A multitask context-aware approach for design lesson-learned knowledge recommendation in collaborative product design

To proactively assist engineers in finding and reusing massive design lesson-learned knowledge (DLK), knowledge recommendation has become a key technology of knowledge management. However, in collaborative product design, complex multitask context information disrupts the perception of engineers’ knowledge needs for every single task. In this situation, traditional knowledge recommendation approach is prone to provide a mixed DLK recommendation list, thus resulting in a lack of pertinence and low accuracy. Facing these challenges, scarcely any reports on context-aware knowledge recommendation in the multitask environment of collaborative product design. Aiming to fill this gap, a multitask context-aware DLK recommendation approach is proposed to assist collaborative product design in a smarter manner. The mutual interference of context information from different tasks is addressed by preprocessing works, multitask knowledge need awareness, DLK recommendation engine, respectively. Therefore, the proposed approach not only effectively acquires engineers’ knowledge needs from different task contexts and pertinently provides the corresponding DLK recommendation list for each task but also guarantees the accuracy of DLK recommendation in multitask context of collaborative product design. To validate the proposed approach, a DLK recommendation system is implemented in a shipbuilding scenario, and some comparative experiments are carried out. Experimental results show that the proposed approach outperforms conventional approaches in the aspects of effectiveness and performance. Therefore, it opens up a promising way to help engineers reuse needed DLK in collaborative product design.