Evidence-Grounded AI Coaching for Competency-Based Packet Tracer/GNS۳ Networking Labs

This paper presents Smart Networking Skills Lab, a web-based training and assessment environment that integrates network simulation (e.g., Cisco Packet Tracer and GNS۳) with an AI-driven coaching layer to support hands-on learning in "Network Equipment Installation and Setup" courses in technical high schools. The system collects learner artifacts-including configuration files, CLI outputs, topology snapshots, and packet captures (pcap)-and applies a hybrid approach that combines rule-based validation with retrieval-augmented AI feedback to (۱) detect common misconfigurations, (۲) generate step-by-step corrective guidance, and (۳) recommend targeted microlearning exercises aligned with competency-based rubrics. The proposed workflow operationalizes "practice → evidence capture automated diagnosis → personalized remediation" to reduce feedback latency and strengthen troubleshooting literacy. A formative evaluation design is proposed, including expert review of feedback accuracy and classroom-based measurement of time-to-competency, error frequency, and quality of troubleshooting paths. Results are expected to show improved practical performance, more consistent assessment, and higher learner confidence compared with conventional workshop instruction. The study contributes a scalable framework for Al-supported network-skills training that bridges simulation evidence and pedagogically grounded assessment in vocational education.

کلیدواژه ها:

vocational education ، networking skills ، AI coach ، intelligent tutoring systems ، learning analytics ، troubleshooting ، microlearning ، network simulation