I crate a file that represents 16000 edges with information of start node, end node and cost in a table.
In MATLAB i want to solve shortest path with PSO algorithm. How i can solve it?
Is there a particular reason why you want to use particle swarm optimization in this case?
If what you really want is to find the least-cost paths, you may consider using Dijkstra's algorithm. Here is one possible implementation, written in MATLAB (there are many): https://www.mathworks.com/matlabcentral/fileexchange/20025-dijkstra-s-minimum-cost-path-algorithm
As you only have 16000 edges, this is likely to be fast enough.
If it is not fast enough, you'll need an optimized, compiled implementation of Dijkstra's algorithm. I am not aware of any such implementations for MATLAB, though evidently there are some for Python (see my question here: C-based implementation of Djikstra's or Fast Marching for QGIS/Python?).