Assgm 2 - Traffic Assignment problem - MM
Assgm 2 - Traffic Assignment problem - MM
My parameters: [[Assgm 2 - Parameters - MM.pdf]]
Name | ID | File_XX.jpg | O-D Flow | Aug. factor for |
||
---|---|---|---|---|---|---|
Matteo Meloni | ca41844hp | 18 | 2 | 0,1863 | 30 | 2.9430 |
You are assigned the following network
The Volume delay function is the same for each link, and it's equal to:
where:
is the flow for the given link
According to the assigned parameters, the values are as follows:
❗❗ Keep in mind, parameters have different names in the AMPL code...
What we have to do
Solve a 02 - Traffic Assignment Problem - MM using AMPL.
- Solve the whole problem (Beckmann's problem) minimizing the objective function in the MinCM2.mod file
- Solve implementing the Frank-Wolfe Method in the .run file and using solvers only to solve the subproblem
Beckmann's problem - Assgm 2
Given the volume delay function:
the Beckmann's problem is
Subject to
where
The objective function can be written explicitly:
In AMPL, the Beckmann's problem is problem
problem Q: v_k, v, N, flux_total, Vnl; # Traffic Assignment (Equilibrium problem) Definition
Which uses:
the link flow vector (what we called in problem definition before) node vector flow of commodity (we have 2 commodities in our case) Beckmann's problem objective function
The objective function in the .mod file appears as
minimize Vnl: sum { (i,j) in links} (gamma * c[i,j] + cap[i,j] * (1/3) * v[i,j]^3);
which corresponds to
where:
Augmenting factor for in red links the old parameter the old parameter
Last question
Pick 4 OD pairs.
For each OD pair, pick 3 paths.
Picked OD pairs and paths:
- (11, 17)
- (11, 12) (12, 15) (15, 16) (16, 17)
- (11, 12) (12,13) (13, 17)
- (11, 12) (12, 15) (15, 13) (13, 17)
- (11, 14)
- (3,)