PlanProjU: Automated Planning System for University Projects

Effective implementation of university extension projects requires a structured ap-proach with explicit objectives, dependencies, and resource constraints. Automated planning, particularly Hierarchical Task Network (HTN) planning, provides an opera-tional method to schedule complex activities by decomposing high-level goals into ex-ecutable tasks. This paper presents a tool that converts university project documents into BPMN 2.0 declarative processes and automatically produces HTN planning in-puts. Following the Design Science Research methodology, the architecture integrates (i) a pre-planner parser that extracts activities, roles, and precedence relations from BPMN models, (ii) generators that create domain and problem files, and (iii) the HTN planners SHOP 2 and PyHOP to synthesize executable plans. The system was validated with three categories of projects provided by two public universities in Colombia: Universidad Nacional de Colombia and Universidad de Caldas. The platform produces multiple alternative plans for each project and reports plan length and solution-search cost, enabling direct comparison across planners. Results show that the proposed workflow reduces manual scheduling effort, improves consistency of implementation roadmaps, and supports evidence-based selection of implementation strategies under different constraints. These capabilities help extension offices formalize knowledge, audit decisions, and reuse plans across initiatives. Traceability links BPMN elements to planning tasks.