作者: Chandra Chekuri , Sanjeev Khanna
关键词: Polynomial-time approximation scheme 、 Open problem 、 Approximation algorithm 、 Algorithm 、 Computer science 、 Dynamic programming 、 Completion time 、 Scheduling (computing) 、 Time complexity 、 Weighting 、 Mathematical optimization
摘要: We consider the well known problem of scheduling jobs with release dates to minimize their average weighted completion time. When multiple machines are available, machine environment may range from identical (the processing time required by a job is invariant across machines) at one end spectrum unrelated on each specified an arbitrary vector) other end. While strongly NP-hard even in case single machine, constant factor approximation algorithms for most general machines. Recently PTAS was discovered parallel [1]. In contrast, MAX SNP-hard [11]. An important open determine approximability intermediate uniformly related where has speed and it takes p/s process size p s. resolve complexity this obtaining PTAS. This improves earlier ratio (2 + Ɛ).