论文信息 - Parallel Algorithms for Some Optimization Problems Arising in Air Pollution Modeling

Parallel Algorithms for Some Optimization Problems Arising in Air Pollution Modeling

We will consider some optimization problems arising in air pollution modeling. An analogy with similar optimization problems, which can be formulated as Quadratic Optimization and Mixed Integer Linear Programming problems and arising in placement of components on a Printed Circuit Board (PCB) under different geometrical and technological constraints, will be made. We will also describe an efficient parallel Branch and Bound (B&B) algorithms to solve these MILP problems in a MIMD distributed memory environment. The algorithms are scalable and run on a cluster of workstations under PVM. The efficiency of the algorithms has been investigated and the test results on some placement tasks and from MIPLIB library are presented.

V. N. Alexandrov | J. L. Alonso | V. Alexandrov

[1] Per S. Laursen. Simple Approches to Parallel Branch and Bound , 1993, Parallel Comput..

[2] Vassil N. Alexandrov,et al. Parallel Branch and Bound Algorithms for Integer and Mixed Integer Linear Programming Problems under PVM , 1997, PVM/MPI.

[3] Panos M. Pardalos,et al. SCOOP: Solving Combinatorial Optimization Problems in Parallel , 1996, Solving Combinatorial Optimization Problems in Parallel.

[4] Laurence A. Wolsey,et al. Integer and Combinatorial Optimization , 1988 .

[5] William J. Cook,et al. Computational experience with parallel mixed integerprogramming in a distributed environment , 1999, Ann. Oper. Res..

[6] Djangir A. Babayev,et al. Reducing the number of variables in Integer and Linear Programming Problems , 1994, Comput. Optim. Appl..

[7] Bernard Gendron,et al. Parallel Branch-and-Branch Algorithms: Survey and Synthesis , 1994, Oper. Res..

[8] Georg Sigl,et al. GORDIAN: VLSI placement by quadratic programming and slicing optimization , 1991, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..