Singapore Traffic Speed Bands
A saved CSV holding one day of road speeds in Singapore, which the supplied analysis attributes to the Land Transport Authority (LTA). Each row is one directional road link (one direction of travel along one road segment): its ID, road name, two endpoints and 288 speed bands, one every five minutes. A speed band is a speed range, from band 1 (0–9 km/h) to band 8 (70 km/h and above).
This file is the exact input to the analysis shown on the Key Findings page.
What the file records
- Spatial scope
- Singapore road links in this extract
- Time spacing
- Every five minutes, according to the column header
- Clock times
- 00:00 to 23:55, from each value’s position in the list
- Collection date
- Not recorded in the file
- Source
- LTA DataMall · Traffic Speed Bands
This copy was imported on 30 September 2026: the date it was received, not the date the traffic was measured.
According to the supplied HTML, the extract comes from version 3 of LTA’s Traffic Speed Bands API (the online service that publishes the readings). The API was queried repeatedly through the day, and the answers were rearranged into one series of readings per link.
| Record | What is known |
|---|---|
| Publisher attribution | LTA DataMall · Traffic Speed Bands |
| Original collector / access date | Not recorded in the CSV or the supplied package. |
| Measurement technology | The file does not identify per-link instruments, calibration or upstream processing. |
| Cadence | The header specifies five-minute intervals; there are 288 values in every row. |
| Timezone | Not stored. Clock times follow the CSV header, with no conversion to UTC. |
| Custody | Supplied ZIP archive → retained source CSV → analysis input, checked against its SHA-256 fingerprint. |
Each row is one directional road link and each column is one five-minute slot, so each cell holds one link’s speed band in one slot (1 = slowest, 8 = fastest, 0 = no reading). Road names and coordinates are kept beside the matrix as labels.
The matrix · 96 representative rows
96 links picked at even steps through the range of daily average band, slowest at the top. All 288 slots are shown, from 00:00 to 23:55. Colour runs from blue (slow) to red (fast); grey means no reading. Click a cell, or use Road link and Time slot, to select a link and a slot.
The selected link’s complete day
Horizontal axis: hour of the day. Vertical axis: speed band. The orange dot marks the selected time slot.
The preview shows 96 of the 58,780 rows; the analysis and Key Findings use all of them. The CSV’s index column # records the original row order and is not used in the analysis.
| CSV field | Meaning | Use |
|---|---|---|
| # | Original row index | Not used in the analysis |
| LinkID | Directional road-link identifier | Ties each profile to its cluster and map position |
| RoadName | Road name | A label; several links can belong to one road |
| OriginX / OriginY | Start longitude / latitude | In degrees; averaged with the end point to place the link on the map |
| EndingX / EndingY | End longitude / latitude | In degrees; averaged with the start point to place the link on the map |
| SpeedSeries(starts at 0:00 every 5min) | Text-encoded list of 288 integer bands | Converted into the 288 numbers the analysis uses |
The file has no road category (RoadCategory), exact km/h values, calendar date or per-slot timestamps.
The supplied source gives the speed range of bands 1–8, listed below. Band 0 is not described there; in this file it appears only in the 32 links whose readings are all 0, where it marks a missing reading. Band 8 has no upper limit, so averages cannot be converted to km/h; the charts show them in band units.
| Band | Meaning in supplied source | Readings | Share |
|---|---|---|---|
| 0 | Undocumented / no reading | 9,216 | 0.054% |
| 1 | 0–9 km/h | 35,926 | 0.212% |
| 2 | 10–19 km/h | 2,582,074 | 15.253% |
| 3 | 20–29 km/h | 5,020,334 | 29.656% |
| 4 | 30–39 km/h | 3,873,296 | 22.880% |
| 5 | 40–49 km/h | 2,399,789 | 14.176% |
| 6 | 50–59 km/h | 1,363,439 | 8.054% |
| 7 | 60–69 km/h | 727,857 | 4.300% |
| 8 | 70+ km/h | 916,709 | 5.415% |
Counts cover all 16,928,640 cells (58,780 links × 288 slots).
| Step | Decision | Effect |
|---|---|---|
| Parse | Convert each stored list into 288 integers | 58,780 × 288 numeric matrix; no time-slot imputation |
| Check | Require integer values from 0 through 8 and unique LinkID | Stop before analysis if the file has any other structure |
| No readings | Mark the 32 all-zero rows as missing readings; keep them, as the original analysis did | Cluster 0 contains only these 32 links |
| Features | Use the bands as they are; no centring or scaling | Similarity reflects both speed level and daily shape |
| Coordinates | Average each pair of endpoints | Used only to place links on the map |
| Outputs | Keep the link IDs and the original row order | Coordinates, cluster labels, curves and map stay matched to the same links |
| Check | Result |
|---|---|
| Rows / unique LinkID | 58,780 / 58,780 |
| Series length | Every row contains exactly 288 values |
| Missing CSV fields | 0 |
| All-zero rows | 32 links · 9,216 cells |
| Other profiles containing a 0 | 0 |
| Values in remaining profiles | Integers in the range 1–8 |
- These checks establish consistency of the supplied file, not completeness of the original collection.
- All-zero traces indicate absent readings here; the cause cannot be determined from the CSV.
- One undated day cannot establish a typical weekday, weekend, seasonal pattern or trend.
- Road links can share geography and road identity; rows are not independent observations of separate roads.
- No road-category labels or independent congestion measurements were supplied.
Analysis source: SGRoadSpeedbands.csv · 57.33 MB. The Workflow page explains how its daily profiles become the clusters, curves and map shown in Key Findings.
SHA-256
55769b334249dc35bb860a180bf5a8309f0cf2d24b927f852d18f7c67cbf7609
Source attribution: LTA DataMall · Traffic Speed Bands. Supplied under Singapore Open Data Licence v1.0; retain source attribution and the licence link when reusing the data. The data is provided as-is. This work does not imply LTA endorsement.
Source and licence pages checked 2026-09-30.