Generalized robust window data envelopment analysis approach for dynamic performance measurement under uncertain panel data

This paper proposes a robust window data envelopment analysis (RWDEA) approach for assessing the dynamic performance of decision making units (DMU) in the presence of panel and uncertain data. To present the RWDEA method, generalized data envelopment analysis (GDEA) model, window analysis (WA) method, and robust optimization (RO) approach are taken into account. The proposed RWDEA approach can be used under different returns to scale (RTS) assumptions, including constant returns to scale (CRS), non-increasing returns to scale (NIRS), non-decreasing returns to scale (NDRS), and variable returns to scale (VRS). Notably, the RWDEA model is linear and can fully rank DMUs under deep uncertainty. To solve and show the validity of the proposed approach, the RWDEA model is implemented for evaluating the efficiency of the intellectual capital of 10 automotive and parts manufacturing companies. The results indicate that the RWDEA approach is applicable and useful for the dynamic efficiency assessment of DMUs in the presence of uncertain panel data. The RWDEA approach, by considering the uncertainties in the data and using panel data, provides more reliable results in comparison with the classical DEA models.