论文信息 - Fast neighborhood search for the single machine total weighted tardiness problem

Fast neighborhood search for the single machine total weighted tardiness problem

Most successful heuristics for solving 1||@?w"jT"j are based on swap moves. We present an algorithm which improves the complexity of searching the swap neighborhood from O(n^3) to O(n^2). We show that this result also improves the complexity of the recently developed dynasearch heuristics.

James B. Orlin | Özlem Ergun

[1] Chris N. Potts,et al. A Branch and Bound Algorithm for the Total Weighted Tardiness Problem , 1985, Oper. Res..

[2] James B. Orlin,et al. New neighborhood search algorithms based on exponentially large neighborhoods , 2001 .

[3] E. Lawler. A “Pseudopolynomial” Algorithm for Sequencing Jobs to Minimize Total Tardiness , 1977 .

[4] Roberto Tadei,et al. An enhanced dynasearch neighborhood for the single-machine total weighted tardiness scheduling problem , 2004, Oper. Res. Lett..

[5] Chris N. Potts,et al. Dynasearch - interative local improvement by dynamic programming: Part I, The traveling salesman problem , 1995 .

[6] Chris N. Potts,et al. An Iterated Dynasearch Algorithm for the Single-Machine Total Weighted Tardiness Scheduling Problem , 2002, INFORMS J. Comput..

[7] Jan Karel Lenstra,et al. Complexity of machine scheduling problems , 1975 .