A Computational Analysis of Ceasefire Dynamics: Network Modelling of Violent Incident Data in South Sudan (2018–2023)

This original research employs computational methods to analyse the structural dynamics of conflict and peace in South Sudan. Using a network science approach, we model relationships between armed actors and violent incidents derived from the Armed Conflict Location & Event Data Project (ACLED) dataset for the period 2018–2020. The methodology constructs temporal bipartite networks, applies community detection algorithms, and calculates centrality metrics to identify key conflict drivers and the stability of ceasefire arrangements. Results reveal persistent, latent conflict networks that operate beneath formal peace agreements, highlighting specific actor clusters and geographic hubs that undermine the Revitalised Agreement on the Resolution of the Conflict in South Sudan (R-ARCSS). The discussion critically assesses the utility of computational modelling for conflict early warning and the limitations of top-down peace processes, concluding with policy-relevant insights for conflict mediation.