Scheduling meeting solved by neural network and Min-conflict heuristic
Bouhouch Adil, Chakir Loqman, Abderrahim El Qadi · 2016
The Meeting scheduling problem (MSP) is an important vehicle of communication in business, factories and team works. MSP requires a careful balance between the individual personal preferences and the organization. In this paper we propose a new approach to solve meeting scheduling problem. The proposed network combines the characteristics of neural networks and conflicts minimization approach. This approach is divided into three steps: The first concerns formulating a meeting scheduling problem as a CSP problem, then we reformulate this CSP as a quadratic problem under linear constraint(QP). The second step applies the continuous Hopfield network to solve the QP. The later step involves the Min-conflict heuristic to improve the solution given by the second step. The performance of the proposed approach with direct approach which uses CHN alone are compared over some generated meeting instances.