We study the problem of constructing a novel framework for dynamically organizing mobile nodes in wireless ad-hoc networks into clusters where it is necessary to provide robustness in the face of topological changes caused by node motion, node failure and node insertion/removal. The main contribution of our work is a new strategy for clustering a wireless AD HOC network and improvements in WCA (Weighted Clustering Algorithm) [3], a well-known algorithm. We first derive mathematically a new cluster size bound to replace δ in WCA and a simple node stability model. Thereafter, we prove their efficiencies. Our contribution also extends previous work to replace the degree- difference used initially in WCA to provide load-balance in wireless AD HOC with a new more efficient and consistent model which helps to decrease the number of clusters. We show that our algorithm outperforms WCA in terms of cluster formation and stability. The non-periodic procedure for clusterhead election is invoked on-demand, and is aimed to reduce the computation and communication costs. We strive to provide a trade-off between the uniformity of the load handled by the clusterheads and the connectivity of the network.