TY - GEN
T1 - A lagrangian relaxation based heuristic for solving the length-balanced two arc-disjoint shortest paths problem
AU - Li, Yanzhi
AU - Lim, Andrew
AU - Ma, Hong
PY - 2005
Y1 - 2005
N2 - We consider a HAZMAT transportation problem, which is modeled as the length-balanced two arc-disjoint shortest paths problem (LB2SP). The objective function of LB2SP is expressed as a weighted sum of two terms, i.e., the sum of the path lengths and the positive length difference between the paths. We demonstrate that LB2SP is NP-Hard, and formulate it as an Integer Programming (IP) model. We develop a Lagrangian relaxation based heuristic (LRBH) for LB2SP. Computational experiments are conducted to compare the performance of LRBH with the CPLEX solver, showing that the LRBH is efficient for LB2SP. © Springer-Verlag Berlin Heidelberg 2005.
AB - We consider a HAZMAT transportation problem, which is modeled as the length-balanced two arc-disjoint shortest paths problem (LB2SP). The objective function of LB2SP is expressed as a weighted sum of two terms, i.e., the sum of the path lengths and the positive length difference between the paths. We demonstrate that LB2SP is NP-Hard, and formulate it as an Integer Programming (IP) model. We develop a Lagrangian relaxation based heuristic (LRBH) for LB2SP. Computational experiments are conducted to compare the performance of LRBH with the CPLEX solver, showing that the LRBH is efficient for LB2SP. © Springer-Verlag Berlin Heidelberg 2005.
UR - http://www.scopus.com/inward/record.url?scp=33745623978&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-33745623978&origin=recordpage
U2 - 10.1007/11589990_196
DO - 10.1007/11589990_196
M3 - 32_Refereed conference paper (with ISBN/ISSN)
SN - 3540304622
SN - 9783540304623
VL - 3809 LNAI
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 1323
EP - 1326
BT - AI 2005: Advances in Artificial Intelligence
PB - Springer Verlag
ER -