The invention proposes a lithium-ion battery URL prediction method based on Box-Cox transformation and Monte-Carlo simulation, and the method comprises the steps: employing the Box-Cox transformation to transform the capacity of a battery, constructing a linear model between a capacity transformation value and the number of cycles, and carrying out the recognition of model parameters and model uncertainty through the least square algorithm, wherein the uncertainty of the residual life is generated through the Monte-Carlo simulation. The method can reduce online aging data needed by the precise residual life prediction, wherein the online aging data needed by the precise residual life prediction is just at least 30% of the total attenuation data size of the battery.