Competitive neural network to solve scheduling problems

Most scheduling problems have been demonstrated to be NP-complete problems. The Hopfield neural network is commonly applied to obtain an optimal solut ion in various different scheduling applications, such as the traveling sal esman problem (TSP), a typical discrete combinatorial problem. Hopfield neu ral networks, although providing rapid convergence to the solution, require extensive effort to determine coefficients. A competitive learning rule pr ovides a highly effective means of attaining a sound solution and can reduc e the effort of obtaining coefficients. Restated, the competitive mechanism reduces the network complexity. This important feature is applied to the H opfield neural network to derive a new technique, i.e. the competitive Hopf ield neural network technique. This investigation employs the competitive H opfield neural network to resolve a multiprocessor problem with no process migration, time constraints (execution time and deadline), and limited reso urces. Simulation results demonstrate that the competitive Hopfield neural network imposed on the proposed energy function ensures an appropriate appr oach to solving this class of scheduling problems. (C) 2001 Elsevier Scienc e B.V. All rights reserved.