Wyniki wyszukiwania - BazTech

Ograniczanie wyników

1 Foundations of Computing and Decision Sciences

1 2009

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

A frontier-based neural network for assessing the efficiency of activity units

Curram S. P., Athanassopoulos A. D., Shale E. A.

Foundations of Computing and Decision Sciences

2009

Vol. 34, No. 1

23-39

This paper uses neural networks to assess the relative efficiency of activity units. A new formulation is proposed whereby the neural network fit is obtained by minimising one-sided errors which are attributed to technical inefficiency, effectively modelling a frontier. Efficiency ratings are given to rank activity units on a (0, 100] scale. Experiments on artificial data are employed to test the proposed method for its accuracy. These results are also compared with the corresponding efficiency ratings obtained from data envelopment analysis. Factors relevant to methodology selection for implementation are discussed. The empirical results of the paper support the potential role of neural networks as a reliable method of frontier efficiency assessment