Bayesian probabilistic models for corporate context, with an application to internal audit activities

TL;DR

This paper introduces a Bayesian network-based approach for fair performance evaluation in corporate internal audit activities, providing a transparent, explainable, and generalizable analytical methodology for data-driven decision-making.

Contribution

It develops a methodological framework for advanced analytics in corporate settings, including feature selection, model inference, and selection, applicable beyond the specific case study.

Findings

Successful implementation of Bayesian network for internal audit evaluation

Development of a transparent and explainable analytical toolbox

Framework applicable to various corporate decision-making contexts

Abstract

In this paper we present a business case carried out in Poste Italiane, in the context of fair performance evaluations of human resources engaged in internal audit activities. In addition to the development of a Bayesian network supporting the goal of the Internal Audit unit of Poste Italiane, the work has led to the development of a methodological approach to advanced analytics in corporate context, whose usefulness goes well beyond the specific use case described here. We thus present the different stages of such analytical strategy, from feature selection, to model structure inference and model selection, as a general toolbox that allows a completely transparent and explainable process to support data-driven decisions in business environments.