The cost-to-time ratio problem for large or infinite graphs

Abstract It is an important problem in graph theory and its applications to find the cycles and paths of maximal or minimal cost-to-time ratio in a doubly weighted directed graph. This problem is equivalent to the solution of certain systems of linear inequalities. An algorithm is suggested for the determination of the maximal cost-to-time ratio which works for finite and infinite graphs. It is based on the famous Ford-Bellman algorithm and its extensions by Karp [9] and the author [6;7]. The computational properties of our algorithm mainly depend on the maximal number of arcs of the simple paths, not on the number of vertices of the graph.