Modeling ESG risks for supply chain due diligence from media monitoring data using bivariate penalized approximate likelihood

Read full text Summarise with VIRAT
Authors: Joscha Krause, Cathrine Aeckerle-Willems, Stéphane Cortina, Yuejun Guo, Ruben J. Marques, Felix Viroli
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

New legislation requires European companies to perform supply chain due diligence by regularly assessing the risk of their suppliers committing ESG violations, such as relying on child labor. Assessments are typically done using public risk indicators for countries and industries in which the suppliers operate. However, the approach often does not represent the actual risk stemming from a particular supplier accurately. Moreover, risk indicators are usually only updated annually and do not reflect current developments. Online data collected from media monitoring on the suppliers can augment these indicators to provide more accurate and timely assessments. We propose a bivariate penalized temporal logit mixed model for combining static risk indicators with dynamic media data to predict ESG violation risks in the supply chain of a company. Model parameter estimation is performed with a two-step approximate likelihood algorithm. The methodology is tested in Monte Carlo experiments. An application to real-world data is provided.

Fields of Study & Subjects

Sustainable Supply Chain Management Supply Chain Resilience and Risk Management Statistical Methods and Inference

Publication details

Journal / source
Figshare
Publication date
22 March 2026
Language
EN