Context
Mercitalia Rail, the rail-freight company of the FS Group, was replacing three legacy crew-management systems — ALPI, GENTE, and VESTE — with a single integrated platform: IVU.Rail. The programme also introduced optimisers for pairing and cyclic grids, weekly scheduling, and repairing, and had to align the new system with the wider MIR ecosystem.
It was a multi-party transformation: client, vendor, and internal teams, with a long phase plan and a lot of decisions that needed to stay aligned. I joined FSTechnology as a project management intern for five months, during the closing phase of the programme.
My role
I supported eight phases of the programme — 3, 4, 7, 8, 9, 10, 11, and 12 — spanning system foundations, optimisation algorithms, and final module rollouts. I contributed directly to closing phases 4, 7, and 10. My work sat between the people making decisions and the people building the system.
The work was not the optimiser. It was making sure the people building it agreed on the problem.
What I did
- Project governance and phase management. Supported end-to-end execution across eight critical phases, from system foundations to optimisation algorithms and final module rollouts. Helped prepare the documentation needed to close phases 4, 7, and 10.
- Operational coordination. Acted as a bridge between client, vendor, and internal teams: streamlined technical communication, aligned decision-making, and followed up on open points in the most delicate phases.
- Requirements analysis and technical synthesis. Analysed complex business processes and technical documentation; translated constraints into clear, actionable material for governance and decision-making boards.
- Optimiser scope. Supported the definition and documentation of pairing and cyclic-grid rules, the weekly and repairing optimisers, and the GO interfaces — at requirements and governance level, not as the implementation owner.
Result
Three phases closed with direct contribution, five more had support, and the programme moved through its closing phase in 2026. The useful outcome was not a single feature: it was a clearer decision trail between a client, a vendor, and an internal team working on a system that thousands of daily rail operations would depend on.
What I would change
I would introduce a single decision log earlier, make ownership between client and vendor explicit from phase one, and keep one source of truth for requirements instead of reconstructing context near each closure. Those are process changes, but they are the difference between a project that closes and a project that can be maintained.
Notes
This was an enterprise project. I have kept the description at the level of process and technology and omitted client data, internal screens, and commercial details. I am happy to talk through the approach in a call.