Logistics Route Optimizer

The problem

A delivery truck in the Mahikeng area of the North West needs the cheapest route between towns. Trying every path wastes time; a plain shortest-path search ignores what is known about the road ahead. Built for an Artificial Intelligence module.

How it works

The network is modelled as a graph of 13 towns and 17 roads — Mahikeng, Mmabatho, Lichtenburg, Zeerust, Coligny and others — with the distance of each road in kilometres.

A* search ranks every candidate by f(n) = g(n) + h(n):

• g(n) — the real distance travelled so far,
• h(n) — a traffic-based estimate of the cost still to go,

and always expands the most promising town next, using a priority queue (heapq). It returns the optimal path, the order in which towns were explored and the total cost.

Verification

A separate test case replays the found route as a simulated drive, and a quick verification script re-runs the search without the animation.