In response to the issue of insufficient demand for logistics facilities during holidays or the temporary or permanent closure of logistics facilities due to sudden events such as epidemics, natural disasters, etc., this paper proposes a planning model and solution algorithm based on multi-objective dynamic decision-making to optimize the emergency transport plan and structure of the e-commerce logistics network. Firstly, an LSTM model is utilized to predict the transportation demand for a certain period in the future. Decision variables are introduced to indicate whether to open or close new routes after closing a logistics facility. The objective functions are set to minimize the number of routes with changes in cargo volume, minimize the load imbalance of each route, and minimize the cumulative total of parcels that cannot circulate normally. Considering factors such as normal circulation of parcels, circulation capacity, transportation capacity after opening or closing facilities, and time constraints, a multi-objective planning model is constructed. Finally, the simulated annealing algorithm is employed for solution, obtaining a circulation plan that meets the requirements. Simulation experiment results demonstrate that the model can rapidly and accurately respond to the paralysis of the e-commerce logistics network caused by sudden events, showcasing high efficiency and practical application value.