Postdoc in Trustworthy Multi-Model Optimization for Railway Timetabling (ICAI Robust)

Delft University of Technology (TU Delft); Netherlands; On-site; EUR 47,000 - 62,000; Full time

Benefits offered by this job

Relocation support

ABP pension

Education opportunities

Flexible leave

Job summary

Delft University of Technology (TU Delft) invites applications for a postdoctoral researcher in the ICAI Robust RAIL Lab to advance multi-model optimization for railway timetabling and capacity planning. The role bridges theory and a large real-world transport system, in collaboration with ProRail and NS, under the NWO LTP-ROBUST programme.

You will develop trustworthy optimization proxies, explore market-design aspects of train path allocation, and publish results while shaping your own

Qualifications

• PhD in optimization, OR, ML or closely related field.
• Strong mathematical and algorithmic background.
• Conversational Dutch preferred for collaboration with Dutch Railways and ProRail.
• Interest in theory-to-real-world sustainable transport systems.

Responsibilities

• Bridge models across macroscopic to microscopic levels for rail optimization.
• Develop multi-model optimization and self-certifying proxies.
• Work with ProRail, NS and TU Delft researchers; publish research results.

Skills

Mathematical/algorithmic grounding

Research mindset

Interdisciplinary collaboration

Dutch conversational level

Education

PhD in optimization/OR/ML or related field

Job description

Delft University of Technology (TU Delft)

Organisation/Company Delft University of Technology (TU Delft) Research Field Engineering » Computer engineering Engineering » Systems engineering Researcher Profile Recognised Researcher (R2) Application Deadline 31 Oct 2026 - 22:59 (UTC) Country Netherlands Type of Contract Temporary Job Status Not Applicable Hours Per Week 40.0 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

Offer Description

ICAI Robust RAIL Lab is looking for a postdoctoral researcher to help bridge that gap, along two connected directions.
Job description
The Dutch railway network is approaching its capacity limits, and the timetable that keeps it running is one of the most consequential optimization problems in the country — spanning many orders of magnitude in detail, from macroscopic capacity-planning down to millisecond-level microscopic simulation. No single algorithm spans that whole range on its own.
ICAI Robust RAIL Lab — a joint initiative of TU Delft, Utrecht University, and NS/ProRail under the NWO LTP-ROBUST programme — is looking for a postdoctoral researcher to help bridge that gap, along two connected directions.

• Trustworthy multi-model optimization. Combine models at different levels of abstraction — macroscopic capacity planning, mesoscopic scheduling, microscopic simulation — into a single, computationally tractable pipeline, without sacrificing trust in the result. Draw on recent work on self-certifying optimization proxies — fast machine learning surrogates that carry their own provable optimality guarantee, falling back to an exact solver only on the instances that need it — and on the broader idea of building a relaxation that's fine-grained exactly where a query needs it and coarse elsewhere. This hybrid, multi-model philosophy motivates decision support in other domains as well.
• Timetabling for a multi-operator railway. ProRail's system for allocating rail capacity is shifting from an operator-request model to one where train paths are effectively "sold" to multiple, competing railway operators — turning capacity allocation into a genuine market-design problem. How do you allocate scarce train paths fairly, efficiently, and strategy-proofly across operators who each hold private information about their own needs? This is a live mechanism-design question, not a hypothetical one: it echoes the FCC's landmark two-sided spectrum incentive auction.

You'll work with researchers in the Algorithmics section, headed by Mathijs de Weerdt, contacts at ProRail and NS, and some of the lab's researchers (PhD tracks on hub planning, real-time rescheduling, crew rostering, energy-efficient operation) — with freedom to shape your own research agenda within this scope.

What we offer: a postdoctoral position for 16 fte months (working days per week and exact duration negotiable), embedded in a well-resourced, industry-connected lab.

Job requirements

• A recently obtained PhD in optimization, operations research, machine learning, or a closely related field.
• A strong mathematical/algorithmic grounding.
• A genuine interest in working at the interface of theory and a large, real-world system for sustainable transport.
• Due to your collaboration with Dutch Railways and ProRail, a conversational level of spoken Dutch is strongly preferred.

Conditions of employment

• Duration of contract is 16 months. Temporary.
• A job of 32-40 hours per week.
• Salary and benefits are in accordance with the Collective Labour Agreement for Dutch Universities.
• An excellent pension scheme via the ABP.
• The possibility to compile an individual employment package every year.
• Discount with health insurers on supplemental packages.
• Every year, 232 leave hours (at 38 hours). You can also sell or buy additional leave hours via the individual choice budget.
• Plenty of opportunities for education, training and courses.
• Partially paid parental leave
• Attention for working healthy and energetically with the vitality program.

Will you need to relocate to the Netherlands for this job? TU Delft is committed to make your move as smooth as possible! The HR unit, Coming to Delft Service , offers information on their website to help you prepare your relocation. In addition, Coming to Delft Service organises events to help you settle in the Netherlands, and expand your (social) network in Delft. A Dual Career Programme is available, to support your accompanying partner with their job search in the Netherlands. .