Optimization method of additive network DEA model in big data environment — Based on two-stages model perspective

被引：0
|
作者：
Wei Y. [1 ]
Yang M. [2 ,3 ,4 ]
Liang L. [2 ,3 ,4 ]
Yu Y. [1 ]
机构：
[1] School of Business, Nanjing Audit University, Nanjing
[2] School of Management, Hefei University of Technology, Hefei
[3] Ministry of Education Key Laboratory of Process Optimization and Intelligent Decision Making, HeFei University of Technology, Hefei
[4] Anhui Laboratory of Intelligent Interconnection System, Hefei University of Technology, Hefei
来源：
Xitong Gongcheng Lilun yu Shijian/System Engineering Theory and Practice | 2023年 / 43卷 / 11期
基金：
中国国家自然科学基金;
关键词：
big data; data envelopment analysis; optimal solution; two-stages;
D O I：
10.12011/SETP2022-2028
中图分类号：
学科分类号：
摘要：
Additive network DEA model is one kind of highly nonlinear programming problem, which is difficult to be solved directly. The existing heuristic algorithm cannot be applied to obtain the exact solution of the model, and will consumes a lot of computing resources and times in the big data environment. In order to solve these two issues, the presented study mainly does two aspects of work. Firstly, aiming at the issue of solving precision of the model, a solving method of quadratic fractional programming is proposed in this study, which decomposes the model into a finite quadratic programming, and it is proved theoretically that the presented method can be applied to obtain the exact solution of the model. Then, in view of the issue of the slow solving speed in the big data environment, this work optimized the model’s constraints by reducing the number of constraints and the consumption of computing resources while the feasible region remained unchanged, so as to improve the solving speed of the model. The numerical case results show that the proposed method can effectively improve both of the calculation accuracy and speed of the additive two-stages DEA model. © 2023 Systems Engineering Society of China. All rights reserved.
引用
收藏
页码：3294 / 3306
页数：12
相关论文
共 28 条
  • [1] Ali A I., Streamlined computation for data envelopment analysis[J], European Journal of Operational Research, 64, 1, pp. 61-67, (1993)
  • [2] Dula J H, Helgason R V., A new procedure for identifying the frame of the convex hull of a finite collection of points in multidimensional space[J], European Journal of Operational Research, 92, 2, pp. 352-367, (1996)
  • [3] Dula J H, Helgason R V, University S M, Et al., An algorithm for identifying the frame of a pointed finite conical hull[J], ORSA Journal on Computing, 10, 3, pp. 323-330, (1998)
  • [4] Barr R S, Durchholz M L., Parallel and hierarchical decomposition approaches for solving large-scale data envelopment analysis models[J], Annals of Operations Research, 73, pp. 339-372, (1997)
  • [5] Zhu Q, Wu J, Song M., Efficiency evaluation based on data envelopment analysis in the big data context[J], Computers & Operations Research, 98, 1, pp. 291-300, (2018)
  • [6] Dula J H., A method for data envelopment analysis[J], INFORMS Journal on Computing, 23, 2, pp. 284-296, (2011)
  • [7] Dula J H, Lopez F J., DEA with streaming data[J], Omega, 41, 1, pp. 41-47, (2013)
  • [8] Dula J H, Thrall R M., A computational framework for accelerating DEA[J], Journal of Productivity Analysis, 16, 1, pp. 63-78, (2001)
  • [9] Chen W, Cho W., A procedure for large-scale DEA computations[J], Computers and Operations Research, 36, 6, pp. 1813-1824, (2009)
  • [10] Chen W, Lai S., Determining radial efficiency with a large data set by solving small-size linear programs[J], Annals of Operations Research, 250, 1, pp. 147-166, (2017)
123