Hybrid recommendations and dynamic authoring for AR knowledge capture and re-use in diagnosis applications

In Industry 4.0, integrated data management is an important challenge due to heterogeneity and the lack of structure of numerous existing data sources. A relevant research gap involves human knowledge integration, especially in maintenance operations. Augmented Reality (AR) can bridge this gap, but it requires improved augmented content to enable effective and efficient knowledge capture. This paper proposes dynamic authoring and hybrid recommender methods for accurate AR-based reporting. These methods aim to provide maintainers with augmented data input formats and recommended datasets for enhancing the efficiency and effectiveness of their reporting tasks. The proposed contributions have been validated through experiments and surveys in two failure diagnosis reporting scenarios. Experimental results indicated that the proposed reporting solution can reduce reporting errors by 50% and reporting time by 20% compared to alternative recommender and AR tools. Besides, survey results suggested that testers perceived the proposed reporting solution as more effective and satisfactory for reporting tasks than alternative tools. Thus, proving that the proposed methods can improve the effectiveness and efficiency of diagnosis reporting applications. Finally, this paper proposes future works towards a framework for automatic adaptive authoring in AR knowledge transfer and capture applications for human knowledge integration in the context of Industry 4.0.