CFLP-Instance-Generator

Published: 1 August 2018| Version 1 | DOI: 10.17632/vhst7gvh7j.1
Contributor:
Andreas Klose

Description

C-code for generating test problem instances for the Capacitated Facility Location Problem according to a proposal made by Cornuejols et al. (1991). The test instances where used in Klose and Goertz (2007) and Goertz and Klose (2012). References: Cornuejols G, Sridharan R, Thizy J-M (1991) A Comparison of Heuristics and Relaxations for the Capacitated Plant Location Problem. Eur J Oper Res 50:280-297. doi: 10.1016/0377-2217(91)90261-S Klose A, Goertz S (2007) A branch-and-price algorithm for the capacitated facility location problem. Eur J Oper Res 179:1109-1125. doi: 10.1016/j.ejor.2005.03.078 Goertz S, Klose A (2012) A simple but usually fast branch-and-bound algorithm for the capacitated facility location problem. INFORMS J Computing 24:597-610. doi: 10.1287/ijoc.1110.0468

Files

Steps to reproduce

After compiling the code, the program (named gencflp) can be used from a terminal by typing: gencflp "inputfile" "path" where: "inputfile" is an ascii file providing information about the test problem instances which "gencflp" should generate. This file must provide the following infomation: (1) Seed, i.e. an integer number specifying the seed number for the random number generator (if Seed=0 a seed number is generated automatically) (2) For every problem class: #customers #depot sites ratio problem name where "ratio" is the desired ratio of total capacity to total demand. "path" is an optional argument specifying the output path where the generated data files should be stored. Example: gencflp gencflp.input1 (generates the test instances from Klose and Goertz, 2007) gencflp gencflp.input2 (generates the test instances from Goertz and Klose, 2012)

Institutions

Aarhus Universitet Science and Technology

Categories

Operations Research, Nonlinear Programming, Mixed Integer Programming, Facility Location

Licence