Awadh Pratap Singh

Abstract titled "Computing and Predicting Efficiency of Educational Institutions Using Machine Learning Data Envelopment Analysis" selected in The International Conference on Nonlinear Applied Analysis & Optimization (2022) at the Department of Mathematical Sciences, IIT (Banaras Hindu University), Varanasi during December 19-22, 2022, under the aegis of Indian Institute of Technology (BHU).

Data envelopment analysis (DEA) models are used to compute the relative efficiency of the decision-making units. Various methods have been proposed in the literature to rank decision-making units (DMUs) employing DEA. However, suppose a new DMU needs to be known for its efficiency or inefficiency; the DEA analysis would have to be re-conducted, requiring a large amount of computation. Our study aims to establish a relationship between the DEA method and machine learning (ML) algorithms. It proposes an approach that combines DEA with ML (ML-DEA) algorithms to measure and forecast the DMU's efficiency. Three ML-DEA algorithms will be discussed: the DEA model combined with support vector machines (SVM-DEA) with random forests (RF-DEA), respectively. To illustrate the applicability of the proposed model, the performance of Indian Institutes of Technologies (IITs) will be measured, estimated, and compared with the efficiency scores obtained by the DEA model.

The proposed bi-objective fuzzy data envelopment analysis (BOFDEA) model is a new approach to assess the performance efficiency of decision-making units (DMUs) in uncertain environments using ?-cuts. The model is based on fuzzy data envelopment analysis (FDEA) and considers two objectives and a solution method and ranking system are provided. Generally, the efficiency score obtained for a DMU using the ?-cut approach is an interval. Intervals are partially ordered sets, due to which ranking intervals is a challenging task. The proposed BOFDEA model with ?-cuts provides the efficiency of DMUs in the crisp form, not the form of intervals. Due to this, ranking DMUs with the proposed method's help becomes very easy and less computationally. We have validated the proposed model through numerical examples and showcased its practicality by demonstrating a real-world application in education. 

https://conferences.iitbhu.ac.in/ICNAAO2022/imageIITBHU/AcceptedAbstracts.pdf