Content uploaded by Amir Reza Mahdavi
Author content
All content in this area was uploaded by Amir Reza Mahdavi on Jun 25, 2019
Content may be subject to copyright.
1
modares.ac.ir@amirrezamahdavi
armamdoohi@modares.ac.ir
Grove School of EngineeringCity College of New Yorkmallahviranloo@ccny.cuny.edu
2
3
G(N,E)N
E
i
1
m
2
m
3
m
4
m
m1
m2
m3
m4
400 m
600 m
800 m
4
i
yi
ij
x
ij
i
C
ij
c
iijij
ij
d
min
l
max
l
i
piTS
N
max Z= ∑piyi(∑Ciyi+ ∑Cij𝑑𝑖𝑗xij)
⁄
s.t.
∑xij = ∑yi-1
∑xiji,j ∈S ≤ ∑yi
i ∈S / {k} k∈S⊆N, |S|≥2
lmin ≤ dij ≤ lmax
yi=1 i∈T⊆N
yi=0 , 1 i∈N
xij=0 , 1 (i,j)∈E
max Z= ∑piyi(∑Ciyi+ ∑Cij𝑑𝑖𝑗xij)
⁄
5
s.t.
∑𝑥𝑖𝑗𝑖 ∈𝑁/{𝑖} = yi
∑𝑥𝑖𝑗𝑗 ∈𝑁/{𝑗} = y𝑗
∑xiji,j ∈δS ≥y𝑘 k∈S⊆N
6
T
T
7
8
9
[1]
Pulido, D., Darido, G., Munoz-Raskin, R. and Moody, J., “The urban rail development handbook.”, World
Bank Group Press, 2018.
[2]
Laporte, G., Mesa J. A., Ortega F. A., and Perea F.. "Planning rapid transit networks.", Socio-Economic
Planning Sciences, Vol. 45, No. 3, pp. 95-104, 2011.
[3]
Dufourd, H, Gendreau M., and Laporte G.., "Locating a transit line using tabu search.", Location Science,
Vol. 4, No. 2, pp.1-19, 1996.
[4]
Bruno, G., Gendreau, M. and Laporte, G., “A heuristic for the location of a rapid transit line.”, Computers
& Operations Research, Vol.29, No. 1, pp.1-12, 2002.
[5]
Labbé, M., Laporte, G., Martín, I.R. and Gonzalez, J.J.S., “The ring star problem: Polyhedral analysis and
exact algorithm.”, Networks: An International Journal, Vol.43, No. 3, pp.177-189, 2004.
[6]
Kermansshahi, S., Shafahi, M., Mollanejad, Y. and Zangui, M., “Rapid transit network design using
simulated annealing.”, 12th World conference of transportation research, Lisbon, Portugal, 2010.
[7]
Gutiérrez-Jarpa, G., Obreque, C., Laporte, G. and Marianov, V., “Rapid transit network design for optimal
cost and origin–destination demand capture.”, Computers & Operations Research, Vol. 40, No, 12,
pp.3000-3009, 2013.
[8]
Laporte, G. and Pascoal, M.M., “Path based algorithms for metro network design.”, Computers &
Operations Research, Vol. 62, pp.78-94, 2015.
[9]
Canca, D., De-Los-Santos, A., Laporte, G. and Mesa, J.A., “A general rapid network design, line planning
and fleet investment integrated model.”, Annals of Operations Research, Vol. 246, No. 1, pp.127-144, 2016.
[10]
Cadarso, L. and Marín, Á., “Rapid transit network design considering risk aversion.”, Electronic Notes in
Discrete Mathematics, Vol. 52, pp.29-36, 2016.
[11]
Wei, Y., Jin, J.G., Yang, J. and Lu, L., “Strategic network expansion of urban rapid transit systems: A bi‐
objective programming model.”, Computer‐Aided Civil and Infrastructure Engineering, pp.1-13, 2018.
[12]
Costa, A., Cordeau, J.F. and Laporte, G., “Steiner tree problems with profits.”, INFOR: information systems
and operational research, Vol. 44, No. 2, pp.99-115, 2006.
[13]
Feillet, D., Dejax, P. and Gendreau, M., “Traveling salesman problems with profits.”, Transportation
science, Vol. 39, No. 2, pp.188-205, 2005.
10
Application of a mathematical programming model for development of Tehran metro
network
Amir R. Mahdavi1, A.R. Mamdoohi2 and M. Allahviranloo3
1- MSc Graduate of Transportation Planning, Tarbiat Modares University, Tehran, Iran. Email:
amirrezamahdavi@modares.ac.ir
2- Associate Professor, Civil and Environment Faculty, Tarbiat Modares University, Tehran, Iran. Email:
armamdoohi@modares.ac.ir
3- Assistant Professor, Grove School of Engineering, City College of New York, New York, USA. Email:
mallahviranloo@ccny.cuny.edu
Abstract
Designing or Development of a transit network is one of the most important challenge of transportation
engineers and urban planners. Among various types of transit, metro is so noticeable because it is green and
massive transit. Metro lines or network design problem is so complicated because a metro network has
numerous stakeholders with different purposes and it is very costly. In this paper, two binary non-linear
mathematical programming models are proposed for designing circular and non-circular lines. The
objective is to maximize the ratio of population coverage per construction cost. A heuristic method is
introduced for solving these problems. Four non-circular corridors are considered to improve population
coverage and accessibility of the Tehran metro network, and two circular corridors are defined to enhance
connectivity and robustness. Results indicate that the total length of new lines and the total number of
stations are equal to 140.2 km and 102, respectively. The total construction cost is 47.2 thousand billion
Tomans. Consequently, for one billion Tomans investment to construct these new lines, the total population
coverage of the Tehran metro network is increased equal to 36 persons.
Key words: Metro line design model, Metro network development model, Development heuristic
algorithm, Binary non-linear mathematical programming