Branch‐and‐bound approach for optima localization in scheduling multiprocessor jobs
作者: Alexander KononovPolina KononovaAlexander Gordeev
作者单位: 1Temporary Research Group “Optimization” Sobolev Institute of Mathematics 4, Akad. Koptyug Avenue 630090 Novosibirsk Russia
2 Department of Mechanics and Mathematics Novosibirsk State University 2, Pirogova Str. 630090 Novosibirsk Russia
刊名: International Transactions in Operational Research, 2020, Vol.27 (1), pp.381-393
来源数据库: Wiley Journal
DOI: 10.1111/itor.12503
关键词: Multiprocessor jobsBranch‐and‐bound algorithmOptima localization
原始语种摘要: Abstract(#br)We consider multiprocessor task scheduling problems with dedicated processors. We determine the tight optima localization intervals for different subproblems of the basic problem. Based on the ideas of a computer‐aided technique developed by Sevastianov and Tchernykh for shop scheduling problems, we elaborate a similar method for the multiprocessor task scheduling problem. Our method allows us to find an upper bound for the length of the optimal schedule in terms of natural lower bound. As a byproduct of our results, a family of linear‐time approximation algorithms with a constant ratio performance guarantee is designed for the NP‐hard subproblems of the basic problem, and new polynomially solvable classes of problems are found.
全文获取路径: Wiley  (合作)

  •  分公司分支
  • multiprocessor 多处理机
  • Branch 分公司分支
  • and 分公司分支
  • scheduling 
  • localization 定位
  • bound 
  • schedule 负载图表
  • computer 电子计算机
  • technique 技术
  • dedicated 专用