losdos

Level-of-service attribute estimation using dodgr with the Ordnance Survey Multimodal Routing Network.

losdos is an R package that computes distance and time estimates for origin-destination pairs across multiple transportation modes (walk, bicycle, and car) using the UK Ordnance Survey Multimodal Routing Network (OSMRN).

This method is described in detail in the paper:

Roberts, H. S., Calastri, C., Batley, R. (under review) "Evaluating open-source approaches for estimating level-of-service attributes in transport choice modelling". Manuscript submitted for publication.

The package incorporates:

Time-varying and traffic-sensitive car speeds across 14 different time periods, based on real average speed data
Slope-aware walk times (Weidmann model)
Slope-aware bike times (Parkin-Rotheram model) that are traffic-sensitive and include the option to walk-and-push
Turn restrictions integrated from the OSMRN
Minimum-time routing via the dodgr package

All distances and times are computed along minimum-time paths and returned in standard units (metres for distance, minutes for time).

Installation

Prerequisites

You will need:

R version 4.0 or above
OSMRN data files: osmrn.gpkg — OS Multimodal Routing Network (GeoPackage format)
Study area boundary boundary.gpkg — (GeoPackage format, EPSG:27700 (BNG))

OSMRN Dataset

This package is designed to work with the Multimodal Routing Network, published by the Ordnance Survey. The dataset is not included in this package and must be obtained separately by the user.

Source: Official website
- It is also available for educational and research purposes through the EDINA Digimap platform
License: Use of the dataset is subject to the Ordnance Survey's licensing terms. Users are responsible for reviewing and complying with these terms.
Format expected: The package expects the dataset in Geopackage format as published, with no further adjustment by the user.

Install from GitHub

# Install devtools if you haven't already
if (!require("devtools", quietly = TRUE)) install.packages("devtools")

# Install losdos
devtools::install_github("harrysroberts/losdos")

Usage

There are two main functions for computing attributes:

osmrn_trip_attributes() — Compute distance and time by each mode for a set of defined trips with fixed origins, destinations and time periods
osmrn_matrix_attributes() — Compute walk/bike/car distance and time for all pairs of origins and destinations across specified time periods

Basic Examples

Example of osmrn_trip_attributes():

library(losdos)

# Prepare your trips dataset
trips <- data.frame(
  id = c(1, 2, 3),
  period = c("MoFr09001200", "MoFr19002200", "SaSu14001900"),
  from_easting = c(429180, 427750, 435741),
  from_northing = c(434731, 435747, 432124),
  to_easting = c(429906, 430454, 430731),
  to_northing = c(433271, 433532, 441858)
)

# Compute mode and time-specific attributes
results <- osmrn_trip_attributes(trips)

# View output
results

#        id       period ... walk_distance walk_time bike_distance bike_time car_distance car_time
# 1       1 MoFr09001200 ...      2143.134  26.32291      2145.671  10.09154     2588.870 10.59448
# 2       2 MoFr19002200 ...      4226.703  52.46928      4079.125  14.82283     4181.384 10.42310
# 3       3 SaSu14001900 ...     13172.694 164.69240     13061.157  44.89270    15626.795 28.32776

Example of osmrn_matrix_attributes():

library(losdos)

# Prepare your origins dataset
origins <- data.frame(
  id = c(1, 2, 3),
  easting = c(429180, 427750, 435741),
  northing = c(434731, 435747, 432124)
)

# Prepare your destinations dataset
destinations <- data.frame(
  id = c(1, 2, 3),
  easting = c(429906, 430454, 430731),
  northing = c(433271, 433532, 441858)
)

# Compute mode and time-specific attributes for periods of interest
results <- osmrn_matrix_attributes(
  origins,
  destinations,
  periods =  c("MoFr09001200", "MoFr19002200", "SaSu14001900")
)

# View output
results

#    origin destination       period ... walk_distance walk_time bike_distance bike_time car_distance  car_time
# 1       1           1 MoFr09001200 ...      2143.134  26.32291      2145.671 10.091541     2588.870 10.594476
# 2       1           1 MoFr19002200 ...      2143.134  26.32291      2145.671  9.686857     2588.870  8.842980
# 3       1           1 SaSu14001900 ...      2143.134  26.32291      2148.427  9.844002     2588.870  9.014744
# 4       1           2 MoFr09001200 ...      2158.624  26.53713      2158.624 11.197209     2551.594  8.479723
# 5       1           2 MoFr19002200 ...      2158.624  26.53713      2158.624  9.109990     2551.594  6.614542
# 6       1           2 SaSu14001900 ...      2158.624  26.53713      2158.624  9.572531     2551.594  7.865528 
# 7       1           3 MoFr09001200 ...      8761.434 110.58688      8664.031 32.905502     9717.246 18.881990
# 8       1           3 MoFr19002200 ...      8761.434 110.58688      8664.031 31.731030     9717.246 17.302694
# 9       1           3 SaSu14001900 ...      8761.434 110.58688      8664.031 32.861117     9717.246 18.127696
# ...
# 25      3           3 MoFr09001200 ...     13172.694 164.69240     13061.157 45.259649    15626.795 30.226330
# 26      3           3 MoFr19002200 ...     13172.694 164.69240     13061.157 44.158099    15626.795 26.941794
# 27      3           3 SaSu14001900 ...     13172.694 164.69240     13061.157 44.892704    15626.795 28.327760

Input Requirements

The trips data frame must include:

id — Unique trip identifier
period — Time period code (one of 14 periods: e.g., "MoFr09001200", "SaSu14001900")
from_easting, from_northing — Origin coordinates (EPSG:27700)
to_easting, to_northing — Destination coordinates (EPSG:27700)

The origins data frame must include:

id — Unique origin identifier
easting, northing — Origin coordinates (EPSG:27700)

The destinations data frame must include:

id — Unique destination identifier
easting, northing — Destination coordinates (EPSG:27700)

Available Time Periods

The OSMRN includes speeds for 14 time periods:

Weekdays (Mo-Fr):

MoFr04000700, MoFr07000900, MoFr09001200, MoFr12001400
MoFr14001600, MoFr16001900, MoFr19002200, MoFr22000400

Weekends (Sa-Su):

SaSu04000700, SaSu07001000, SaSu10001400, SaSu14001900
SaSu19002200, SaSu22000400

Output

The osmrn_trip_attributes() function returns the input trips data frame with additional columns:

Column	Description
`walk_distance`	Walking distance (metres)
`walk_time`	Walking time (minutes)
`bike_distance`	Cycling distance (metres)
`bike_time`	Cycling time (minutes)
`car_distance`	Driving distance (metres)
`car_time`	Driving time (minutes)

The osmrn_matrix_attributes() function returns a data frame with one row per origin-destination-period combination, including the same distance and time columns as above.

Custom Walk/Bike Speeds

# Use custom speeds
results <- osmrn_trip_attributes(
  trips,
  walk_speed = 5.0,        # km/h
  bike_speed = 20.0        # km/h
)

Data Setup

Place your OS-MRN data files in the working directory:

working_directory/
├── input/
│   └── raw/
│       ├── boundary.gpkg        # Study area boundary
│       └── osmrn.gpkg           # OS Multimodal Routing Network

The first call to osmrn_trip_attributes() will automatically:

Process the raw OSMRN files
Filter to your study area boundary
Compute walk, bike and car distance and time attributes for each trip
Save the processed networks in input/processed/

Subsequent calls will reuse the pre-processed data for speed.

Both functions also allow for caching and retrieving modal networks from input/processed/ to speed up repeated analyses. The logical argument make_cache = TRUE will save the processed networks for future use, while use_cache = TRUE (default setting) will load from cache if available.

Package Functions

Main Function

osmrn_trip_attributes() — Compute distance and time by each mode for origin-destination pairs
osmrn_matrix_attributes() — Compute walk/bike/car distance and time for all origin-destination pairs across specified periods

Internal Functions

create_base_network() — Build base network with all time-of-day variations
generate_modal_networks() — Extract modal (walk/bike/car) networks, one for each time period in the case of bike and car
generate_origin_links() — Create links connecting trip origins to the nearest network node
generate_destination_links() — Create links connecting trip destinations to the nearest network node
generate_augmented_networks() — Append origin/destination links to each modal network
generate_dual_networks() — Convert to dual representation with turn restrictions
compute_trip_attributes() — Compute distance/time of each trip by each mode via dodgr routing
compute_matrix_attributes() — Compute distance/time by each mode for all origin-destination pairs and specified periods via dodgr routing

References

The package implements models from:

Walking times: Weidmann (1993) doi: 10.3929/ethz-a-000687810
Cycling times: Parkin and Rotheram (2010) doi: 10.1016/j.tranpol.2010.03.001

as implement in MATSim by Horni et al. (2016) doi: 10.5334/baw

Citation

If you use this package in your research, the following citations are appreciated:

Paper describing the method:

Roberts, H. S., Calastri, C., Batley, R. (under review) "Evaluating open-source approaches for estimating level-of-service attributes in transport choice modelling". Manuscript submitted for publication.

This software:

Roberts, H.S. (2026) “losdos”. Zenodo. doi:10.5281/zenodo.21222208.

Dependency:

This package builds on dodgr. Please also cite that package.

Padgham, M. (2019) "dodgr: An R package for network flow aggregation." Transport Findings, 2(14). doi:10.32866/6945

OSMRN dataset:

This package is intended to be used in conjunction with the Ordnance Survey Multimodal Routing Network. Users of this dataset are encouraged to cite their use of this dataset.

Ordnance Survey (2026) "Multi-modal Routing Network". url:https://www.ordnancesurvey.co.uk/products/os-multi-modal-routing-network

BibTeX:

@article{roberts_under_review_evaluating,
  author      = {Roberts, Harry Samuel and Calastri, Chiara and Batley, Richard},
  title       = {Evaluating open-source approaches for estimating level-of-service attributes in transport choice modelling},
  year        = {under review},
  note        = {Manuscript submitted for publication}
}

@software{roberts_2026_21222208,
  author      = {Roberts, Harry Samuel},
  title       = {losdos},
  month       = jul,
  year        = 2026,
  publisher   = {Zenodo},
  version     = {v1.0.0},
  doi         = {10.5281/zenodo.21222208},
  url         = {https://doi.org/10.5281/zenodo.21222208},
}

@Article{padgham_2019_dodgr,
  journal     = {Transport Findings},
  doi         = {10.32866/6945},
  publisher   = {Network Design Lab},
  title       = {dodgr: An R package for network flow aggregation},
  author      = {{Mark Padgham}},
  year        = {2019},
  month       = {2},
}

@misc{ordnance_survey_multi-modal_2026,
  title       = {Multi-modal {Routing} {Network}},
  url         = {https://www.ordnancesurvey.co.uk/products/os-multi-modal-routing-network},
  urldate     = {2026-03-18},
  author      = {{Ordnance Survey}},
  year        = {2026},
}

License

MIT License. See LICENSE file for details.

Author

Harry Roberts (ts22hr@leeds.ac.uk)

Institute for Transport Studies, University of Leeds

Name		Name	Last commit message	Last commit date
Latest commit History 19 Commits
R		R
inst		inst
man		man
.Rbuildignore		.Rbuildignore
.gitignore		.gitignore
DESCRIPTION		DESCRIPTION
LICENSE		LICENSE
NAMESPACE		NAMESPACE
README.md		README.md

Navigation Menu

Repository files navigation

losdos

Installation

Prerequisites

OSMRN Dataset

Install from GitHub

Usage

Basic Examples

Input Requirements

Available Time Periods

Output

Custom Walk/Bike Speeds

Data Setup

Package Functions

Main Function

Internal Functions

References

Citation

License

Author

About

Releases

Packages

Contributors

Languages

Folders and files

Latest commit

History

Repository files navigation

losdos

Installation

Prerequisites

OSMRN Dataset

Install from GitHub

Usage

Basic Examples

Input Requirements

Available Time Periods

Output

Custom Walk/Bike Speeds

Data Setup

Package Functions

Main Function

Internal Functions

References

Citation

License

Author

About

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors

Languages