Title: Assessing and forecasting method of financial efficiency in a free industrial economic zone

Authors: Tomás José Fontalvo-Herrera; Enrique Delahoz-Dominguez; Orianna Fontalvo-Echavez

Addresses: Faculty of Economics Sciences, University of Cartagena, Cartagena, Colombia ' Industrial Engineering Department, Engineering Faculty, Universidad Tecnológica de Bolívar, Cartagena, Colombia ' Engineering Faculty, Universidad del Norte, Barranquilla, Colombia

Abstract: Industrial free zones are key to the economic progress of developing countries, making the evaluation and forecast of efficiency in these organisations relevant. This research proposes a three-phase method to evaluate and forecast the financial efficiency of the business profiles of companies belonging to the free economic zone of Cartagena - Colombia. The first phase consisted of a cluster analysis to determine representative groups among the companies analysed. In the second phase, financial efficiency is measured for each of the clusters found in phase 1. Finally, in phase 3 a machine learning model is trained and validated to predict the belonging of a company to a category of financial efficiency - cluster. The results show the creation of two business clusters, with an average efficiency of 49.8% and 14.6% respectively. The random forest model has an accuracy of 95% in the validation phase.

Keywords: data envelope analysis; DEA; clustering; machine learning; random forest; efficiency.

DOI: 10.1504/IJPQM.2021.115694

International Journal of Productivity and Quality Management, 2021 Vol.33 No.2, pp.253 - 270

Received: 09 Oct 2019
Accepted: 25 Dec 2019
Published online: 17 Jun 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article