Evidence and indexing
Records are classified from the available event history. Missing start notifications do not establish that no physical works took place.
Discover current and upcoming works, explore historical change, compare streets and organisations, and understand what the records actually prove.
Only the visible map area is requested; where many records overlap, the view is capped and marked as partial. Proposed works may change or never start.
Multiple permits can share one work reference; counts above are distinct work references, not individual notifications.
| Road / ward | Interpretation | Promoter / authority | Management | Reported dates (UTC) | Work reference | |
|---|---|---|---|---|---|---|
| Loading… | ||||||
Watch how reported roadworks and traffic management evolved. Recorded-start evidence is the default; proposed-only schedules are optional.
Period-wide counting: A work is counted if its reported period overlaps any part of the day, 7-day interval or month, even briefly. Works spanning two periods appear in both; do not sum these counts into an annual unique-work total. The map shows up to 1,200 source periods per frame; a cap is clearly identified. Historical playback is a reconstruction from retained events. Date-only and proposed intervals are not independently verified activity. A missing stop is bounded and labelled estimated, not treated as open indefinitely. Signals and closures are recorded categories, not equipment or traffic-delay measurements.
Rank streets by distinct work references rather than raw notifications; examine repeat interventions and non-overlapping, event-bounded work windows.
Street grouping uses USRN where supplied, otherwise the supplied street name within a ward. Numbers are not a measure of congestion or journey delay.
Compare reported work volumes, recorded starts, traffic-management types, geographical coverage and event-bounded activity windows.
Shared filters and selected time scope apply to organisation totals.
Promoters and highway authorities are separate roles. Statistics count distinct official work references, not separate digs or measured road-user impact.
Evidence-labelled summaries and local community bulletins; their filter controls are described below. Useful for neighbourhood conversations and social posts.
The older summary cards follow shared filters. Community bulletins use their own period, geography and evidence controls, independently of the shared filters.
Every story carries its source, review status and limitations. Reported traffic management does not establish measured delays or exact operating hours.
Repeat works, disruption calendars, possible coordination, planning accuracy, shareable local reports and seasonal trends — grounded in distinct work references and retained source evidence.
A recorded work-start is evidence of a reported start, not an independent site observation. Reported temporary-light and closure windows are not measured equipment operating time or traffic delay. Historic recovered archives can be incomplete.
Records are classified from the available event history. Missing start notifications do not establish that no physical works took place.
Recovered archive months reviewed by the site operator are included in published results. Original ZIP recovery flags and hashes are retained for independent audit.
Source documents: DfT Street Manager Open Data: messages, topics, archives and subscription methods · DfT official guidance · Government transport data listing · OpenStreetMap map attribution and licence. These links document the underlying sources, not the correctness or completeness of this local reconstruction.
Reporting freshness: The page's as-of timestamp and operational indexes are distinct from the latest independently received DfT SNS notification. A silent Activity feed must be investigated rather than described as working.
Historical monthly/weekly intervals: Time Machine periods now count every distinct source work reference whose reported window overlaps that period. A work can be counted in several successive periods; monthly totals are not additive. Map locations and capped record lists are indicative.
Distinct work IDs combine permits with the same official work reference, falling back to an individual permit reference if the work ID is missing. They are not a count of independent excavation sites.
Works that happened require explicit recorded start evidence; a supplied actual date without the corresponding event is shown separately. A cancelled or refused permit with no recorded start is not proof that no physical activity occurred.
Temporary signals and closures are reported traffic-management categories. They do not measure when physical equipment was switched on, actual traffic delay or observed hours of disruption.
Historic completeness varies by month, feed and archive integrity. The dashboard includes operator-reviewed recovered archive months; missing notifications are still possible, and no-activity claims require separate evidence.