Path Cover Problems with Length Cost
Lin, Guohui
DOI: https://doi.org/10.1007/s00453-023-01106-2
IF: 0.909
2023-03-07
Algorithmica
Abstract:For a graph , a collection of vertex-disjoint (simple) paths is called a path cover of G if every vertex is contained in exactly one path of . The Path Cover problem ( PC for short) is to find a minimum cardinality path cover of G . In this paper, we introduce generalizations of PC , where each path is associated with a weight (cost or profit). Our problem, Minimum (Maximum) Weighted Path Cover [ MinPC (MaxPC) ], is defined as follows: Let . Given a graph and a weight function that defines a weight for each path based on its length, the objective of MinPC (MaxPC) is to find a path cover of G such that the total weight of the paths in is minimized (maximized). Let L be a subset of U , and be the set of paths such that each path is of length . We consider Min PC with binary cost, i.e., the cost function is if ; otherwise, . We also consider Max PC with , if ; otherwise, . Many well-known graph theoretic problems such as the Hamiltonian Path and the Maximum Matching problems can be modeled using Min PC and Max PC . In this paper, we first show that deciding whether Min PC has a 0-weight solution is NP-complete for planar bipartite graphs of maximum degree three, and consequently, (i) for any constant , there is no polynomial-time approximation algorithm with approximation ratio for Min PC unless P NP, and (ii) Max PC is NP-hard for the same graph class. Next, we present a polynomial-time algorithm for Min PC on graphs with bounded treewidth for a fixed k . Lastly, we present a 4-approximation algorithm for Max PC , which becomes a 2.5-approximation algorithm for subcubic graphs.
computer science, software engineering,mathematics, applied