README.md

In AdrienHdz/traveltimeCLT: Estimate and predict travel time on road networks

traveltimeCLT

Installation

The package is still under development in the Alpha stage.

Install from GitHub with:

# install.packages("devtools")
devtools::install_github("AdrienHdz/traveltimeCLT")

Example

This package includes a small data set (tripset) that aggregates map-matched anonymized mobile phone GPS data collected in Quebec city in 2014 using the Mon Trajet smartphone application developed by Brisk Synergies Inc. The precise duration of the time period is kept confidential.

View the data with:

library(traveltimeCLT)
library(traveltimeHMM)
library(data.table)

data(trips)
head(trips)
trip    time    timeBins    tt  logspeed    length  linkId
<int>   <chr>   <chr>   <dbl>   <dbl>   <dbl>   <int>
15  2014-05-05 07:32:41.000 MorningRush 8.00    1.801   48.422  24088
15  2014-05-05 07:32:50.000 MorningRush 64.49   0.147   74.703  23470
15  2014-05-05 07:33:54.000 MorningRush 17.34   2.621   238.410 34576
15  2014-05-05 07:34:11.000 MorningRush 4.54    2.875   80.557  34586
15  2014-05-05 07:34:16.000 MorningRush 3.03    2.905   55.259  34583
15  2014-05-05 07:34:19.000 MorningRush 8.56    2.810   142.211 6704

Transforming the variables

trips <- as.data.table(trips)
trips$speed <- exp(trips$logspeed)

Splittig data into train and test sets.

test.trips <- create_test_trips(M = 500, trips, min.n = 1)
test = trips[trip %in% test.trips]
train = trips[!trip %in% test.trips]

Creating rules and time-bins

myrules = list(
  list(start='6:30', end= '9:00', days = 0:6, tag='MorningRush'),
  list(start='15:00', end= '18:00', days = 0:6, tag='EveningRush'))

mytimebins = c("MorningRush", "EveningRush", "Other")

Fit the model using the following code:

ttCLTmodel <- traveltimeCLT(data.train = train, 
                M = 1000,
                L = 2,
                bin = "MorningRush",
                rules = myrules,
                data.timebins = mytimebins)

To predict on the test set and get estimations:

ttCLTresults <- predict_traveltimeCLT(obj.traveltime = ttCLTmodel,
                      data.test = test,
                      bin = "MorningRush",
                      rules = myrules)

References

Elmasri, M., Labbe, A., Larocque, D., Charlin, L, 2020. “Prediction intervals for travel time on transportation networks”.

https://arxiv.org/abs/2004.11292

AdrienHdz/traveltimeCLT documentation built on Dec. 31, 2020, 9:46 a.m.