The weighted supermatrix comes out from the combination of the unweighted supermatrix and the control hierarchy. The latter scores the priority of a cluster over all the clusters to which it is connected: therefore, the control hierarchy is an n*n matrix, with n=number of clusters in the network. To build the control hierarchy matrix, first of all you choose a cluster Ci. Then, all other clusters connected with Ci are pairwise compared (with AHP) to determine their impact on Ci; their weights are listed in the control hierarchy matrix. For all the clusters that are not connected to Ci the corresponding element in the control hierarchy is set to 0.
Now, all elements in the block corresponding to the intersection between cluster Ci and cluster Cj in the unweighted supermatrix can be multiplied by the weight of Ci over Cj listed in the control hierarchy matrix. In this way, you can obtain your weighted supermatrix.
1- Saaty, T. L. (2004). Decision making—the analytic hierarchy and network processes (AHP/ANP). Journal of systems science and systems engineering, 13(1), 1-35.
2- Jharkharia, S., & Shankar, R. (2007). Selection of logistics service provider: An analytic network process (ANP) approach. Omega, 35(3), 274-289.
3- Saaty, T. L. (2001). Analytic network process. In Encyclopedia of Operations Research and Management Science (pp. 28-35). Springer US.