An improved elastic net method for the traveling salesman problem

An elastic net method is presented for finding traveling-salesman tours; the method improves on the convergence properties of a recent model proposed by R. Durbin and D. Willshaw (1987). It combines the notion of symmetric forces described by the author in an earlier paper (1981) with an annealing schedule derived directly from the input data. The author demonstrates that the method converges about 50 times faster than the Durbin-Willshaw model on patterns of up to 50 cities. It produces either optimum tours or tours within a few percent of optimum. Several search strategies for larger problems are suggested.<>