Assessing VR-Enhanced Technician Training in a Manufacturing Plant

Thesis Abstract

The study investigates the effectiveness of virtual reality (VR)-enhanced technician training within a large-scale automotive parts manufacturing plant, addressing persistent gaps in on-the-job performance, safety incident rates, and training turnaround times observed after traditional instruction approaches. The core aim is to determine whether VR-based simulation and tactile feedback can improve technical competence, safety adherence, and operational efficiency compared with conventional training modalities. Specific objectives include (1) evaluating changes in procedural accuracy and time-to-competency for maintenance technicians; (2) assessing safety knowledge retention and incident exposure during post-training tasks; (3) examining trainee engagement, cognitive load, and perceived realism of VR environments; (4) identifying organizational factors that influence VR training adoption, including supervision quality and access to equipment; and (5) estimating the return on investment (ROI) and cost-effectiveness of implementing VR training at scale. The research adopts a mixed-methods design anchored in a quasi-experimental framework complemented by qualitative interviews. The population comprises all maintenance technicians employed at the plant over a 24-month period, with a sample of 120 technicians who undergo VR-enhanced training and 120 matched controls receiving standard instruction. The sampling strategy combines purposive selection for VR cohorts and propensity score matching to ensure equivalence on demographic variables, prior experience, and baseline competency. Data collection instruments include standardized performance checklists aligned with ISO 45001 safety requirements, time-to-competency metrics from training records, pre- and post-intervention knowledge tests, validated cognitive-load scales, and semi-structured interview guides. Instrument validity is established through content validity reviews by a panel of subject-matter experts and pilot testing with 20 technicians. Reliability checks include Cronbach’s alpha for internal consistency of knowledge tests and inter-rater reliability (Cohen’s kappa) for practical performance assessments. Quantitative analysis employs multivariate analysis of covariance (MANCOVA) to compare post-training outcomes between VR and traditional groups while controlling for covariates such as prior experience and age. Mediation analysis investigates whether cognitive load and perceived realism mediate training effectiveness. A difference-in-differences (DiD) approach evaluates changes in safety incidents, maintenance task durations, and error rates pre- and post-intervention. The qualitative component utilizes thematic analysis of interview transcripts to extract insights on experiential realism, user acceptance, and organizational enablers and barriers to VR integration. Triangulation is achieved by comparing qualitative themes with quantitative results to illuminate convergent and divergent evidence. Expected findings indicate significant improvements in procedural accuracy, faster time-to-competency, and higher safety knowledge retention for the VR group relative to controls, with moderate to large effect sizes (partial eta-squared estimates in the 0.10–0.25 range). VR training is anticipated to reduce mean task completion times by 15–25% and decrease reportable safety incidents by 20–30% within six months post-training. The cognitive load is expected to be managed effectively when VR scenarios incorporate adaptive difficulty and haptic feedback, with trainee engagement positively moderating performance gains. The study also anticipates identifying organizational factors—such as leadership support, technician-to-trainer ratio, and equipment availability—that influence VR efficacy and sustainability. The study contributes to knowledge by providing empirical evidence on the comparative effectiveness and cost-effectiveness of VR-enhanced technician training in a real-world manufacturing setting, integrating concepts from experiential learning theory, social constructivism, and cognitive load theory. It offers a concrete implementation blueprint for scaling VR training, including criteria for scenario development, learner progression metrics, and governance structures for technology adoption. The main conclusion posits that VR-enhanced training yields superior competence and safety outcomes when embedded within a supportive learning ecosystem and aligned with organizational goals. Recommendations include investing in modular VR content aligned to critical maintenance procedures, establishing standardized evaluation protocols, ensuring ongoing trainer facilitation, and conducting longitudinal ROI analyses to guide broader deployment across the manufacturing network.