Recent advances in the standards of object-centric event logs offer opportunities to develop new process mining tools to create process models that depict relations between different object types of a business process. Process mining enables the discovery of process models from system event logs, which has advantages compared to traditional process modeling since it is more precise, faster, and data-driven. Business processes are related to multiple objects. Therefore, a new standard called Object-Centric Event Log - OCEL 2.0 has emerged that enables relating an event to multiple objects to avoid the drawbacks faced in traditional process mining, which considers only one object type. Causal nets or C-nets is a modeling language specifically created for process mining and largely used both in the backend and for model representation because it avoids inconsistencies present in other modeling languages. However, there is a gap in discovering object-centric models in a language that deals with noise and supports concurrency and choice. Hence, this thesis investigates how object-centric Causal nets can be discovered and represented to overcome current limitations. It uses the design science research paradigm to develop algorithms for discovering and visualizing object-centric Causal nets. The artifact is demonstrated by discovering an object-centric Causal nets model from an OCEL 2.0 log in two scenarios. The evaluation is done by comparing an object-centric Causal nets model (OCCN) to an object-centric Petri net model (OCPN) through a survey for quantitative and qualitative assessment. The technology acceptance model - TAM is used to measure users’ acceptance. The results show that the artifact generates object-centric Causal nets models that are perceived as useful and easy to use, understandable, with positive user feedback. The final considerations include directions regarding future work.