What is Inverse DEA?

What can InverseDEA do?

▷  Inverse DEA (inverse data envelopment analysis) discusses such problems: among a group of decision making units (DMU),

If we increase certain inputs and assume that the DMU maintains its current efficiency level, how much more outputs could the unit produce? Or,

If the outputs need to be increased to a certain level and the efficiency of the DMU remains unchanged, how much more inputs should be provided?

--Wei Q., et al. (2000)

Wei, Q., et al. (2000). An inverse DEA model for inputs/outputs estimate.European Journal of Operational Research 121(1): 151-163

▷ Inverse DEA can be generalized as follows

Assume that the DMU maintains its current efficiency level,

•If we change the input level (such as increasing all inputs, decreasing all inputs, increasing some inputs while decreasing some inputs and other inputs unchanged), how much outputs could the DMU produce?

•If we expect the output level to be changed (such as all outputs increased, all outputs decreased, some outputs increased while some outputs decreased and other outputs unchanged), how much inputs should be allocated?

Assume that we want the DMU to achieve a pre-set efficiency level,

•If the input level is given (the inputs may be increased, decreased, or some increased, some decreased and others unchanged), how much outputs should the DMU produce?

•If we set the output level (the outputs may be increased, decreased, or some increased, some outputs decreased and others unchanged), how much inputs should be allocated?

▷  With InverseDEA, you can solve all the left problems, including

Assume that the DMU maintains its current efficiency level,

•Given changed inputs, compute the targeted outputs

•Given changed outputs, compute the targeted inputs

Assume that you want the DMU to achieve a pre-set efficiency level,

•Given changed inputs, compute the targeted outputs

•Given changed outputs, compute the targeted inputs

The problems can be solved using

•Input-oriented model

•Output-oriented model

The problems can be solved under technology of

•Constant returns to scale (CRS)

•Variable returns to scale (VRS)

•Non-increasing returns to scale (NIRS)

•Non-decreasing returns to scale (NDRS)

*1: Here “changed inputs/outputs” means that the inputs/outputs may be increased, decreased, some increased, some decreased and others unchanged.

*2: Super-efficiency and undesirable outputs are supported.

*3: Software Platform: Windows only, and the 64-bit MS Office Professional is required.