In-silico pre-diagnosis of transport infrastructure safety
SAMRoute is an explainable generative AI that helps transport infrastructure stakeholders model and assess the safety of transport infrastructure before incidents occur.
Nov 26, 2025Rail Live 2025, Madrid, ES
Conversations and exchanges on railway infrastructure.
Nov 20, 2025OVHSummit, Paris, FR
Back again this year.
Nov 19, 2025Salon des Maires et des Collectivités Locales, Paris, FR
Conversations, exchanges, and real-world feedback.
Nov 18, 2025SOLUTRANS 2025, Lyon, FR
Conversations, exchanges, and real-world feedback in Lyon.
Nov 5, 2025RNTP 2025, Orléans, FR
Exchanges with public transport stakeholders.
Nov 4, 2025Journées de la filière auto (PFA — Filière Automobile & Mobilités)
Meetings with PFA, automotive & mobility sector.
Sep 11, 2025MOUV'INDUSTRIE, Circuit des 24h du Mans, FR
Exchanges with partners.
Jun 16, 2025UITP 2025, Hamburg, DE
Discussions about SAMRoute.
Apr 2, 2025SITL 2025, Paris, FR
Attendance, networking and discussions about SAMRoute.
Feb 6, 2025FLOTAUTO, Paris, FR
Attendance, networking and discussions about SAMRoute.
Jan 22, 2025ATEC ITS, Paris, FR
Attendance, networking and discussions about SAMRoute.
Nov 28, 2024OVHSummit, Paris, FR
Present at the annual event of our technical sponsor (thank you!).
Nov 20, 2024Mayors and Local Authorities Fair, Paris, FR
Discussions about the SAMRoute software, networking.
Nov 6, 2024Smart City Expo, Barcelona, ES
Discussions about the SAMRoute software, networking.
Oct 16, 2024Paris Motor Show, Paris, FR
Discussions about the SAMRoute software, networking.
Oct 1–4, 2024Livestock Summit, Clermont-Ferrand, FR
On-site presence to discuss agrivoltaics.
Nov 12–18, 2023AGRITECHNICA, trade fair for agricultural machinery, DE
Immersion to understand risks and solutions.
Nov 10, 2023SRA Technical Day (France Assureurs), Paris, FR
Presentation of the global road risk scoring and Q&A.
Evaluate risk in a systematic and comprehensive way
Across 100% of the network — including the 80% that are often neglected
Here’s how practices evolve with SAMRoute:
- Traditionally, road safety intervention is reactive. With SAMRoute, your approach becomes proactive, aiming to anticipate incidents before they occur.
- Today, analyses focus only on critical areas, neglecting the rest of the network. With SAMRoute, the evaluation now covers 100% of the network, offering a complete view.
- Today, road safety management is fragmented and limited by administrative boundaries. With SAMRoute, the analysis is comprehensive and accessible, exceeding administrative boundaries.
- Today, data is often fragmented and hard to compare. With SAMRoute, the data is now complete and comparable.
Now, covers 100% of the infrastructure
Allow analysis before incidents happen
Better the comparability across zones
Features of the SAMRoute platform and its risk score
The SAMRoute score estimates the annual risk of having at least one accident per road segment every 100 meters, globally, in a linear and multifactorial way.
72.5 million km
global coverage
85 million images
with signs, markings, objects
20 million accident events
FR, NL, UK, US
34 risk factors
linear and explainable risk score
Four model variants
auto, moto, bicy, peds
3.3 billion risk scores
predicted globally
top 10% has 90% positives
positive predictive value
Country-specific
Data layers
An explorer to visualize, filter, summarize, compare
And an API for direct access to data
The SAMRoute user interface (UI) allows users to:
- add and remove layers
- compare two maps side by side
- aggregate data in the form of heatmaps (for certain layers)
- summarize active layers using KPIs (analytics)
- filter by administrative geographic zones
- group geographic areas and compare these groups
- select segments or areas exceeding risk thresholds (clipping)
- measure the risk of a route between point A and point B (routing).
Road Risk LayerAdmin BoundariesHeatmapPOIs: Railroad crossingsML diagnostics: ROCSatellite: NASASatellite: ESARouting: From A to B360: Case ReviewL'API SAMRoute
Part of the data used and integrated into SAMRoute
OpenStreetMap: It's the primary data of SAMRoute, which describes the road network topology and the base annotations.
NASA: The gridded population of the world (GPW) layer helps visualize the expected population density
ESA (European Space Agency): The worldcover layer helps visualize the estimated land cover next to the road infrastructure.
Mapillary by META: The street view layer to see street photos as well as detections by pattern recognition (AI).
ONISR (French Gov.): SAMRoute uses that data as accidentology layer and to evaluate the risk score performance.
Rijkswaterstaat (Gov. of The Netherlands): SAMRoute uses that data as accidentology layer for the Netherlands.
United Kingdom: SAMRoute integrates as part of its accidentology layer the historical accident data from the UK.
Global Transit Data (GTFS): As a proof-of-concept (PoC, in France), we have integrated trips and stops from transit data.
Resources and case study
Why road safety remains a major challenge
Because the total cost of road insecurity represents 1.5 to 2% of GDP in Western countries. In France, it's 37.9 billion euros (or 1.7% of GDP), a massive burden.
The French road network spans 1 to 2 million km.
- And across this network, the total cost of road insecurity is 56.2 million euros / year / 100,000 inhabitants (1.7% of GDP).
- Road safety efforts focus on a critical fraction of the network, leaving the rest under-analyzed.
- Administrative boundaries constrain road safety efforts.
- Road safety is traditionally reactive. Accidents are rare events, leading to a lack of data to guide actions.
- Finally, regulations are evolving as the EC 2019/1936 directive mandates proactive evaluation of critical road infrastructure by the end of 2024.
Massive cost of road insecurity (FR)
Efforts constrained on a fraction of the network
Efforts limited by admin boundaries
Reactive efforts (limited data)
Important roads to evaluate proactively
Frequently Asked Questions
Precise answers to your questions
How does SAMRoute evaluate road safety?
What types of data does SAMRoute use for road scoring?
Is SAMRoute software suitable for all types of road networks?
How can we integrate SAMRoute into our current management system?
Can SAMRoute be used for large-scale road network analyzes?
What is the accuracy of the safety scoring provided by SAMRoute?
How does SAMRoute assist in decision-making for road maintenance?
Does the software offer customization features for specific use cases?
Is there a demo version of SAMRoute that we can try?
What is the data update process in SAMRoute?
How does SAMRoute handle changes in road infrastructure over time?
Does the software provide support and training for new users?
How does SAMRoute contribute to the overall improvement of road safety?
Is the software available in multiple languages?
Are there any case studies or user testimonials for SAMRoute?
The reasons to select SAMRoute
— Why SAMRoute could be an innovative solution to improve the safety of your organization's transport infrastructure?
Because:
- SAMRoute evaluates 100% of the road network, including often overlooked segments.
- You are not limited by administrative boundaries.
- You stay ahead of new regulations, such as the EC 2019/1936 directive, which mandates proactive road assessment for major roads by 2024.
- By identifying high-risk segments early, you have the chance to avoid human and material costs resulting from accidents.
- The risk score from SAMRoute is linear and explainable, unlike AI models that are non-linear and opaque.
See what SAMRoute can do for you
Tell us about yourself and we'll connect you with an expert from SAMRoute who can share more about the product and answer your questions.
Plans tailored to each network
Each plan provides access to SAMRoute in your area of interest, with the option to add more within the plan's limits. For pricing tailored to larger networks, contact us.
Contact us
Our contact detailsOur online presence
Our contact detailsOur online presence