A PROCEDURE FOR CALCULATING THE WEIGHT-MATRIX OF A NEURAL-NETWORK FORRESOURCE LEVELING

In this paper, a new approach for the computation of the weight-matrix of a neural network (NN) for resource leveling (RL) is introduced. Th e proposed method achieves significantly improved efficiency over the conventional technique of employing the functional expressions of the weights, by exploiting the structural properties of the matrices arisi ng in the formulation of the RL problem as a quadratic zero-one optimi zation. These structural properties are identified, and stated in term s of template-matrix contributions of the cost and constraint function s of the quadratic optimization, to the weight-matrix of the NN. It is shown that by using these templates, the weight-matrix can be filled- in directly, based on the early start schedule of a project. (C) 1997 Elsevier Science Limited.