Different times, same results about r5r

[mvpsaraiva]

Hi @mortewle. Thanks for trying our package.
If you are using our sample data, please have in mind that the GTFS provided contains just a few routes. Hence, when calculating a travel time matrix, many travel times will come from walking trips, which should not be affected by the departure time.

I did a small test and, if you look closely, you can see that many travel times are different. Can you try the code below and report the results back, please?

options(java.parameters = "-Xmx4G")

library("r5r")
library("dplyr")

# Start R5R core
r5r_core <- setup_r5(system.file("extdata", package = "r5r"), verbose = FALSE)

# Load points
points_hex <- read.csv(system.file("extdata/poa_hexgrid.csv", package = "r5r"))


# Testing trip date and time
max_walk_distance = 800L
max_trip_duration = 120L

# 8 am
trip_date_time <- lubridate::as_datetime("2019-03-22 08:27:00")
ttm_8am <- travel_time_matrix(r5r_core, points_hex, points_hex, 
                              mode = c("WALK", "BUS"), 
                              departure_datetime = trip_date_time, max_walk_dist = max_walk_distance,
                              max_trip_duration = max_trip_duration, verbose = FALSE)

# 11 am
trip_date_time <- lubridate::as_datetime("2019-03-22 11:13:00")
ttm_11am <- travel_time_matrix(r5r_core, points_hex, points_hex, 
                              mode = c("WALK", "BUS"), 
                              departure_datetime = trip_date_time, max_walk_dist = max_walk_distance,
                              max_trip_duration = max_trip_duration, verbose = FALSE)

# Checking results
full_join(ttm_8am, ttm_11am, by=c("fromId", "toId")) %>% 
  filter(travel_time.x != travel_time.y) %>% 
  View()

full_join(ttm_8am, ttm_11am, by=c("fromId", "toId")) %>% 
  filter(travel_time.x == travel_time.y) %>% 
  View()

[mortewle]

Hi. Looking closer at my results, it seems like all the trips I checked earlier were walking only or had the exact same waiting time for the times I chose, because now I see some divergence. It works with the code you provided as well. Thank you!

Great package by the way!

[mvpsaraiva]

Thanks @mortewle!
All feedback is appreciated.