Approximation algorithms for solving the heterogeneous Chinese postman problem
Jianping Li,Pengxiang Pan,Junran Lichen,Lijian Cai,Wencheng Wang,Suding Liu
DOI: https://doi.org/10.1007/s10878-022-00931-5
2022-11-25
Journal of Combinatorial Optimization
Abstract:In this paper, we consider the heterogeneous Chinese postman problem (the HCPP), which generalizes the k-Chinese postman problem. Specifically, given a weighted graph G=(V,E;w;r)documentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$G=(V,E;w;r)$$end{document} with length function w:E→R+documentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$w:E
ightarrow R^{+}$$end{document} satisfying the triangle inequality, a fixed depot r∈Vdocumentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$rin V$$end{document}, and k vehicles having k nonuniform speeds λ1,λ2,…,λkdocumentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$lambda _{1}, lambda _{2}, ldots ,lambda _{k}$$end{document}, respectively, it is asked to find k tours in G for these k vehicles, each starting at the same depot r, and collectively traversing each edge in E at least once. The objective is to minimize the maximum completion time of vehicles, where the completion time of a vehicle is its total travel length divided by its speed. The main contribution of our paper is to show the following two results. (1) Given any small constant δ>0documentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$delta >0$$end{document}, we design a 20.8765(1+δ)documentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$20.8765(1+delta )$$end{document}-approximation algorithm to solve the HCPP, where the running time required is bounded by a polynomial in the input size and 1δdocumentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$frac{1}{delta }$$end{document}. (2) We present a (1+Δ-1/k)documentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$(1+varDelta -1/k)$$end{document}-approximation algorithm to solve the HCPP in cubic time, where Δdocumentclass[12pt]{minimal}usepackage{amsmath}usepackage{wasysym}usepackage{amsfonts}usepackage{amssymb}usepackage{amsbsy}usepackage{mathrsfs}usepackage{upgreek}setlength{oddsidemargin}{-69pt}egin{document}$$varDelta $$end{document} is the ratio of the largest vehicle speed to the smallest one.
mathematics, applied,computer science, interdisciplinary applications