A simple on-line bin-packing algorithm
Journal of the ACM (JACM)
Simulated annealing and Boltzmann machines: a stochastic approach to combinatorial optimization and neural computing
Information processing in dynamical systems: foundations of harmony theory
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Bounded space on-line bin packing: best is better than first
SODA '91 Proceedings of the second annual ACM-SIAM symposium on Discrete algorithms
Improved space for bounded-space, on-line bin-packing
SIAM Journal on Discrete Mathematics
Approximation algorithms for bin packing: a survey
Approximation algorithms for NP-hard problems
An efficient processor allocation algorithm using two-dimensional packing
Journal of Parallel and Distributed Computing
Heuristic Solution of Open Bin Packing Problems
Journal of Heuristics
A consensus-function artificial neural network for map-coloring
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.00 |
An incremental approach to bin packing is proposed. A harmony theory artificial neural network is employed, whose two-layer architecture prompts the explicit encoding of the allowable placements of the objects in the bins as well the constraints arising from each placement. As a result, bin packing instances of any dimension can be solved, while the restrictions which usually apply to bin packing - and concern fixed object orientation, uniform bin capacity and the relation between bin capacity and maximum object volume - are relaxed. Furthermore, the computationally expensive task of sorting the objects in descending order is not performed. The proposed solutions suggest the exact placements of the objects in the bins. For appropriate parameter values of the harmony theory network, the smallest number of bins required for packing all the objects (i.e. an optimal solution) is consistently determined, while all optimal solutions are settled upon with asymptotically equal probability.