A causal artificial intelligence model for payment delay optimisation in supply chain financing

Supply chain financing (SCF) has become a popular approach for small- and medium-sized enterprises (SMEs) to improve financial resilience. Payment delays within SCF have emerged as a critical operational challenge for both suppliers and SCF providers. This paper aims to integrates causal AI modelling to proposed a framework to discover and optimise the operational treatments for mitigating payment delays in SCF. The proposed framework combines causal machine learning methods such as backdoor adjustment with Inverse Probability Weighting (IPW), and Double Machine Learning models (DoubleML). The proposed causal AI framework implements data-driven learning for the processes of causal discovery, causal effect estimation, and the optimisation of policy trees. This proposed framework aims to establish a cohesive method designed to identify potential treatment effects and assist in making operational decisions to mitigate payment delays. The effectiveness of the proposed framework is demonstrated through a case study on aerospace supply chain network. The generalisability and the industrial insights associated with the case study results have been analysed.