An RM-ODP method to model PBL digital twins for supporting closed-loop supervision

Project-Based Learning (PBL) in software engineering produces heterogeneous evidence across repositories, task boards, classroom activities, and assessments, making timely supervision difficult. This paper presents a two-stage method to model PBL digital twins using the five RM-ODP viewpoints, aligned with ISO/IEC 10746 and ISO/IEC/IEEE 42010. Stage A models the as-is PBL ecosystem using SADT/IDEF0 and identifies native event points and requirement gaps. Stage B introduces a Digital Twin layer that consumes these events, applies data contracts, correlates patterns through Complex Event Processing (CEP), and delivers feedback within the active sprint window without replacing existing processes. The method structures goals and governance (enterprise viewpoint), semantic contracts and traceability (information viewpoint), event services and CEP rules (computational viewpoint), distributed pipelines and immutable snapshots (engineering viewpoint), and implementation choices (technology viewpoint). An institutional case instantiation demonstrates end-to-end applicability from process modeling to operational supervision signals. The result is a reusable procedure for supporting closed-loop supervision, auditable evidence integration, and longitudinal learning analytics in diverse PBL environments. In the current research cycle, Stage A is fully instantiated and Stage B is operational for repository-based evidence; correlation is executed as batch replay over sprint windows rather than as a streaming engine, so that the supervision-signal loop is demonstrated while the intervention-response loop that records how signals change pedagogical practice remains planned. The validation section quantifies this maturity and explicitly separates implemented, partially implemented, and planned capabilities.