Computer codes for the TSP have become increasingly more sophisticated over the years. A conspicuous sign of these improvements is the increasing size of nontrivial instances that have been solved, moving from Dantzig, Fulkerson, and Johnson's solution of a 49-city problem in 1954 up through the solution of a 85,900-point problem 52 years later.
Year | Research Team | Size of Instance | Name | |
1954 | G. Dantzig, R. Fulkerson, and S. Johnson | 49 cities | dantzig42 | |
1971 | M. Held and R.M. Karp | 64 cities | 64 random points | |
1975 | P.M. Camerini, L. Fratta, and F. Maffioli | 67 cities | 67 random points | |
1977 | M. Grötschel | 120 cities | gr120 | |
1980 | H. Crowder and M.W. Padberg | 318 cities | lin318 | |
1987 | M. Padberg and G. Rinaldi | 532 cities | att532 | |
1987 | M. Grötschel and O. Holland | 666 cities | gr666 | |
1987 | M. Padberg and G. Rinaldi | 2,392 cities | pr2392 | |
1994 | D. Applegate, R. Bixby, V. Chvátal, and W. Cook | 7,397 cities | pla7397 | |
1998 | D. Applegate, R. Bixby, V. Chvátal, and W. Cook | 13,509 cities | usa13509 | |
2001 | D. Applegate, R. Bixby, V. Chvátal, and W. Cook | 15,112 cities | d15112 | |
2004 | D. Applegate, R. Bixby, V. Chvátal, W. Cook, and K. Helsgaun |
24,978 cities | sw24798 | |
2006 | D. Applegate, R. Bixby, V. Chvátal, W. Cook, Daniel Espinoza, Marcos Goycoolea, and K. Helsgaun |
85,900 cities | pla85900 |