25:["$","title",null,{"children":"Road safety in Tower Hamlets | Emile Mathieu | Observable"}] 4b:T868,# Road safety in Tower Hamlets ## Road collisions statistics in Tower Hamlets before and after the Low Traffic Neighbourhoods The Department for Transport (Dft) recently released their updated *Casualties in Greater London Report* ([2023](https://content.tfl.gov.uk/casualties-in-greater-london-2023.pdf)), based on their updated [London collision statistics](https://www.data.gov.uk/dataset/cb7ae6f0-4be6-4935-9277-47e5ce24a11f/road-safety-data) alongside a [dashboard](https://app.powerbi.com/view?r=eyJrIjoiMTIzYzRiZDMtZDZiYS00OWYyLTk3YTQtN2RkMGNiZTRmOTJkIiwidCI6IjFmYmQ2NWJmLTVkZWYtNGVlYS1hNjkyLWEwODljMjU1MzQ2YiIsImMiOjh9) to visualise said statistics. This report opens with: "In 2023 London experienced the fewest people killed (95) on our roads outside of the pandemic." (p. 3)" This is obviously great news, but there were still 3,710 people killed or seriously injured in 2023 in Greater London, so we haven't reached vision zero yet! ### What we did We wanted to follow up on this report, and in particular see whether the Bethnal Green/Weavers LTN had an impact on safety, and in particular on road collisions. ### How we did it We've based our analysis on Tower Hamlets' LTNs being operational from 1st July 2021 — though some of the new road layouts were being modified well after this date. We looked at overall road collision in the LTN, across the borough, and also across neighbouring inner boroughs (i.e. Hackney and the City of London). For the data and analysis, we exclusively used the DfT updated [road collision data](https://www.data.gov.uk/dataset/cb7ae6f0-4be6-4935-9277-47e5ce24a11f/road-safety-data). We've selected data from 1 Jan 2017, up to 31 Dec 2023 as that's the latest available data. *Disclaimer: This analysis was done by volunteers working with the [Save Our Safer Streets campaign](https://heartbg.uk). Where possible, we've sought advice and comment from third parties with relevant experience or expertise. We ask that Tower Hamlets council undertake professional and verified analysis to understand the impacts of the LTNs and to determine the borough's needs and how best to serve its communities.*4c:T57b,

Key Results

Serious collisions have strongly decreased: Within the area of the LTN there has been a significant decrease in serious collisions relative to the rest of Tower Hamlets and adjacent inner boroughs since the LTN opened. Since the scheme has been introduced in July 2021, there has only been one serious collision inside the LTN, located at the edge of the area and outside the area covered by LTN control measures.

Looking at boundary roads, there are no statistically significant changes in number of collision after the new road layouts: We used a test known as a Boschloo exact test to check whether there were statistical differences between periods before and after the LTNs. The results of this analysis indicate there were no statistical differences in response times. In other words, LTNs did not lead to increases (nor decreases) road safety on boundary roads.

4d:T572,Plot.plot({ color: { type: "categorical", legend: true, scheme: "pastel1", // legend: "ramp", tickFormat: (d) => severity_int_to_str(d), domain: [1, 2, 3], }, y: { // label: "↑ No. of collisions in" + " " + wardSelect1, label: "↑ No. of casualties in" + " " + wardSelect1, grid: true }, x: { label: "Month and Year →", thresholds: "d3.utcMonth", tickFormat: d3.utcFormat("%b-%y") }, marks: [ // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", filter: (d) => d.Borough === boroughCompare, fill: "Gray", opacity: 0.2})), // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", filter: (d) => d.WardName === wardSelect1, fill: "IncidentGroup"})), // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", filter: (d) => d.WardName === wardSelect1, fill: "accident_severity"})), Plot.rectY(collisions, Plot.binX({y: "sum" }, {y: "number_of_casualties", x: "Month", filter: (d) => (wardSelect1 === "") ? true : d.WardName === wardSelect1 , fill: "accident_severity"})), // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", fill: "IncidentGroup"})), Plot.ruleY([0]), Plot.ruleX([new Date(preLTN)], {strokeOpacity: 1., stroke: "red"}), Plot.ruleX([new Date(postLTN)], {strokeOpacity: 1., stroke: "red"}) ], marginBottom: 45, marginTop: 35, insetLeft: 5 })4e:T829,beforeAfterResponseTimes1 = Plot.plot({ x: { label: "", // domain: ["pre LTN", "in between", "post LTN"], domain: ["pre LTN", "post LTN"], // tickFormat: (d) => d.split("-")[0] }, // y: { // label: wardSelect1 + " " + "Response times ↑ (secs)", // grid: true, // inset: 3, // type: "utc", // tickFormat: d3.utcFormat("%M:%S"), // domain: [0,380000] // }, y: { // label: "↑ No. of collisions in" + " " + wardSelect1, label: "↑ No. of casualties in" + " " + wardSelect1, grid: true }, color: { type: "categorical", scheme: "pastel2" }, marks: [ // Average response time for all TH // Plot.ruleY(collisions, { // y: meanResposeTimeMS, // x1: "Period", // x2: "Period", // stroke: "red", // strokeWidth: 0.5, // strokeOpacity: 0.1 // }), // Average response times for specificied ward before and after LTNs Plot.barY(collisions, Plot.groupX( {y: "sum"}, { x: "Period", y: "number_of_casualties", sort: null, fill: "Period", fillOpacity: 0.3, filter: (d) => ((wardSelect1 === "") ? true : d.WardName === wardSelect1) & d.utcTime >= new Date('2017-12-01') })), // Average response times before and after LTNs // Plot.barY(collisions, // Plot.groupX( // {y: "mean"}, { // x: "Period", // y: "number_of_casualties", // sort: null, // fill: "Period", // fillOpacity: 0.3, // })), // Plot.dotY(collisions, // Plot.groupX( // {y: "sum"}, { // x: "Period", // y: "number_of_casualties", // stroke: "Period", // strokeOpacity: 1, // fill: "black", // strokeWidth: 3, // r: 4, // })) ], marginLeft: 50, // space to the left of the chart marginBottom: 30, // space below the chart insetBottom: 33, // space between the x-axis and the marks insetLeft: 5, width: 400, // 240 })4f:T608,{ const severityIndexSelect = get_severity_index(severitySelect) return Plot.plot({ color: { type: "categorical", legend: true, scheme: "pastel2", // legend: "ramp", domain: ["Inside LTN", "Boundary LTN", "Rest of Tower Hamlets"] }, y: { // label: "↑ No. of collisions in" + " " + wardSelect1, label: "↑ No. of " + severitySelect + " casualties", grid: true, }, x: { label: "Month and Year →", thresholds: "d3.utcMonth", tickFormat: d3.utcFormat("%b-%y") }, marks: [ // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", filter: (d) => d.Borough === boroughCompare, fill: "Gray", opacity: 0.2})), // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", filter: (d) => d.WardName === wardSelect1, fill: "IncidentGroup"})), // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", filter: (d) => d.WardName === wardSelect1, fill: "accident_severity"})), Plot.rectY( collisions, Plot.binX({y: "sum" }, {y: "number_of_casualties", x: "Month", filter: (d) => (['Inside LTN', 'Boundary LTN', 'Rest of Tower Hamlets'].includes(d.Location)) & (severityIndexSelect.includes(d.accident_severity)), fill: "Location"})), // Plot.rectY(collisions, Plot.binX({y: "count" }, {x: "Month", fill: "IncidentGroup"})), Plot.ruleY([0]), Plot.ruleX([new Date(preLTN)], {strokeOpacity: 1., stroke: "red"}), Plot.ruleX([new Date(postLTN)], {strokeOpacity: 1., stroke: "red"}), ], marginBottom: 45, marginTop: 35, insetLeft: 5 }) }50:T997,{ const severityIndexSelect = get_severity_index(severitySelect) return Plot.plot({ x: { label: "", // domain: ["pre LTN", "in between", "post LTN"], domain: ["pre LTN", "post LTN"], // tickFormat: (d) => d.split("-")[0] }, // y: { // label: wardSelect1 + " " + "Response times ↑ (secs)", // grid: true, // inset: 3, // type: "utc", // tickFormat: d3.utcFormat("%M:%S"), // domain: [0,380000] // }, y: { // label: "↑ No. of collisions in" + " " + wardSelect1, label: "↑ No. of " + severitySelect + " casualties", grid: true }, color: { legend: true, type: "categorical", scheme: "pastel2", domain: ["Inside LTN", "Boundary LTN", "Rest of Tower Hamlets"] }, marks: [ // Average response time for all TH // Plot.ruleY(collisions, { // y: meanResposeTimeMS, // x1: "Period", // x2: "Period", // stroke: "red", // strokeWidth: 0.5, // strokeOpacity: 0.1 // }), // Average response times for specificied ward before and after LTNs Plot.barY(collisions, Plot.groupX( {y: "sum"}, { x: "Period", y: "number_of_casualties", sort: null, fill: "Location", fillOpacity: 1., filter: (d) => (['Rest of Tower Hamlets'].includes(d.Location)) & (severityIndexSelect.includes(d.accident_severity)) & d.utcTime >= new Date('2017-12-01') })), Plot.barY(collisions, Plot.groupX( {y: "sum"}, { x: "Period", y: "number_of_casualties", sort: null, fill: "Location", fillOpacity: 0.9, filter: (d) => (['Boundary LTN'].includes(d.Location)) & (severityIndexSelect.includes(d.accident_severity)) & d.utcTime >= new Date('2017-12-01') })), Plot.barY(collisions, Plot.groupX( {y: "sum"}, { x: "Period", y: "number_of_casualties", sort: null, fill: "Location", fillOpacity: 0.8, filter: (d) => (['Inside LTN'].includes(d.Location)) & (severityIndexSelect.includes(d.accident_severity)) & d.utcTime >= new Date('2017-12-01') })), // })), ], marginLeft: 50, // space to the left of the chart marginBottom: 30, // space below the chart insetBottom: 33, // space between the x-axis and the marks insetLeft: 5, width: 400, // 240 }) }51:T80b,{ const LTN = new GeoJSON().polygon([ [-0.057133, 51.532572], [-0.064693, 51.531395], [-0.072525, 51.530846], [-0.074711, 51.529827], [-0.074914, 51.529258], [-0.076487, 51.527835], [-0.077971, 51.526941], [-0.078077, 51.526306], [-0.077161, 51.524729], [-0.077214, 51.523735], [-0.075217, 51.523701], [-0.074064, 51.523917], [-0.069619, 51.525771], [-0.064638, 51.526525], [-0.055430, 51.527547], [-0.05610, 51.5302544] ]).featureCollection; const radius = 0.1 / zoom const circle = d3.geoCircle().center([-0.065216,51.528524]).radius(radius).precision(2)() const severityIndexSelect = get_severity_index(severitySelect2) return Plot.plot({ width: width, height: width * 0.8, projection: { type: "mercator", // domain: statemesh, domain: circle, }, // color: {scheme: "YlGnBu"}, color: { type: "categorical", legend: true, // scheme: "pastel1", legend: "ramp", domain: [1, 2, 3], range: ['#ff0000', '#ff7400', '#ffc100'], tickFormat: (d) => d == 1 ? "Fatal" : (d == 2 ? "Serious" : "Slight") }, style: "overflow: visible;", marks: [ // Plot.density(collisions, {x: "longitude", y: "latitude", bandwidth: 1, fill: "density", filter: (d) => d.Period === periodSelect}), Plot.geo(statemesh, {strokeOpacity: 0.5}), // Plot.geo(nation, {stroke: "#CCC", strokeOpacity: 0.3}), // Plot.dot(collisions, {x: "longitude", y: "latitude", r: 1, fill: "currentColor", filter: (d) => d.accident_severity == 1}) Plot.dot(collisions, {x: "longitude", y: "latitude", r: 'number_of_casualties', fill: "accident_severity", filter: (d) => d.Period === periodSelect & severityIndexSelect.includes(d.accident_severity)}), // & d.latitude >= 51.510000 & d.latitude <= 51.54 & d.longitude >= -0.09000 & d.longitude <= -0.035000 Plot.geo(LTN, { fill: "green", opacity: 0.3},), // Plot.geo(circle, {stroke: "red", strokeWidth: 2}), // Plot.dot(collisions, {x: "longitude", y: "latitude", r: 1, fill: "currentColor"}), ] }) }52:T708,

	Before (Jan 2017 - Jun 2020)		After (Jul 2021 - Dec 2023)

Severity	Number inside LTN	Number elsewhere in Tower Hamlets	Number inside LTN	Number elsewhere in Tower Hamlets	Ratio	p-value
Fatal & Serious	9	586	1	430	0.15	<0.05
Slight	71	3373	51	2334	1.04	0.83

Severity	Number on boundary roads	Number elsewhere in Tower Hamlets	Number on boundary roads	Number elsewhere in Tower Hamlets	Ratio	p-value
Fatal & Serious	70	586	48	430	0.93	0.74
Slight	358	3373	219	2334	0.88	0.17

53:[{"id":"d5c3efcee6f37b0969a4719b9e4de65a4be2bc9beef0c7479cbe116c681c9e91f27dd247b1f378251c451dac6e8d8f6a7c5899472cd1e251e9c8dbaabaf86442","href":"https://static.observableusercontent.com/files/d5c3efcee6f37b0969a4719b9e4de65a4be2bc9beef0c7479cbe116c681c9e91f27dd247b1f378251c451dac6e8d8f6a7c5899472cd1e251e9c8dbaabaf86442","name":"image.png","type":"image/png","lastModified":1728903754280.978,"size":1240933},{"id":"c266c323dfd92909efd581fab6d5b76f5dc2081b00a5ab8f1dedc2b73cc7c85c128b54ae6924a4a79aa932987ec235efc2336682eebbd336b19806d617a808e6","href":"https://static.observableusercontent.com/files/c266c323dfd92909efd581fab6d5b76f5dc2081b00a5ab8f1dedc2b73cc7c85c128b54ae6924a4a79aa932987ec235efc2336682eebbd336b19806d617a808e6","name":"TH-LTN-collision@3.csv","type":"text/csv","lastModified":1728853908592.885,"size":1954888},{"id":"b5febb5407ce19be25477d4011e4408a74e9044cba4efbed57e50b99b0c44b716528e1adc02ff3790a579b11d4ae4f365e670ce2ea45d850446624488a380130","href":"https://static.observableusercontent.com/files/b5febb5407ce19be25477d4011e4408a74e9044cba4efbed57e50b99b0c44b716528e1adc02ff3790a579b11d4ae4f365e670ce2ea45d850446624488a380130","name":"londonBoroughs.json","type":"application/json","lastModified":1728857320568.398,"size":102115}] 47:{"id":"5cc2f85e1154f95c","slug":"road-safety","title":"Road safety in Tower Hamlets","thumbnail":"f7c2e8fdca0f7c0de1e2947f4d298ffd1787b3739e7638a52a33eb679aa8d15a","access_level":"public","create_time":"$D2024-09-30T18:53:02.892Z","update_time":"$D2024-10-27T19:44:19.316Z","view_time":null,"first_public_time":"$D2024-09-30T18:53:02.892Z","has_unpublished_edits":null,"fork":true,"stars":0,"starred":false,"owner":{"type":"team","id":"d8d4735a38ee536e","login":"emile-mathieu","name":"Emile Mathieu","avatar_url":"https://avatars-next.observableusercontent.com/35ffe7a03ed5ec3496d971ae25cee385b584b5c27f1ca97b5fde0152339cd71a","tier":"starter_2024","bio":"","followed_by_you":false,"follows_you":false},"version":2408,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"fcfd6e2483c61df2","login":"emile-mathieu","name":"Emile Mathieu","avatar_url":"https://avatars-next.observableusercontent.com/fdaef8cca9f4b97113845f3dff9333b3fb28cd25f2115478976d180788ff3ade","tier":null,"bio":"","followed_by_you":false,"follows_you":false}],"members":[{"type":"individual","id":"fcfd6e2483c61df2","login":"emile-mathieu","name":"Emile Mathieu","avatar_url":"https://avatars-next.observableusercontent.com/fdaef8cca9f4b97113845f3dff9333b3fb28cd25f2115478976d180788ff3ade","tier":null,"role":"editor"},{"type":"team","id":"d8d4735a38ee536e","login":"emile-mathieu","name":"Emile Mathieu","avatar_url":"https://avatars-next.observableusercontent.com/35ffe7a03ed5ec3496d971ae25cee385b584b5c27f1ca97b5fde0152339cd71a","tier":"starter_2024","role":"viewer"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"Road safety in Tower Hamlets","cells":[{"id":0,"value":"$4b","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":1257,"value":"$4c","pinned":false,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2260,"value":"## Liveable Streets \n\nThe Liveable Streets programme was introduced under the previous Tower Hamlet mayorship, with the aim \n> ‘to reduce people making short-cuts through residential streets to eliminate “rat runs” and encourage more sustainable journeys to improve air quality and road safety.’\n\nPractically, this involves a jigsaw of measures, such as traffic filters, pocket parks, dedicated cycle lanes etc.\nThe vast majority of these measures were eventually _not_ implemented, apart from the Bethnal Green area where ~80% were enacted here.\nSee map below.\n","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2255,"value":"FileAttachment(\"image.png\").image({width: 1500})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1877,"value":"## Data analysis\n\n We plot the aggregate number of collisions over 6 month periods, split into different collision _severities_, and with the option to filter by ward.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":1130,"value":"viewof wardSelect1 = Inputs.select(wardsAll, {label: \"Choose a Ward\"})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1600,"value":"$4d","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1635,"value":"$4e","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1916,"value":"We break the collision data with respect to the _severity_ of the collision, as defined [here](https://www.gov.uk/government/publications/reported-road-casualty-statistics-background-quality-report/reported-road-casualty-statistics-background-quality-report). You can filter the data via the toggle below:","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":1717,"value":"viewof severitySelect = Inputs.select(['All', 'Fatal', 'Fatal / Serious', 'Serious', 'Serious / Slight', 'Slight'], {value: 'Fatal / Serious', label: \"Collision severity\"})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1710,"value":"$4f","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1939,"value":"We can observe that since the LTN has been in place (2nd vertical red line), there has been only 1 serious collision inside the LTN.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2324,"value":"$50","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2233,"value":"Below is a geospatial visualisation of road collisions, where the colour indicates the severity of the collision and dot radii the number of casualties. The boundary of the LTN studied is indicated in green.\nYou can toggle data from before or after the traffic measures, and adjust the map zoom.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2353,"value":"viewof severitySelect2 = Inputs.select(['All', 'Fatal', 'Fatal / Serious', 'Serious', 'Serious / Slight', 'Slight'], {label: \"Collision severity\"})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2018,"value":"viewof periodSelect = Inputs.select(['pre LTN', 'post LTN'], {label: \"Time period\"})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2204,"value":"viewof zoom = Inputs.range([1., 10], {step: 0.01, label: \"Zoom\"})","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1946,"value":"$51","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1912,"value":"We followed the methodology from [The Impact of Introducing Low Traffic Neighbourhoods on Road Traffic Injuries\n](https://findingspress.org/article/18330-the-impact-of-introducing-low-traffic-neighbourhoods-on-road-traffic-injuries) which studied the impact of low traffic neighbourhoods in Waltham Forest on road safety using the same collision dataset from DfT.\nWe use an [Boschloo's exact test](https://en.wikipedia.org/wiki/Boschloo%27s_test) to test the hypothesis that the number of collisions in an area did change significantly relative to another area after the introduction of the LTN.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":1831,"value":"$52","pinned":false,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2241,"value":"We observe that serious collisions have statistically significantly decreased inside the LTN relative to the rest of the borough, with only [one serious collision](https://www.cyclestreets.net/collisions/reports/2023010482104/) since the introduction of the traffic calming measures, which happened at south western edge of the LTN, at the junction between Redchurch St and Club Row, that is [outside any of the measures/interventions that reduce traffic](https://bethnalgreenlondon.co.uk/wp-content/uploads/sites/2/2023/10/image.jpeg).\nHowever, slight collisions have stayed stable thorough.\nFurthermore, we observe that on boundary roads overall number of collisions have not changed.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":1254,"value":"## See more analysis\n- [London Fire Brigade response times](https://observablehq.com/@soss/lfb-response-times-in-tower-hamlets)\n- [Ambulance response times](https://i.heartbg.uk/LAS-response-time-analysis)\n- [Air quality analysis](https://i.heartbg.uk/air-quality-analysis)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":682,"value":"## Appendix","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2087,"value":"import { GeoJSON } from \"@jwolondon/geojson-generator\"","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1948,"value":"london = FileAttachment(\"londonBoroughs.json\").json()","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1990,"value":"tower_hamlet_geojson = {\nconst tower_hamlet_geojson = london\n// tower_hamlet_geojson.bbox = [-0.5103750689005356, 51.28676016315083, 0.3340155643740317, 51.69187411690989]\ntower_hamlet_geojson.bbox = [-0.078000, 51.510000, -0.035000, 51.536322]\ntower_hamlet_geojson.objects.boroughs.geometries = london.objects.boroughs.geometries.filter(function(x) { return (x.id == \"Tower Hamlets\") | (x.id == \"Hackney\") | (x.id == \"City of London\") });\nreturn tower_hamlet_geojson\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1959,"value":"statemesh = topojson.mesh(tower_hamlet_geojson, tower_hamlet_geojson.objects.boroughs)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1902,"value":"wardsAll = [\"\", \"BETHNAL GREEN\", \"BLACKWALL & CUBITT TOWN\", \"BOW EAST\", \"BOW WEST\", \"BROMLEY NORTH\", \"BROMLEY SOUTH\", \"CANARY WHARF\", \"ISLAND GARDENS\", \"LANSBURY\", \"LIMEHOUSE\", \"MILE END\", \"POPLAR\", \"SHADWELL\", \"SPITALFIELDS & BANGLATOWN\", \"ST DUNSTAN'S\", \"ST. KATHARINE'S & WAPPING\", \"ST PETER'S\", \"STEPNEY GREEN\", \"WEAVERS\", \"WHITECHAPEL\"]","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1693,"value":"collisions = FileAttachment(\"TH-LTN-collision@3.csv\").csv({typed: true})","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2315,"value":"d3.sum(collisions.filter(d => d.Period === \"pre LTN\"), d => d[\"number_of_casualties\"])","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":2253,"value":"// collisions_mean = d3.sum(collisions.filter(d => d.Period === \"pre LTN\"), d => d[\"number_of_casualties\"])\ncollisions_mean = {\n const end = d3.max(collisions.filter(d => d.Period === \"pre LTN\"), d => d[\"utcTime\"])\n const start = d3.min(collisions.filter(d => d.Period === \"pre LTN\"), d => d[\"utcTime\"])\n const nb_year = (end - start) / (365 * 24 * 60 * 60 * 1000)\n return d3.sum(collisions.filter(d => d.Period === \"pre LTN\"), d => d[\"number_of_casualties\"] / nb_year)\n}","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1676,"value":"preLTN = new Date('2020-06-01') // Assumed date LTNs installed","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":161,"value":"postLTN = new Date('2021-07-01') // Assumed date LTNs installed","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1708,"value":"// responseTimesMinusE2 = responseTimes.filter(d => d.Postcode_district !== \"E2\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1619,"value":"collisions","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"","database":"$undefined"},{"id":2377,"value":"function severity_int_to_str(x){\n if (x == 1) {\n return 'Fatal'\n } else if (x == 2) {\n return 'Serious'\n } else if (x == 3) {\n return 'Slight'\n } else {\n return ''\n }\n \n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":1788,"value":"function get_severity_index(severity_text) {\n let severityIndexSelect\n if (severity_text == 'Fatal') {\n severityIndexSelect = [1]\n }\n else if (severity_text == 'Fatal / Serious') {\n severityIndexSelect = [1, 2]\n }\n else if (severity_text == 'Serious') {\n severityIndexSelect = [2]\n }\n else if (severity_text == 'Serious / Slight') {\n severityIndexSelect = [2, 3]\n }\n else if (severity_text == 'Slight') {\n severityIndexSelect = [3]\n }\n else if (severity_text == 'All') {\n severityIndexSelect = [1, 2, 3]\n }\n else {\n severityIndexSelect = [1, 2, 3]\n }\n return severityIndexSelect\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W53","stdlib":"1"}} 24:null