Metro 3 in Brussels under uncertainty: scenario-based public transport accessibility analysis
summary
The gist
Metro Line 3 in Brussels is one of Europe’s most debated infrastructure projects, marked by escalating costs, delays…
In short
This study assessed Metro Line 3's impact using three scenarios: current network, partial build, and full completion. It found that while full implementation offers significant accessibility gains, these benefits are unevenly distributed, disproportionately helping lower-income neighborhoods. The analysis also confirmed that the project's benefits are robust against variations in daily schedules.
Key concepts
- Scenario-based assessment
- The study tested three different versions of Metro 3: the existing network, a partial construction plan, and the full planned route to Bordet. This allowed researchers to compare how accessibility changes under different potential outcomes rather than just one fixed plan.
- Distributional Diagnostics
- These are statistical tools used to map *where* benefits occur geographically. The paper used methods like Deciles of ∆T and Difference of empirical CDFs to show exactly which neighborhoods gain the most travel time reductions, revealing that gains are not spread equally.
- Temporal Robustness
- This checks if the accessibility improvements hold true even when service schedules change slightly. By testing three departure times (morning peak, evening peak, and weekend midday), the study ensured that the positive effects of Metro 3 aren't just a fluke on one specific day.
- Reliability Metrics
- These metrics measure how consistent and dependable travel times are. The analysis showed that the full metro 3 provides both substantial speed increases and consistent reliability improvements, meaning users can expect better performance regardless of when they travel.
Terminology used across episodes
This episode discusses
- Metro 3 in Brussels under uncertainty: scenario-based public transport accessibility analysis · Paper Radio
The paper
Metro 3 in Brussels under uncertainty: scenario-based public transport accessibility analysis · Read on arXiv
Brecht Verbeken, Arne Vanhoyweghen, Vincent Ginis
Vrije Universiteit Brussel · Harvard University · imec-SMIT
Metro Line 3 in Brussels is one of Europe's most debated infrastructure projects, marked by escalating costs, delays, and uncertainty over completion. Yet no public accessibility appraisal exists to inform this policy debate. This paper provides a scenario-based assessment of the spatial and distributional accessibility impacts of Metro 3 using schedule-based public transport data. Official GTFS feeds from all regional operators were combined and adapted to represent three scenarios: (i) the current network, (ii) partial implementation of Metro 3 (conversion of the southern premetro section), and (iii) full implementation to Bordet. Accessibility was measured as public transport travel time between 647 evenly spaced 500 m points covering the Brussels-Capital Region. Simulations were conducted for morning, evening, and weekend midday periods, each with three departure times (t-10, t, t+10), capturing robustness to short-term timetable variation. Results show substantial but uneven accessibility gains, with the largest improvements occurring in neighborhoods with below-average incomes. Temporal robustness analysis highlights variability in accessibility outcomes, underscoring the need to account for uncertainty in departure timing. These findings suggest that Metro 3 has the potential to reduce socio-spatial inequalities in accessibility, providing transparent evidence for a project where public debate is dominated by cost concerns.
Transcript
Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.
Ruth: I'm Ruth.
Warren: And I'm Warren, here with Iris.
Ruth: We're looking at the paper: "Metro 3 in Brussels under uncertainty".
Warren: Here's the gist of the paper. Metro Line 3 in Brussels is one of Europe’s most debated infrastructure projects, marked by escalating costs, delays, and uncertainty over completion <ref:2509.22223#pg2> This paper provides a scenario-based assessment of the spatial and distributional accessibility impacts of Metro 3 using schedule-based public transport data <ref:2509.22223#pg2> Results show substantial but uneven accessibility gains, with the largest improvements occurring in neighborhoods with below-average incomes <ref:2509.22223#pg2>.
Ruth: The study employs a scenario-based assessment of Metro 3 using schedule-based public transport data <ref:2509.22223#pg2> Three scenarios are simulated: (i) the current network, (ii) partial implementation of Metro 3 (conversion of the southern premetro section), and (iii) full implementation to Bordet <ref:2509.22223#pg2>.
Warren: Accessibility was measured as public transport travel time between 647 evenly spaced 500 m points covering the Brussels-Capital Region <ref:2509.22223#pg2>.
Ruth: Simulations were conducted for morning, evening, and weekend midday periods, each with three departure times (t−10, t, t+ 10), capturing robustness to short-term timetable variation <ref:2509.22223#pg2>.
Title and authors: Ruth: So we’re looking at the paper "Metro three in Brussels under uncertainty: scenario-based public transport accessibility analysis," and it sets up this whole simulation to test how different versions of Metro three affect access across Brussels <ref:2509.22223#pg1,Metro 3 in Brussels under uncertainty: scenario-based public transport accessibility analysis>.
Warren: Exactly. They take schedule-based public transport data and create three distinct scenarios for Metro three: the status quo, a partial build converting just the southern premetro section, and a full build extending all the way to Bordet <ref:2509.22223#pg1>.
Iris: The paper immediately points out that this isn't just about checking if travel times get shorter overall; it’s specifically focused on how those changes shift accessibility for different groups of people.
Ruth: Right. And what they highlight right away is that these accessibility improvements aren't spread evenly across the city, because they cluster in specific areas, especially in neighborhoods that are already more densely populated and have lower incomes one <ref:2509.22223#pg3>.
Warren: They use three diagnostic tools to map out this shift: deciles of change, the shift function of aggregated times, and the difference in empirical cumulative distribution functions to characterize how the entire distribution of travel times moves under each scenario one <ref:2509.22223#pg3>.
Iris: That specific approach lets them look beyond just averages and see the actual shape of how access changes for different populations one <ref:2509.22223#pg3>.
Ruth: And when they map out where these benefits show up geographically, they find that the pattern shifts depending on whether you're looking at a partial conversion or a full line build one <ref:2509.22223#pg3,a partial conversion or a>.
Warren: So, the geography of the benefit isn't just one simple area; it’s spread differently for each scenario, which is important context for planning one <ref:2509.22223#pg3>.
Iris: They show that for the full line, you get this larger continuous swath of improved access spreading north-east into areas like Schaerbeek and Evere compared to the more southern focus seen in the partial build one <ref:2509.22223#pg3>.
Ruth: That gives us a clearer picture of where public investment should aim if Brussels wants to maximize how much benefit spreads across the region one <ref:2509.22223#pg3>.
Warren: So, essentially, they’re using simulation to map out not just travel times, but actual social outcomes related to access and equity for Metro three one <ref:2509.22223#pg3>.
The paper's summary: Ruth: Now let’s talk about what this research itself improves over just looking at the feasibility of the project, because the authors spend a lot of time explaining why their method is stronger than just a simple study one <ref:2509.22223#pg3>.
Warren: They focus heavily on temporal robustness by using multiple time slices and day types to capture variability in service schedules, which helps you avoid those snapshots that might hide downside risk one <ref:2509.22223#pg3>.
Iris: That’s a big methodological improvement because it addresses the uncertainty in the timetable itself, not just relying on a single static plan one <ref:2509.22223#pg3>.
Ruth: Exactly. They test the stability of these accessibility benefits across morning peak, evening peak, and Saturday periods to confirm that those results aren't just lucky for one specific time of day one <ref:2509.22223#pg3>.
Warren: And they show that for the full conversion scenario, this robustness leads to consistent reliability improvements—specifically a negative three point zero second change in RBI and a change in CE of negative thirty-five point six zero one.
Iris: So, what this means is that if you’re planning something like this, you can be more confident that the benefits hold up even when the service isn't running perfectly on schedule one <ref:2509.22223#pg3>.
Ruth: It’s about showing that a full conversion delivers both substantial speed increases and consistent reliability improvements at the same time, which is a much stronger argument than just looking at one metric one <ref:2509.22223#pg3>.
Warren: I agree, and this contrasts nicely with the partial conversion scenario where they found modest gains in travel speed but ran into a risk of reliability deterioration one <ref:2509.22223#pg3>.
Iris: That contrast is key; it’s showing that the choice of implementation level matters not just for speed, but for overall system resilience one <ref:2509.22223#pg3>.
Ruth: So, the paper improves the debate by providing a way to quantify exactly how different implementation choices trade off accessibility gains against reliability risks one <ref:2509.22223#pg3>.
Warren: It’s giving policymakers a tool to see precisely where they are balancing those competing factors in their decision-making process one <ref:2509.22223#pg3>.
The paper's improvements: Ruth: Wrapping up this discussion on the paper, what do we actually have here for the world because of how they did things? The main implication is that time savings alone aren't going to justify Metro three on a narrow, short term accounting one <ref:2509.22223#pg3>.
Warren: They argue that Brussels needs this resilient north-south backbone if they want to enable a coherent surface network redesign and truly de-risk their climate and mode share goals one <ref:2509.22223#pg3>.
Iris: It’s about addressing those structural accessibility deficits in disadvantaged neighborhoods through infrastructure, which is a huge social argument one <ref:2509.22223#pg3>.
Ruth: The paper demonstrates that the distributional and resilience benefits are robust across these changes in departure time perturbations and do not degrade reliability one <ref:2509.22223#pg3>.
Warren: They’ve provided a transparent, reproducible accessibility review of Metro three which is the main contribution here for anyone trying to evaluate these projects one <ref:2509.22223#pg3>.
Iris: So, the core value lies in combining temporal robustness design with risk-adjusted diagnostics, creating a pipeline for GTFS-based counterfactual network evaluation under uncertainty one <ref:2509.22223#pg3>.
Ruth: It’s a lot of data to digest, but it makes a really strong case that the project needs to be seen as more than just a way to get faster travel times one <ref:2509.22223#pg3>.
Warren: And remember, the back-of-the-envelope monetisation shows that time alone can't justify it, so the justification has to be on wider system benefits like network option value and reliability one <ref:2509.22223#pg3>.
Iris: It’s a solid piece of analysis that puts the distributional impact front and center, showing how infrastructure decisions affect different parts of society one <ref:2509.22223#pg3>.
Conclusion: Ruth: So we’ve seen how this study on "Metro three in Brussels under uncertainty: scenario-based public transport accessibility analysis" shows that full implementation gives big gains for low-income neighborhoods, but you have to look at the trade-offs <ref:2509.2233#pg1>.
Warren: Right, and it’s not just about speed; the paper is really showing how the full build gives you both a speed increase and consistent reliability improvement across morning peaks, evening peaks, and weekends <ref:2509.2233#pg1>.
Iris: I'm interested in how they handled that uncertainty with the schedule data; they test three different departure times for each period to make sure the benefits aren't just a lucky snapshot <ref:2509.2233#pg1>.
Ruth: Exactly, and that temporal robustness means the distributional impacts are stable, which is important because it shows those accessibility gains aren't fragile <ref:2509.2233#pg1>.
Warren: But you have to remember the limitation they state: they didn't use demand models or boarding loads, so this is a lower bound for how much people actually benefit <ref:2509.2233#pg1>.
Iris: I think that’s fair; treating every travel time pair as equally important is a way to keep the welfare estimates conservative <ref:2509.2233#pg1>.
Ruth: So, the main implication for policy is that Brussels needs this north-south backbone to reduce structural deficits and de-risk their climate goals <ref:2509.2233#pg1>.
Warren: It’s a lot of capital, but the paper suggests that if you’re going to do it, you need to look at those wider system benefits like network option value and decarbonisation <ref:2509.2233#pg1>.
Iris: I just want to check one thing on the methodology; they use a GTFS based routing approach, which is cool because it’s reproducible, but the scenario information is built from internal counterfactuals which isn't official <ref:2509.2233#pg1>.
Ruth: That’s a fair caveat; you have to be careful when you rely on simulation feeds that aren't the official plans <ref:2509.2233#pg1>.
Warren: Well, what this paper really contributes is a generalisable pipeline for doing this kind of GTFS based counterfactual network evaluation when there's a lot of uncertainty involved <ref:2509.2233#pg1>.
Iris: It’s solid work because it combines temporal robustness design with risk adjusted diagnostics, which is what really matters for making these kinds of big decisions <ref:2509.2233#pg1>.
More episodes
- Daily Summary for 2026-11-19
- 2608.21274-Recommendation Quality and the Concentration of Consumption: Experimental Evidence from Netflix
- 2408.02757-A nonparametric test for diurnal variation in spot correlation processes
- 2306.16393-High-Dimensional Canonical Correlation Analysis
- 2605.17299-Geometric Brownian motion with intermittent entries and exits
- 2607.04753-Fooling Yourself: how narratives shape beliefs
- 2408.14671-Inference on Treatment Effects with Repeatedly Mismeasured High-Dimensional Controls
- 2201.06898-Difference-in-Differences Estimators for Treatments Continuously Distributed at Every Period
- 2501.15307-Influence Functions: A Unified Differentiation-Based Framework
- The day in conversation for 2026-11-19