Integrating Relative Efficiency Models with Machine Learning Algorithms for Performance Prediction

Bibliographic Data

ID16910951
AuthorsMarcos Gonçalves Perroni (0000-0003-1702-8176, Pontifícia Universidade Católica do Paraná), Claudimar Pereira Da Veiga (0000-0002-4960-5954, Fundação Dom Cabral, corresponding author), Elaine Forteski (Universidade do Contestado), Diego Antonio Bittencourt Marconatto (0000-0002-9458-9199, Fundação Dom Cabral), Wesley Vieira Da Silva (0009-0003-0913-702X, Universidade Federal de Alagoas), Carlos Otávio Senff (0000-0001-7408-0072, Pontifícia Universidade Católica do Paraná), Zhaohui Su (0000-0003-2005-9504, Southeast University)
Year2024
Volume14
Issue2
Publication date2024-04-01
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueSAGE Open (JOURNAL)
Journal identifiersISSN: 2158-2440 • E-ISSN: 2158-2440
PublisherSAGE Publications (PUBLISHER • US)
DOI10.1177/21582440241257800
OpenAlexW4399735906
LanguageEN
Citations received2
References cited44

Predicting operational performance enables organizations to develop operational effectiveness goals considering different combinations of resources. Measuring performance is consolidated with advances in relative efficiency analysis techniques, including data envelopment analysis (DEA) and stochastic frontier analysis (SFA), albeit these methods lack predictive capability. This paper proposes an approach for performance prediction by integrating relative efficiency measurement models with machine learning algorithms. Data analyses were conducted using data provided by the energy assessment project offered to small and medium-sized manufacturing companies in the United States ( n 7,548) using sales as the output, with the inputs being the number of employees, hours of operation, electricity, natural gas, cost of electricity, and cost of natural gas. Performance was estimated differently, employing parametric (SFA) and non-parametric (DEA) methods. The prediction benchmarking process occurred by adopting machine learning algorithms: regression (LM), support vector machine (SVM), K-nearest neighbor (KNN), linear discriminant analysis (LDA), random forest (RF), and decision tree (DT). The findings showed that it is possible to identify the best prediction algorithm associated with a performance model. However, the performance prediction may differ if different strategies for measuring performance or machine learning model configurations are used. In addition, SFA-LOG and SVM had the best performance for regression, and DEA-VRS/IRS excelled with random forest; the RF algorithm was the best fit across all performance approaches. The error rate depends on the algorithm and the performance model, and the number of classes must be reduced to obtain a higher success rate

Algorithm · Machine learning · Computer Science · Efficiency Analysis Using DEA · Energy Efficiency and Management · Forecasting Techniques and Applications · Artificial Intelligence

Tools

Open DOIOpen Access