Study on Improved DEA-Based MPI and Its Application in Dynamic Performance Evaluation

Malmquist productivity index (MPI) has been applied to express the dynamic performance over time, which can provide valuable information about the effectiveness of an industry. However, there exist scarcities about multiple correlation problems of variables and limitation of relationship between Dec...

Full description

Saved in:
Bibliographic Details
Main Authors: Fu Bo-xin, Song Xiang-qun, Guo Zi-jian, Fang Zhuo
Format: Conference Proceeding
Language:English
Subjects:
Conference management
Containers
Data envelopment analysis
data envelopment analysis (DEA)
Decision making
dynamic performance evaluation
Engineering management
Industrial relations
malmquist productivity index (MPI)
Performance analysis
Principal component analysis
principal components analysis (PCA)
Productivity
Technology management
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Malmquist productivity index (MPI) has been applied to express the dynamic performance over time, which can provide valuable information about the effectiveness of an industry. However, there exist scarcities about multiple correlation problems of variables and limitation of relationship between Decision Making Units (DMU) and input/output indexes. In this paper, a modification approach of DEA-Based MPI is proposed by using principal components analysis (PCA) and other relevant ways. A real data set about ten container ports in China over the period of 2001 to 2005 has been used to validate the method proposed in this paper. It is necessary to adopt effective policies and measures to improve the performance of productivity and competitiveness of this industry, and to promote the sustainable development of container accumulation ability of hinterland in different regions. The improved DEA-based MPI approach provides a practical tool to assist policies design and make strategic decisions for ameliorating the total performance in many fields.
ISSN:2155-1847
DOI:10.1109/ICMSE.2007.4421888