Coalition structure generation with given required bound based on coalition combination

Coalition formation is a key topic in multi-agent systems. Most of these researches concentrate on how to form a coalition and assign the revenues of the coalition through negotiation. Another researching branch is to study the optimal division of agents into coalitions (the pairwise disjoint subsets) so that the sum of the revenues of all coalitions is maximal. One may prefer a coalition structure that maximizes the sum of the values of the coalitions, but often the number of coalition structures is too large to allow exhaustive search for the optimal one. When practical applications can present required real bound on the worst case, and how to attain this demand via partial search? This paper analyzes the relations among coalition structures in depth and presents on a novel algorithm based on coalition combination: the bound K(n)¿2 can be attained with searching of those coalition structures whose cardinality structure is in the coalition combination set CCs(n, k) . Finally, experiments show that the new algorithm is obviously better than existing algorithms.