Efficiency estimation and reduction potential of the Chinese construction industry via SE-DEA and artificial neural network

Abstract

PurposeThe objective of this study is to evaluate the overall technical efficiency, labor efficiency, capital efficiency and equipment efficiency of 30 Chinese construction sectors to foster sustainable economic growth in the construction industry.Design/methodology/approachThis study employed the super-efficiency data envelopment analysis (SE-DEA) and artificial neural network model (ANN) to evaluate the industrial performance and improvement potential of the Chinese regional construction sectors from 2000 to 2017.FindingsResults showed that the overall technical and capital efficiencies displayed relatively stable patterns. Equipment efficiency presented a relatively huge fluctuation during the sample period. Meanwhile, labor, capital and equipment efficiencies could potentially improve in the next five years. A spatial examination of efficiencies implied that the economic level was still a major factor in determining the efficiency performance of the regional construction industry. Beijing, Shanghai and Zhejiang were consistently the leading regions with the best performance in all efficiencies. Shandong and Hubei were critical regions with respect to their large reduction potential of labor, capital and equipment.Research limitations/implicationsThe study focused on the regional efficiency performance of the construction industry; however, it failed to further deeply discover the mechanism that captured the regional inefficiency. In addition, sample datasets used to predict might induce the accuracy of prediction results. Qualitative policy implications failed to regress the efficiency performance of the industrial policy variables. These limitations will be discussed in our further researches.Practical implicationsEnhancing the overall performance of the Chinese construction industry should focus on regions located in the western areas. In comparison with labor and capital efficiencies, equipment efficiency should be given priority by eliminating outdated equipment and developing high technology in the construction industry. In addition, the setting of the national reduction responsibility system should be stratified to account for regional variations.Originality/valueThe findings of this study can provide a systematic understanding for the current and future industry performance of the Chinese construction industry, which would help decision makers to customize appropriate strategies to improve the overall industrial performance with the consideration of regional differences.

Data science and productivity: A bibliometric review of data science applications and approaches in productivity evaluations

“…Azadeh et al (2015) incorporated artificial neural networks (ANNs), genetic algorithms (GAs), and multivariate analysis to develop a flexible algorithm for the optimisation of machinery productivity. Most recently, Yuan et al (2020) evaluated industrial productivity and its improvement potential using a super-efficiency data envelopment analysis (SE-DEA) and ANN. For the random forest category, Liu et al (2018) focussed on the improvement of construction productivity by adopting a generalised additive model (GAM) and random forest to access the impact of outdoor ambient environmental factors on construction productivity.…”

A new cross-efficiency DEA approach for measuring the safety efficiency of China's construction industry

“…The construction industries in 31 regions of Chinese mainland are selected as research subjects. To evaluate their safety efficiency under the strategy of safety priority, the paper constructs an indicators system of inputs, desirable outputs, and undesirable outputs based on the literature summary and the reality of construction industry (Kang et al, 2020;Yang et al, 2022;Yuan et al, 2020;Zhou et al, 2019;Zhang et al, 2021). The specific indicators are shown as follows.…”

Application of Hybrid Approach in Banking System: An Undesirable Operational Performance Modelling

“…The findings concluded that DEA and ANN with genetic algorithm is the best approach to predict performance with 94% accuracy. Yuan et al (2020) studied the overall efficiency of Chinese construction companies by incorporating super efficiency DEA-ANN to identify and predict improvement potential. The study concluded that sustainability competitiveness exists within Indian hospitality industry.…”