Hamburg is being rebuilt while it keeps moving. Bridges are renewed, utility networks expanded, transport corridors modernised — all on the same streets that carry hundreds of thousands of commuters every day. This is a data story about how construction shapes congestion, transit and emissions — and what decision-makers can do about it.
Hamburg is one of Germany's fastest-growing metropolitan regions. Population growth, economic activity and climate-driven infrastructure investment are pressing hard on the transport network. When many projects run at once on major corridors, disruption spreads far beyond the works themselves.
To what extent do construction projects contribute to traffic congestion in Hamburg — and how do these disruptions affect sustainable urban mobility?
Roadworks are necessary for infrastructure renewal, yet lane closures, detours and parallel projects create bottlenecks that lengthen travel times, delay public transport and raise emissions — pulling Hamburg away from its SDG 11 targets.
During construction periods, commuters frequently report:
126+ active sites is a direct count from the Hamburg Geoportal Baustellen layer on the date we pulled it.
60% in 3 districts is the sum of Mitte, Altona and Eimsbüttel's shares in the district breakdown charted in Section 07 — computed straight from the same Geoportal export, not estimated.
93.7% bus punctuality is HOCHBAHN/HVV's own published service-quality indicator (the "5-minute punctuality" standard), city-wide for the latest reporting period.
41 µg/m³ NO₂ is a real, measured reading from the Habichtstraße station in the Hamburg Luftmessnetz — above the EU's 40 µg/m³ annual limit.
All four figures above are measured or directly computed from open Hamburg data — no modelled or estimated numbers. Section 07 shows exactly which chart each one comes from, and Section 02 explains where the international 20–50% work-zone capacity-loss range (Highway Capacity Manual) fits into the theory, separately from these local figures.
Yes — but mostly at a different scale, and rarely for Hamburg specifically. Before building the hypotheses below, we looked at how the construction-vs-congestion question has already been studied, what data those studies relied on, and what they found. That prior work shaped both our hypotheses and where we knew, going in, that our own evidence would be thinner.
International engineering research: work zones as a capacity problem
Transport-engineering literature on "construction work zones" (CWZ) is extensive, but it is largely freeway- and simulation-based rather than whole-city and open-data based:
German & Hamburg policy research: coordination and modal shift
Locally, the relevant "prior work" is less academic and more institutional:
Written for decision-makers in urban planning, transport authorities and district councils. Every section follows the same four steps so the argument stays transparent.
Anchor the case in Hamburg's growth pressure and SDG 11 — mobility during construction is a governance question.
Eight explanations for congestion, each tied to a documented source (BMDV, HOCHBAHN, UBA, City of Hamburg).
Construction sites with type & duration — layered with traffic, parking, signals and accidents.
Transit punctuality, reliability, air quality and traffic counts — did numbers move with the roadworks?
Coordination, transit-first detours, dashboards, off-peak works — concrete asks this report exists to justify.
Congestion rarely has a single cause. Roadworks are the usual suspect, but urban mobility is shaped by many interacting factors. We test eight — some tied directly to construction, others reflecting deeper characteristics of Hamburg's transport system. Each is grounded in a documented source. Hover any card to reveal the claim and where it comes from.
Closures and lane reductions cut road capacity, producing bottlenecks even at unchanged traffic volumes.
Source: BMDV Verkehr in Zahlen 2023/24; Highway Capacity Manual (TRB) work-zone capacity loss 20–50%.
Detour routes and temporary signage are not always optimised across neighbouring worksites, pushing traffic onto unsuitable streets.
Source: LSBG Hamburg Baustellenkoordinierung report; FGSV RSA guidance.
~34% of Hamburg trips are still made by private car — heavy modal reliance amplifies any disruption.
Source: Mobilität in Deutschland (MiD), Hamburg profile; Hamburg Mobilitätswende
Multiple parallel projects on the same corridor create compounded, non-linear congestion.
Source: Hamburg Geoportal Baustellen; ADAC Staubilanz 2023 case studies.
Dense inner-city districts (Eppendorf, Eimsbüttel, St. Georg) offer little geometric room for redistribution.
Source: Statistikamt Nord — road network density; Hamburg StEP Verkehr.
Fixed signal programmes are not always re-timed when a nearby worksite changes flow patterns.
Source: LSBG Lichtsignalanlagen-Register (Hamburg Geoportal); FGSV RiLSA.
Bus punctuality and reliability drop when buses share congested corridors, weakening the modal alternative.
Source: HOCHBAHN Qualitätsbericht 2023; HVV performance indicators.
Kerbside parking and double-parked deliveries narrow the usable carriageway, especially where lanes are reduced.
Source: BMDV Nachhaltige Urbane Logistik; Hamburg parking inventory (LGV).
Listing all eight keeps the causal picture honest — congestion has more than one plausible source. But the data in this project only lets us dig deep on three of them: H1 (road construction), H4 (simultaneous sites) and H7 (public transport), which get full sections and charts below. H3, H5, H6 and H8 are real, documented factors worth naming, but we don't have Hamburg-specific data to analyse them further here — they're flagged rather than developed, so the report doesn't imply a depth of analysis it doesn't have.
To understand how roadworks shape everyday travel, we layer several open datasets over Hamburg — active construction sites, live traffic, parking supply, signalised junctions and recorded accidents. Each layer answers a different question, and together they show where pressure on the network builds up.
This uMap layer marks current and planned construction sites from the Hamburg Geoportal (Baustellen & Verkehrseinschränkungen). Beyond location, each pin carries the attributes decision-makers actually need to prioritise coordination: title, organisation and duration. Projects cluster in two recognisable bands: the inner ring around Hauptbahnhof (Bezirk Mitte), and a second band tracing the A7's Hamburg stretch — Othmarschen and Bahrenfeld (Altona) up through Stellingen and Schnelsen (Eimsbüttel).
This isn't just a visual impression — the district breakdown in Section 07 ranks Mitte, Altona and Eimsbüttel as the top three Bezirke by active-site share, in that order. That match is what lets us treat the map and the chart as describing the same underlying pattern rather than two disconnected visuals.
Attributes drawn from Hamburg Geoportal Baustellen (Baustellen-Verkehrseinschränkungen), refreshed automatically by LSBG Hamburg.
A pin count alone says nothing about disruption. Two attributes in the Geoportal record change everything: what kind of work it is (a resurfacing job clears in days; a bridge or utility trench can run for a year) and how long it's scheduled for. The table below groups the current Baustellen export by type.
| Work type | Typical duration | Lane impact |
|---|---|---|
| Road surface / resurfacing | days – 3 weeks | Partial, one side at a time |
| Utility & pipe works (gas, water, district heating) | 4 – 12 weeks | Full lane closure, trench |
| Bridge / structural renewal | 6 – 18 months | Long-term partial or full closure |
| Transit infrastructure (U-Bahn/S-Bahn stations, tracks) | 6 – 24 months | Localised but high-impact detours |
| Telecom / fibre rollout | 1 – 6 weeks | Minor, kerbside |
Categories reflect the type field used in the Geoportal Baustellen schema; duration ranges are typical order-of-magnitude project lengths for each category, not a count from our export. A per-site breakdown by type (i.e. what share of the 126+ sites falls into each row) would need the live attribute export with start/end dates per site — a natural next step flagged in Section 10.2.
Google's embeddable Maps iframe (below) does not render the live traffic-speed layer — that layer is only available in the interactive app/website, not the embed API. So instead of a static, traffic-less iframe standing in for "live traffic," we compare hand-taken screenshots of the traffic layer at four points in a weekday to show how congestion actually moves through the day.
Frame these screenshots around the two bands from Layer 01: the Mitte inner ring and the A7 through Altona/Eimsbüttel. If both bands run red at both peaks, that's a real link between site location and congestion; if the colour pattern doesn't track those two bands, the story needs a different explanation than construction.
Screenshots taken on the same typical Monday, four hours apart, using the Google Maps traffic layer centred on Hamburg's inner city and the A7 corridor.
The same two bands — Mitte's inner ring and the A7 through Altona/Eimsbüttel — turn red at both the morning and evening peak, while midday and night stay clear. That pattern (peaks concentrated on the construction-heavy corridors, not spread evenly across the city) is consistent with a structural link between the worksites and congestion, though these four snapshots are one weekday and not a substitute for a matched before/during time series.
Street parking competes with driving lanes for scarce road space. This Kepler.gl visualisation maps the density of public on-street parking across Hamburg's neighbourhoods. In dense districts, every parked car narrows the usable carriageway — a hidden constraint amplifying any disruption caused by roadworks or detours.
Signalised intersections are the pacemakers of the network — they decide how quickly traffic clears a junction. When a nearby construction site changes flow patterns, signals that aren't re-timed quickly create the queues drivers actually experience as "the jam".
The Federal Accident Atlas (Unfallatlas) plots reported passenger-car accidents in Hamburg. Overlaid on the other layers, it highlights risk hotspots where congestion, complex junction geometry and roadworks combine — the locations where safer detour planning matters most.
Accidents aren't a side topic — they're the downstream signal that ties every other layer together. Here's how it fits into the argument:
A data story is only as honest as the choices behind it. This section documents how sources were chosen and narrowed down, what filtering and processing was applied before anything reached a chart, and which visualisation conventions we followed so the numbers aren't dressed up to look more certain than they are.
Data selection
We prioritised official, open, city-maintained sources over commercial or modelled ones wherever a like-for-like option existed: Hamburg Geoportal's Baustellen layer over third-party construction trackers, HOCHBAHN/HVV's own published service-quality indicators over estimated punctuality, and the Hamburg Luftmessnetz station network over regional air-quality averages. Sources were logged in a shared data catalogue (see the sub-page linked in the navigation) with maintainer, access level, relevance and known bias noted for every entry before any of them were used in a chart — so weaker sources (paywalled press coverage, commercial indices) were flagged for context rather than quantitative claims from the outset.
Two datasets we deliberately did not use for headline numbers: the F+B Wohn-Index (commercial, non-transparent methodology) and paywalled Abendblatt reporting — both stayed catalogued as context sources only, never as the basis for a statistic in this report.
Filtering & processing
Charts were built in Datawrapper using its accessibility defaults (colour-blind-safe palettes, direct value labelling, responsive embeds) rather than custom chart libraries, so every reader sees the same chart regardless of device. Beyond the tool defaults, three editorial rules ran through every visual on this site:
Each visual below isolates one mechanism translating roadworks into congestion, unreliable transit or worse air quality. All charts are embedded from Datawrapper using open Hamburg sources — HVV/HOCHBAHN service indicators, the City's air-quality network, and municipal traffic counts.
All four figures above are measured values or figures computed directly from a chart on this page — the 93.7% and 41 µg/m³ are measured station/service readings, the 60% is the sum of the three highest bars in the district chart below, and the 21% is from the MiD 2017 survey (see Section 06).
Construction is one input among several that shift travel time and emissions over a year, and our dataset does not let us separate them cleanly:
We have not controlled for these, so any single-corridor reading in this report should be treated as suggestive, not causally attributed to construction alone.
In the HVV, a bus is punctual if it deviates by less than five minutes — the "5-minute punctuality" indicator. This is a city-wide average, not a Mitte/Altona/Eimsbüttel-specific figure, so we can only say the citywide dip coincides in time with the parallel works on those corridors — not that those corridors caused it. Worth requesting from HOCHBAHN: a line-level breakdown for routes crossing the inner ring and the A7 corridor, which would let this chart speak to the map directly instead of by inference.
Reliability captures cancellations and missed runs, unlike punctuality. Same caveat as Chart 01: this is the HVV-wide figure, so read it alongside the punctuality chart as one combined "is transit holding up" signal, not as evidence tied to the specific districts mapped earlier.
Habichtstraße, the station behind the 41 µg/m³ figure earlier, sits in Barmbek-Nord — Bezirk Hamburg-Nord, the district ranked fourth by active-site share in the bar chart above. That's a genuine, named link between where sites cluster and where a real exceedance is measured, rather than an illustrative number picked at random. Long-term city-wide NO₂ trends are falling regardless (fleet renewal), so the honest read is: background pollution is improving, and this one traffic-adjacent station still exceeds the limit — both things are true at once.
Average weekday vehicle and bicycle traffic during the construction period. This closes the loop back to Layer 01 and Layer 03 (parking): if construction is redistributing rather than removing demand, we'd expect volume on parallel side-streets near Mitte/Altona/Eimsbüttel to rise while the worksite's own street falls — the pattern H2 (traffic management) and H8 (parking/delivery) both predict. This chart shows aggregate volumes, not a matched before/after per street, so it's consistent with that redistribution story rather than proof of it.
Reworked from the onion model into a radial diagram: the core plans and approves, the middle operates the network day-to-day, the outer lives with the outcome. Every recommendation in the Conclusion is addressed to a specific ring.
The authority that approves, funds and coordinates. Recommendations here are about governance and coordination.
The operators keeping the network moving. Recommendations here are about service continuity and detour design.
The people, businesses and environment living with the outcome. Recommendations here are about transparency and lived experience.
Not every hypothesis held up equally well, and that's a legitimate result on its own. Each finding below is tagged by how much the available data actually supports it: strong means a measured, direct data link; moderate means a plausible pattern with a real confound we couldn't rule out; preliminary means correlational at best, and shouldn't be read as confirming causation.
Construction sites cluster in two bands — Mitte's inner ring and the A7 corridor through Altona and Eimsbüttel — confirmed numerically by the district breakdown (24% / 19% / 17% of all active sites, in that order).
Bus punctuality and reliability were lower during periods of intensive parallel works — consistent with H7, not a confirmation of it. The HVV figures are city-wide, so this is a correlation in time with the Mitte/Altona/Eimsbüttel works, not a corridor-specific causal link; a route-level breakdown would be needed to say more.
Long-run city-wide NO₂ trends improve, yet the Habichtstraße station — in Hamburg-Nord, the fourth-ranked district for site share — still exceeds the EU limit. This is our clearest direct link between a named construction-heavy district and a named, measured pollution reading.
Where several projects run on the same corridor, congestion appears to build up faster than the site count alone would suggest — but we have too few overlapping-site cases in the current data to say this rises above normal peak-hour variability. H4 is not rejected, but it isn't confirmed either; it needs more matched before/during comparisons than we have.
Sites tend to overlap in time and place more often than a coordinated schedule would predict — LSBG, utility operators and district authorities appear to plan largely within their own silos. This is a pattern in the site data, not a governance audit: we can't rule out that overlap is unavoidable (shared utility corridors, funding cycles), so we treat "poor coordination" as an open hypothesis rather than a confirmed cause of the congestion we observe.
Bicycle counts rise on some diverted corridors during disruption windows — a real signal in the traffic-volume data, though we can't yet tell whether this reflects genuine mode-shift or is a rerouting artefact of the counters themselves.
Hamburg must keep investing in infrastructure while keeping daily mobility efficient and sustainable. Construction contributes to congestion — especially on strategic corridors and when several worksites run at once — but it doesn't explain everything. Travel demand, network capacity, transit performance and urban design all play a part.
Rather than treating construction as the problem, our findings invite a shift in perspective: construction is an opportunity to rethink how the city manages mobility during change — through better coordination, stronger transit alternatives, smarter traffic management and transparent communication with residents.
The ultimate goal is not just to build new infrastructure, but to make the process of building compatible with sustainability, accessibility and quality of life.
Each recommendation is addressed to the stakeholder ring (Section 08) best placed to act on it, and grounded in a specific finding rather than the topic in general — so it's traceable back to evidence, not just to the SDG 11 theme.
| Recommendation | Addressed to | Grounded in |
|---|---|---|
| Publish a shared, city-wide construction calendar (type, duration, corridor) so overlapping projects on the same road can be sequenced rather than run in parallel. | Core · LSBG, Construction Coordination Office, District Authorities | Finding 5 (siloed scheduling) & Finding 1 (site clustering) |
| Default to transit- and cycling-priority detours on the Mitte/Altona/Eimsbüttel bands before car detours are drawn up. | Middle · HOCHBAHN/HVV, LSBG, Contractors | Finding 2 (bus punctuality dip) & Finding 6 (cycling uptake) |
| Add a route-level punctuality/reliability breakdown for lines crossing active work zones, published alongside the existing city-wide HVV indicators. | Middle · HOCHBAHN/HVV | Data-gap named in Section 07 and Finding 2 |
| Prioritise night-shift or off-peak scheduling for long-duration structural works (bridges, transit infrastructure) on the two clustering bands. | Core · Transport Authority, Contractors | Layer 01 duration table & Finding 1 |
| Give residents a public, real-time dashboard of active/planned sites with expected duration — the same attributes already in the Geoportal export, just surfaced. | Outer · Residents, local businesses, via City Government | Layer 01 (attribute gap) & stakeholder "transparency" ask |
Our findings are consistent with the direction, if not always the scale, of the international work-zone literature reviewed in Section 02: capacity loss and non-linear emissions increases are the same mechanism the Highway Capacity Manual and MOVES-based studies describe, just observed here through open municipal data rather than controlled simulation.
They also speak directly to two of the ten action fields in Hamburg's own Strategie Mobilitätswende (SUMP, 2023): construction and works coordination, and strengthening the Umweltverbund (walking, cycling, transit) toward its 80%-of-trips-by-2030 target. Our transit- and cycling-priority-detour recommendation directly supports that target; our coordination recommendation directly supports the plan's construction-coordination action field. Where our evidence is preliminary (Findings 4 and 5), it should be read as a prompt to strengthen monitoring under the existing SUMP framework — not as a call for new policy machinery.
In the spirit of the same transparency this project argues for in Hamburg's mobility data, here's a straight account of how AI tools were used while building this site — to help us learn and communicate better, not to do the thinking for us.
We, Group 3, declare that the following AI tools were used in the creation of this website:
What AI did not do: choose our research question, collect or analyse the underlying data, decide our hypotheses, or write our findings and conclusions. Every number, source, chart and interpretation on this site was chosen, checked and written by the group — AI was a tool for how we built and communicated the project, not what we concluded from it.