A systematic method of remaining useful life estimation based on physics-informed graph neural networks with multisensor data.

•Quantitative guidance for graph-structured data preprocessing is proposed.•Deep learning structure based on ARMAGCN and GRU considering the interactions among multisensor signals with a graph structure is developed.•Physics-informed loss function is proposed to train neural networks, specifically for RUL estimation.•Risk eliminations and high accuracy parallelled for data-driven method in RUL estimation.