Automating business processes in enterprise resource planning (ERP) systems offers significant potential to reduce manual workload and improve efficiency. The limitations of robotic process automation in handling complex business processes have sparked an increasing interest in artificial intelligence-based solutions, particularly those using large language models (LLM). Although existing approaches utilize LLM-based agents for business process automation, they typically do not leverage the benefits of multi-agent collaboration, such as divergent thinking or reasoning. Additionally, important security principles like segregation of duties (SoD) are not considered. Therefore, this study addresses these gaps by proposing the use of multiple LLM-based agents for automating ERP business processes. An error-driven prompt engineering approach is presented to design reliable business process automation involving multiple agents. This approach is applied to the procure-to-pay process, taking into account different multi-agent scenarios to incorporate SoD. Through error-driven prompt engineering, a consistently reliable automation was achieved. Overall, this study aims to facilitate the use of ERP systems in a collaborative environment while ensuring compliance with fundamental security principles.