In this paper, we introduce a secondary evacuation planning problem which is solved by computing flow-dependent shortest path through a known shortest time escape route on a ship. Given a primary escape route, our secondary evacuation planning problem consists in finding the second-shortest escape route, based on the cabin network rebuilding. We suggest a new model for flow-dependent network where the travel time of each link depends on the flow speed and the flow speed depends on the total number of evacuees traversing the link. The model focused on the average evacuation time to travel through the ordered sequence of consecutive arcs by each group, and we proposed a heuristic algorithm to produce sub-optimal secondary evacuation plan. A numerical example is given at last. Results showing that our algorithm can effective supply the flow-dependent network k-shortest path need in reasonable computation times.