In order to solve the problem of less target, static circumstance and real situation cannot be simulated in traditional path planning, under the background of the time-varying conditions, cost, environmental risk and population risk were considered in the multi-objective path optimization and selection. A mathematical model of multi-objective shortest path selection problem with time-varying conditions and multiple departure time (segmentation time) for hazardous goods transportation was established and designed. Considering the complexity of the actual decision-making environment and the limitation of the rationality of the decision-makers， the algorithm of comprehensive k-shortest path and approximating ideal solution sorting method (TOPSIS method) was proposed. The effective path of each single target was determined by k-shortest path algorithm. Then, the effective path to the positive and negative ideal solutions was calculated by TOPSIS method. Finally, an application example was given to prove the effectiveness of the proposed model.