant reduction in computational burden, the proposed model has achieved a more comprehensive ABC inventory classification than the traditional models.
Sustainable efficiency and CO2 reduction potential of China’s construction industry: Application of a three-stage virtual frontier SBM-DEA model
