How to set up and solve the "Nonfixed destination Multiple Depot, multiple TSP (MDmTSP)" Optimization Problem using ortools?

[Hayford K Adjavor]

Hi there,

First of, I am new to the or-tools API for Python package. But I am trying to solve the Multiple Depot, Multiple Traveling Salesman Problem (MDmTSP), a variant of the traditional Traveling Salesman Problem (TSP). The TSP is also akin to the traditional Vehicle Routing Problem (VRP).

My expected solution would look something like this VRP Pickups and Deliveries example - https://developers.google.com/optimization/routing/pickup_delivery - except that there are multiple number of depots and multiple number of vehicles (in my case salesmen) involved. And, the the vehicles (or salesmen) need not return to the originating depot after the end of their tour(s).

The "Non-fixed MDmTSP" mathematical model itself is modeled upon the one found in this paper at - https://www.sciencedirect.com/science/article/pii/S0377221705003188

I am hoping that such a solution method may already exist that someone could point me to, which I could use as an example to build my own.

Thanks in advance.