Search

We present constant factor approximation algorithms for the following two problems: First, given a connected graph $G=(V,E)$ with non-negative edge weights, find a minimum weight spanning tree that respects prescribed upper bounds on the vertex degrees. Second, given prescribed (exact) vertex degrees $d=(d_i)_{i \in V}$, find a minimum weight connected $d$-factor. Constant factor approximation algorithms for these problems were known only for the case that $d_i \geq 2$ for all $i \in V$.

Finding low-cost spanning subgraphs with given degree and connectivity requirements is a fundamental problem in the area of network design. We consider the problem of finding d-regular spanning subgraphs (or d-factors) of minimum weight with connectivity requirements. For the case of k-edge-connectedness, we present approximation algorithms that achieve constant approximation ratios for all d≥2⋅⌈k/2⌉. For the case of k-vertex-connectedness, we achieve constant approximation ratios for d≥2k−1. Our algorithms also work for arbitrary degree sequences if the minimum degree is at least 2⋅⌈k/2⌉ (for k-edge-connectivity) or 2k−1 (for k-vertex-connectivity). To complement our approximation algorithms, we prove that the problem with simple connectivity cannot be approximated better than the traveling salesman problem. In particular, the problem is APX-hard.

International audience; The concept of Reload cost in a graph refers to the cost that occurs while traversing a vertex via two of its incident edges. This cost is uniquely determined by the colors of the two edges. This concept has various applications in transportation networks, communication networks, and energy distribution networks. Various problems using this model are defined and studied in the literature. The problem of finding a spanning tree whose diameter with respect to the reload costs is the smallest possible, the problems of finding a path, trail or walk with minimum total reload cost between two given vertices, problems about finding a proper edge coloring of a graph such that the total reload cost is minimized, the problem of finding a spanning tree such that the sum of the reload costs of all paths between all pairs of vertices is minimized, and the problem of finding a set of cycles of minimum reload cost, that cover all the vertices of a graph, are examples of such problems. In this work we focus on the last problem. Noting that a cycle cover of a graph is a 2-factor of it, we generalize the problem to that of finding an r -factor of minimum reload cost of an edge colored graph. We prove several $NP$-hardness results for special cases of the problem. Namely, bounded degree graphs, planar graphs, bounded total cost, and bounded number of distinct costs. For the special case of r =2 , our results imply an improved $NP$-hardness result. On the positive side, we present a polynomial-time solvable special case which provides a tight boundary between the polynomial and hard cases in terms of r and the maximum degree of the graph. We then investigate the parameterized complexity of the problem, prove W[1]-hardness results and present an FPT-algorithm.

Every 9-regular graph (possibly with multiple edges) with odd edge-connectivity >5 can be edge-decomposed into three 3-factors. If Tutte's 3-flow conjecture is true, it also holds for all 9-regular graphs with odd edge-connectivity 5, but not with odd edge-connectivity 3. It holds for all planar 2-edge-connected 9-regular graphs, an equivalent version of the 4-color theorem for planar graphs. We address the more general question: If G is an r-regular graph, and r = k q where k , q are natural numbers >1, can G be edge-decomposed into k q-factors? If q is even, then the decomposition exists trivially. If k , q are both odd, then we prove that the decomposition exists if G has odd edge-connectivity (size of smallest odd edge-cut) at least 3 k − 2 , which is satisfied if the odd edge-connectivity is at least r − 2 . If q is odd and k is even, then we must require that G has an even number of vertices just to guarantee that G has a q-factor. If we want a decomposition into q-factors, then we also need the condition that, for any partition of the vertex set of G into two odd parts, there must be at least k edges between the parts. We prove that the edge-decomposition into q-factors is always possible if G has an even number of vertices and the edge-connectivity of G is at least 2 k 2 + k .

An (r − 1, 1)-coloring of an r-regular graph G is an edge coloring (with arbitrarily many colors) such that each vertex is incident to r − 1 edges of one color and 1 edge of a different color. In this paper, we completely characterize all 4-regular pseudographs (graphs that may contain parallel edges and loops) which do not have a (3, 1)-coloring. Also, for each r ≥ 6 we construct graphs that are not (r −1, 1)-colorable and, more generally, are not (r − t, t)-colorable for small t.