About SAMRoute
SAMRoute is a SaaS platform that clarifies context and reveals functional dependencies around critical infrastructure points.
This page explains what SAMRoute is (and is not), why it exists, how we deliver, and how we keep things clear and stable over time through a programme with step-by-step gates and sign-off notes.
SAMRoute can run as a programme investment. We deliver something ready to use, checked with you, and documented.
For partnership and co-sponsoring, see our page:
Become a partnerWhat SAMRoute is
SAMRoute gives you a shared reference that shows who depends on what around a critical infrastructure point such as a level crossing.
. What we map?
- The point(s) such as one LX, a group of LX, or a whole list of LX.
- The surroundings (schools, care homes, logistics areas, industrial sites, population, roads…).
- The links between them (dependencies, constraints, detours, exposure).
. What you get (outputs)?
- Simple indicators (KPIs) you can compare across sites.
- Maps and dashboards to explore the context.
- Delivered in steps with checkpoints so it is ready to use.
. Why it matters?
It turns the area around a critical point into something you can describe, compare, and check again next year.
What SAMRoute is NOT
In one sentence, SAMRoute is a decision support reference layer, not a live operational control tool.
. What we are NOT?
- Not real-time operations. It is not a control room and does not replace traffic management tools.
- Not a replacement for regulated processes. It complements your existing safety methods and responsibilities.
- Not a “black box”. Results remain explainable: sources, versions, and computation logic are traceable.
- Not “internal data first”. We can start with open / third-party data, and add internal data only if useful.
. What you can expect?
- A clear perimeter (what is in / out).
- Clear outputs (KPIs, dashboards, fact sheets, exports, APIs).
- Results that are replayable, comparable, and extendable.
Why SAMRoute exists
. The structural issue
A critical point (like a level crossing) is not isolated. Its risk depends on what happens around it, such as roads, nearby activity, people, and other routes. In big systems, this information is spread across many groups (rail, road, cities, departments, private actors). Each group knows its part, but because responsibilities are split, the full picture is hard to keep shared and stable over time.
. The missing layer
What is often missing is a neutral reference that explains, in a simple and repeatable way, who depends on what, where constraints are, and what changes as the territory evolves. SAMRoute builds that reference by linking the point, its context, and the functional dependencies between them, so you can run the same reasoning again, compare sites, and come back next year without starting over.
. Why it matters now
Today, stakeholders need decisions that are easier to explain, easier to share internally, and easier to audit over many years. A stable reference helps teams align faster, reduce guesswork, and keep a clear record of what was assessed and why.
About the founder
. Who is behind SAMRoute?
The SAMRoute project was founded by Fabrice Colas. He is a risk and data specialist (PhD in machine learning). His consulting work is CIR-approved (French Ministry of Research). He also did postdoc research in human genetics, where accuracy and repeatable results are key.
. What triggered SAMRoute?
The trigger was a recurring gap between models and real-world use. Teams can produce studies, but keeping things running is hard: up-to-date data, consistent indicators, and outputs that stay clear and easy to share. SAMRoute closes that gap by linking data, analytics, and decision-ready artifacts in a replayable way.
. Why a platform approach?
SAMRoute is a shared platform because the work repeats. We build the rules, the calculation engine, the views, and the docs once. Then teams can use it again, extend it, and compare places.
. How continuity is ensured
Continuity comes from turning knowledge into code and documentation. We build the whole pipeline (data and indicators) so it works the same way every time. We also publish open documentation so the method stays easy to understand and use again over time.
How we deliver
. Working rhythm
We use a step-by-step approach. We ship something usable early and then improve it with your feedback. We also anticipate the next 2–3 likely questions so each exchange removes the next blocker. We work in gates with clear steps. Each gate ends with a demo or video and a short sign-off note.
. What is delivered
We deliver a shared reference in the platform that maps context and dependencies around each point. Teams get numbers, maps, and views they can use right away, and they can zoom in when they need details.
. Inputs and integration
We can start with no internal inputs. SAMRoute uses open and third-party data first. If it helps, SAMRoute can also add private (non-critical) data to go further.
. Support and responsiveness
Support is best-effort, with agreed hours and a direct line to the team. On-site sessions are possible in Paris and Western Europe if useful. SAMRoute runs on our own servers and on the cloud, with reliability covered by the provider’s SLA.
. Traceability and replayability
At this stage, SAMRoute does not use “black box” models. SAMRoute uses a clear calculation engine and versioned datasets, so results can be re-run and explained. SAMRoute also keeps development and production environments separate.
Governance
. Security and data minimization
To get started, we mainly need the location of the critical points (for example, LX). When it matters, we also need the location of nearby traffic emitters (SEVESO sites, schools, care homes, logistics hubs, and similar). We can also add simple, non-sensitive details for each point (type, shape, signage) and link to public expert reports (for example, BEA-TT) or local equivalents.
Each organization gets its own separate space. Access is protected by a password, with optional one-time codes, and SAMRoute keeps web access logs. For early R&D onboarding, SAMRoute runs on our infrastructure in Ille-et-Vilaine (France). For operational contracts, SAMRoute can run in OVHcloud data centers in Western Europe (Roubaix/Gravelines).
. Traceability
We record which data version was used and which build ran the calculation. Every result has a clear date and version.
. Reversibility and export
Shareability comes from platform access. Teams use the same reference in one place (views, dashboards, links, access rights, and drill down). At TB scale, moving the full reference is not practical, so the platform keeps it shared and stable.
Continuity
. How documentation reduces key-person risk
We maintain a reference of about 100 metrics. Each metric is written in plain language and available in multiple languages. We commit to publishing one public page per metric that explains what it means and how to use it. This documentation helps teams apply the method consistently, without relying on a single expert.
. How engineering keeps continuity
On the engineering side, continuity relies on concrete artifacts. We keep development and production separate. We manage infrastructure as code. We use versioned datasets. We tag releases of the calculation engine. We use basic monitoring and alerts.
. What can be delegated
Delivery is handled directly. The team stays reachable and keeps the service running. Operations are managed internally, so the project does not depend on external partners. As the project scales, two activities can be delegated with documented workflows, namely support and onboarding/training.
Trajectory
. Where we are today
In 2025, SAMRoute was mainly a demo for discussions. Since Nov 2025, SAMRoute has been a usable business demo that people can access and use to get value. The next step is to make it ready for internal use. SAMRoute will strengthen data trust with clear selection rules, stabilize the metric reference, publish documentation, and keep improving the user experience with feedback from field experts.
. What “validated” means
Validation is mostly about user workflows, not just maps. Teams need to search, read results, compare cases, share access, and come back later. We validate this through peer review and real user feedback on the workflow. SAMRoute currently runs in France. When data is available, SAMRoute can extend to other territories. SAMRoute has already tested initial datasets in Spain, such as LX locations and school emitters.
. What comes next
Next, SAMRoute will publish clear documentation for the metrics, with one page per metric. SAMRoute will support the first real workflows with partners, so teams can use it on real cases. SAMRoute will add and refresh datasets, including accident-related sources when available. SAMRoute will extend to more territories when data is available. SAMRoute will also expand analytics and metrics, moving from simple counts to more risk-focused indicators. SAMRoute will keep improving shareability through platform access, so results can circulate across multiple stakeholders.
Resources
Our resources
Our resources
Contact us
Our contact details/-1.7010147,48.1518939,2,0,0/520x240@2x?access_token=pk.eyJ1IjoiZmNvbGFzNjAxMyIsImEiOiJjbGx6NXo2bXUzMDhtM3BwOHZreXB0NWJmIn0.ERd5u_IkrKWSxriAu8CDBg)
Our contact details