Trip planning with detailed_itineraries()



This vignette shows how to do route planning using the detailed_itineraries() function in r5r.

1. Introduction

r5r has some extremely efficient functions to run multimodal routing and accessibility analysis. In general, though, these functions output only the essential information required by most transport planning applications and simulation models. Moreover, the algorithms behind these function return only the optimal route in terms of minimizing travel times and/or monetary costs. Sometimes, though, we would like to do more simple route planning analysis and extract more information for each route. Also, we might be interested in finding not only the fastest route but some other suboptimal route alternatives too.

This is where the detailed_itineraries() function comes in. This function outputs for each origin destination pair a detailed route plan with information per leg, meaning a route taken by a single mode such as a walk to the bus stop. In R5’s documentation these legs are referred to as ‘segments’, a word more usually used to describe small sections on the transport network. Results contain information on the mode, waiting times, travel times and distances for each leg (or ‘segment’ in R5 documentation) and the trip geometry. Moreover, the detailed_itineraries() function can also return results for multiple route alternatives. Let’s see how this function works using a reproducible example.

obs. We only recommend using detailed_itineraries() in case you are interested in finding suboptimal alternative routes and/or need the geometry information of the outputs. If you only want to have route information detailed by trip segments, then we would strongly encourage you to use the expanded_travel_time_matrix() function. More info here.

2. Build routable transport network with setup_r5()

First, let’s load some libraries and build our multimodal transport network. In this example we’ll be using the a sample data set for the city of Porto Alegre (Brazil) included in r5r.

# increase Java memory
options(java.parameters = "-Xmx2G")

# load libraries

# build a routable transport network with r5r
data_path <- system.file("extdata/poa", package = "r5r")
r5r_core <- setup_r5(data_path)

# routing inputs
mode <- c('walk', 'transit')
max_trip_duration <- 60 # minutes

# departure time
departure_datetime <- as.POSIXct("13-05-2019 14:00:00", 
                                 format = "%d-%m-%Y %H:%M:%S")

# load origin/destination points
poi <- fread(file.path(data_path, "poa_points_of_interest.csv"))

3. Detailed info by trip segment for multiple trip alternatives

In this example below, we want to know some alternative routes between a single origin/destination pair. To get multiple route alternatives, we need to set shortest_path = FALSE.

# set inputs
origins <- poi[10,]
destinations <- poi[12,]
mode <- c("WALK", "TRANSIT")
max_walk_time <- 60
departure_datetime <- as.POSIXct("13-05-2019 14:00:00",
                                 format = "%d-%m-%Y %H:%M:%S")

# calculate detailed itineraries
det <- detailed_itineraries(r5r_core = r5r_core,
                            origins = origins,
                            destinations = destinations,
                            mode = mode,
                            departure_datetime = departure_datetime,
                            max_walk_time = max_walk_time,
                            suboptimal_minutes = 8,
                            shortest_path = FALSE)


The output is a data.frame sf object, so we can easily visualize the results.

3.1 Visualize results

Static visualization with ggplot2 package: To provide a geographic context for the visualization of the results in ggplot2, you can also use the street_network_to_sf(( function to extract the OSM street network used in the routing.

# extract OSM network
street_net <- street_network_to_sf(r5r_core)

# extract public transport network
transit_net <- r5r::transit_network_to_sf(r5r_core)

# plot
fig <- ggplot() +
        geom_sf(data = street_net$edges, color='gray85') +
        geom_sf(data = subset(det, option <4), aes(color=mode)) +
        facet_wrap(.~option) + 

# SAVE image
ggsave(plot = fig, filename = 'inst/img/vig_detailed_ggplot.png', 
       height = 5, width = 15, units='cm', dpi=200)

4. A few options:

4.1 Combining orings and destinations

4.2 Keep geometry data in the output

5. Hack for frequency-based GTFS feeds

Please note that the detailed_itineraries() functions does not run on frequency-based GTFS feeds. A simple hack to overcome this problem is to convert your GTFS data from frequencies to time tables. This can be easily done using the gtfstools package. Here is how:


# location of your frequency-based GTFS
freq_gtfs_file <- system.file("extdata/spo/", package = "r5r")

# read GTFS data
freq_gtfs <- gtfstools::read_gtfs(freq_gtfs_file)

# convert from frequencies to time tables
stop_times_gtfs <- gtfstools::frequencies_to_stop_times(freq_gtfs)

# save it as a new file
gtfstools::write_gtfs(gtfs = stop_times_gtfs,
                      path = tempfile(pattern = 'stop_times_gtfs', fileext = '.zip'))

… and now you can use r5r on this

Cleaning up after usage

r5r objects are still allocated to any amount of memory previously set after they are done with their calculations. In order to remove an existing r5r object and reallocate the memory it had been using, we use the stop_r5 function followed by a call to Java’s garbage collector, as follows:

rJava::.jgc(R.gc = TRUE)

If you have any suggestions or want to report an error, please visit the package GitHub page.