A multi-criteria and stochastic robustness analysis approach to compare nations sustainability

https://doi.org/10.1016/j.seps.2021.101159Get rights and content

Highlights

An approach for ranking and clustering countries with similar sustainability challenges.

PROMETHEE II multi-criteria method is used for ranking and an iterative process is used to identify ordered clusters

Robustness analysis helps to identify groups with robust sustainability paths.

The approach is applied to the Human Development Index and the Sustainable Development Goals Index.

Three groups of countries are identified, providing new and helpful visualizations of nations in HDI and SDG.

Abstract

Clustering nations using sustainability indexes has been proposed as a means to support decision-making in politics, academia and business. This starts by computing indexes that describe the social, economic and environmental progress of countries, and continues by ranking and clustering countries that face similar sustainability challenges. Although multi-criteria methods exist to support this process, two main issues have not been addressed in an integrated approach: uncertainty in decision-making parameters and imprecise data sources. In this study, we propose a procedure in which the PROMETHEE II multi-criteria method is used for ranking countries and an iterative algorithm is applied to find ordered clusters. A stochastic robustness analysis procedure helps to identify groups with robust sustainability paths. The approach is applied to the Human Development Index and the Sustainable Development Goals Index. By analyzing the ambiguous assignments of nations to clusters three groups of countries can be identified. First, countries that have very good arguments to be considered in a given cluster. Second, countries which have some merits to be considered in a cluster and where focusing resources would allow them to improve in their current group or jumping into a better one. Third, nations that have very ambiguous arguments to be considered in any cluster such that the support to improve their performance would have to be analyzed in depth on multiple fronts. Thus, policy decision-making could be enhanced with this approach, providing new and helpful visualizations of nations in HDI and SDG. Finally, limitations of this proposal and future research are discussed.

Keywords

HDI
SDG
Clustering
PROMETHEE II
Stochastic

Javier Pereira received his Diploma in Computer Engineering from the Technical University Federico Santa María, Chile (1990) and a PhD in Management Science from the - Paris IX University - Dauphine, France (1995). The last five years he has directed international industrial projects, with development of information technologies, for production planning, control and monitoring systems. He has also participated as lead coordinator of industrial projects in logistics processes, production management and supplier development. His research interests include Information Technology, Supply Chain Management, Decision Analysis and Data Science. Currently, he participates in advanced manufacturing and logistics projects, applying simulation techniques, decision analysis and Data Science, using stochastic techniques (bootstrap DEA, fuzzy DEA, fuzzy-sets).

Pedro Contreras received his Diploma in Business Management and Informatics from the University of Talca, Chile in 1999, a MPhil in Computer Science from The Queen's University of Belfast in 2006 and a PhD from the in University of London in 2010. In 2000 he worked as researcher at the German Institute for Economic Research (DIW) in Berlin. The same year he moved to Belfast to work as research assistant at the Department of Computer Science at The Queen's University of Belfast. After 5 years in 2005 he moved to Royal Holloway, University of London, to work as a research assistant. In 2010 after obtaining his PhD he moved to the private sector where he gained industrial experience in machine learning and distributed systems. Since 2018 he has been working as a Lead Engineer in the Cyber Security Group at The Warwick Manufacturing Group in The University of Warwick. Dr. Contreras has authored many papers in machine learning and has a wide experience in decision automatisation in defence gained through work sponsored with the Defence Science and Technology Laboratory in the UK.

Danielle Costa Morais is an associate professor in the Management Engineering Department at Universidade Federal de Pernambuco (UFPE) since 2007, Director of Post-Graduate Program of Management Engineering at UFPE (2008–2010 and 2013-currently), Director of the research group on Decision and Negotiation for Water Management (DNW). She has been awarded a grant of Productivity in Research by CNPq (Brazilian NRC). Her research interests include MCDM/A, Group Decision and Negotiation, Operational Research and Water Resources Management. She co-authored over 40 scientific papers in reviewed journals. She serves on the editorial board of a few scholarly journals, such as: Group Decision and Negotiation. She has been an active member of the main societies related to Operational Research, MCDM/A and Group Decision, and served the INFORMS MCDM section as a board member.

Pilar Arroyo is Professor of the EGADE Business School of Tecnológico of Monterrey campus Toluca, Mexico. She holds a PhD degree in Business Administration from the Tecnológico de Monterrey in Mexico. She is member of the Mexico National Research System and has published articles on topics such as outsourcing, reverse logistics, supplier development, green marketing, social marketing for health care, and social entrepreneurship. Her research has been published in reputed journals such as the Journal of Supply Chain Management, Journal of Business Process Management, Qualitative Marketing Research, Teaching and Teacher Education, Management Research Review and Journal of Consumer Marketing. She has also authored several books and chapters in books, and presented research articles in several international conferences.

View full text