rDDSProcess: Simulations from day-to-day stochastic process models for...

In MartinLHazelton/transportation: What the Package Does (One Line, Title Case)

Simulations from day-to-day stochastic process models for traffic

Description

This functions generartes sequences of traffic flow patterns drawn from one of a number of day-to-day stochastic process models.

Usage

rDDSProcess(
  ODdemand,
  ODpair,
  A,
  N.days = 100,
  N.sim = 1,
  x.start = NA,
  Alpha,
  Beta,
  pow = 4,
  RUM = "logit",
  theta = 1,
  psi = 0.05,
  process = "Standard",
  variable.demand = F,
  tol = 1e-04,
  verbose = F
)

Arguments

ODdemand	Vector of origin-destination (OD) travel demands
ODpair	Vector indicating the OD pair serviced by each route (ordered by columns of the path-link incidence matrix, A).
A	Path-link incidence matrix
N.days	Number of days to simulate. Defaults to 100.
N.sim	Number of parallel simulation runs. Defaults to 1.
x.start	Initial route flow pattern. Defaults to NA, when the initial flow is set to SUE.
Alpha	Vector of free flow travel time parameters for each link
Beta	Vector of capacity parameters for each link
pow	Polynomial order of cost function for each link. Defaults to 4.
RUM	Choice of random utility model. Can be "logit" (the default) or "probit".
theta	A dispersion parameter. For the logit model, a single value specifying the logit parameter. For the probit model, a vector of standard deviations for the individual link cost errors. Defaults to 1.
psi	Recency parameter, controlling weight assigned to most recent path costs when updating disutility. Defaults to 0.05.
process	Defaults to "Standard", when a time-homogeneous Markoc process is employed in which route choice decisions are based on a utility which is a convex combination of the previous disutility and route costs. Alternatives are "TIDDS1" and "TIDDS2" time-inhomogeneous processes.
tol	Tolerance for convergence assessment for SUE. USed if x.start not specified. Defaults to 1e-4.
verbose	Should progress of SUE algorithm be printed out? Defaults to FALSE.
prob.model	Route choice probability model. One of "ProductMN" (product multinomial, the default), "Poisson", and "Approximate" (a quick and dirty approximation to the others).

Value

The output is a list with components x and u, containing traffic path flows and disutilies respectively. Both x and u are 3-dimensional arrays. The first dimension indexes simulation run (if multiple parallel runs are required); the second indexes network path; and the third indexes day.

Examples

A <- matrix(c(0,1,0,0,1,0,0,0,0,1,1,0,0,0,1,0,0,0,0,1,1,0,0,0,0,0,0,1),ncol=4,byrow=T) 
Alpha <- rep(10,7)
Beta <- rep(50,7)
pow <- rep(4,7)
ODpair <- c(1,1,2,2)
ODdemand <- c(50,50)
theta <- 0.7
N.sim <- 1
N.days <- 100
rDDSProcess(ODdemand,ODpair,A,N.days=N.days,N.sim=N.sim,x.start=c(20,0),Alpha=Alpha,Beta=Beta,pow=pow,RUM="logit",theta=theta,process="TIDDS1",variable.demand=F)
rDDSProcess(ODdemand,ODpair,A,N.days=N.days,N.sim=N.sim,x.start=c(20,0),Alpha=Alpha,Beta=Beta,pow=pow,RUM="logit",theta=theta,process="TIDDS2",variable.demand=F)
rDDSProcess(ODdemand,ODpair,A,N.days=N.days,N.sim=N.sim,x.start=c(20,0),Alpha=Alpha,Beta=Beta,pow=pow,RUM="logit",theta=theta,process="Standard",variable.demand=F)

MartinLHazelton/transportation documentation built on Aug. 5, 2023, 10:28 a.m.