A research team from the University of Waterloo has announced a breakthrough in solving the traveling salesman problem (TSP), which involves calculating the shortest distance between various bars in South Korea. The TSP problem in South Korea consisted of a staggering 81,998 bars, making it the largest TSP problem ever solved to the highest level of optimality.
The calculation revealed that without any breaks, it would take 15,386,177 seconds or 178 days, 1 hour, 56 minutes, and 17 seconds to complete the journey. However, factoring in breaks and stops at different bars would extend the time to multiple years. The team utilized 48 cores of the Intel Xeon Gold 6238 CPU for the computations, starting with an initial path and employing a branch-and-bound search process to gradually optimize the route.
The TSP problem is an NP-hard problem, becoming increasingly difficult to solve rapidly as the size of the problem grows. A straightforward calculation for a TSP problem with 22 points could take thousands of years. Fortunately, there are algorithms that significantly improve optimization, allowing us to tackle larger TSP problems within a feasible time frame. The previous largest TSP problem solved was visiting 57,912 landmarks in the Netherlands.
Source: University of Waterloo
TLDR: A team from the University of Waterloo successfully tackled the largest traveling salesman problem in South Korea with nearly 82,000 bars, utilizing advanced computing power to optimize the shortest route and achieving remarkable results.
Leave a Comment