A Comparison of Three-Stage DEA and Artificial Neural Network on the Operational Efficiency of Semi-Conductor Firms in Taiwan

Hsiang-Hsi Liu,Fu-Hsiang Kuo,Tser-Yieth Chen,Yung-Ho Chiu

doi:10.4236/me.2013.41003

Hsiang-Hsi Liu, Fu-Hsiang Kuo + Show 2 more

Open Access

https://doi.org/10.4236/me.2013.41003

Copy DOI

Abstract

In this study, the data envelopment analysis (DEA), three-stage DEA (3SDEA) and artificial neural network (ANN) are employed to measure the technical efficiency of 29 semi-conductor firms in Taiwan. Estimated results show that there are significant differences in efficiency scores among DEA, 3SDEA and ANN analysis. The advanced setting of the three stages mechanism of DEA does show some changes in the efficiency scores between DEA and ANN approaches. We further find that the environmental factor is still a significant variable to explain technical efficiency inTaiwan, irrespective of whether a DEA, 3SDEA or ANN approach is used.

Highlights

In this paper, we compare traditional data envelopment analysis (DEA), three-stage data envelopment analysis (3SDEA) and artificial neural network analysis (ANN) to estimate technical efficiency indices, and to explore the effect of environmental factors (Fried, Lovell, Schmidt and Yaisawarng, 2002) [1] on technical efficiency for policy purposes in the semi-conductor sector
We find that the mean technical efficiency score of 0.708 and 0.732 during the period 2005-2006 was lower than during the periods 2001-2004, at 0.866, 0.892, 0.911, and 0.844 respectively
We further find that the average pure technical efficiency and scale efficiency are 0.674 and 0.681 from the artificial neural network analysis (ANN), which are significantly lower than the result gained from the threestage DEA approach (0.884; 0.934), or the traditional DEA (0.901; 0.915) (Table 2)

Summary

Introduction

We compare traditional data envelopment analysis (DEA), three-stage data envelopment analysis (3SDEA) and artificial neural network analysis (ANN) to estimate technical efficiency indices, and to explore the effect of environmental factors (Fried, Lovell, Schmidt and Yaisawarng, 2002) [1] on technical efficiency for policy purposes in the semi-conductor sector. In the 1990s, Taiwan government declared the Technology Industry Establishment Promotion Decree which emphasized the importance of the semi-conductor industry as critical in the development of Taiwan’s manufacturing industry. The government included the semi-conductor industry in the ten emerging focus industries in 1998. Under this climate of focus, Taiwan started to increase operational efficiency in the semi-conductor industry, and to place extra funds into the supply market. The objective of this focus was to increase quality of life and reduce financial uncertainty. Taiwan’s semi-conductor companies have confronted accountability issues in the 2000s with administrators bringing some revolutionary changes

Objectives

Results

Conclusion