Editing Selby rail crash (section)

==Investigation==
[[Her Majesty's Railway Inspectorate]], part of the [[Health and Safety Executive]] (HSE), was notified of the crash at 06:55, and four inspectors arrived at the scene at 09:24. Another four inspectors arrived shortly thereafter. The HSE released an interim report on 6 March 2001.<ref name="first-report-guardian"/> It was conducted with the assistance of several entities, including GNER and Freightliner, the operators of the trains; [[Railtrack]], the owner of the railway infrastructure; both [[South Yorkshire Police|South]] and [[North Yorkshire Police]], the [[British Transport Police]], and the [[Highways Agency]].<ref>{{cite journal |date=1 October 2001 |title=UK Health & Safety Executive publishes interim report into Selby rail crash |url=https://www.emerald.com/insight/content/doi/10.1108/dpm.2001.07310dab.002/full/html |access-date=18 August 2024 |journal=Disaster Prevention and Management |volume=10 |issue=4 |doi=10.1108/dpm.2001.07310dab.002 |issn=0965-3562|url-access=subscription }}</ref> Investigators recovered the [[Train event recorder|data recorder]] of the freight locomotive, but the InterCity 225 had not yet been fitted with one under GNER's rolling programme of installation.{{sfn|HSE|2001|p=6}}

===Health and Safety Executive report===
The HSE released their final report into the accident on 4 December 2002.<ref>{{cite news |last=Wainwright |first=Martin |date=5 December 2002 |title=Rail crash inquiry calls for monitors |url=https://www.theguardian.com/uk/2002/dec/05/selby.railtravel |access-date=24 August 2024 |work=The Guardian |issn=0261-3077}}</ref> It stated that no defects were found with any of the railway infrastructure that could have contributed to the outcome of the crash, and that the motorway complied with the standards at the time.{{sfn|HSE|2002|p=7}}

At the time of the impact with the car, the InterCity 225 was estimated to have been travelling at a speed of {{convert|120|mph|abbr=on}} to {{convert|125|mph|abbr=on}}. After the initial collision, the train then slowed to an estimated speed of {{convert|88|mph|abbr=on}} prior to the impact with the freight train, which was estimated to have been travelling at {{convert|54|mph|abbr=on}}. With an estimated closing speed of {{convert|142|mph|abbr=on}}, the collision between the trains is the highest-speed railway incident that has occurred in the UK.{{sfn|HSE|2002|pp=10, 34}}

The HSE report considered the [[crashworthiness]] of the trains. The InterCity 225 had 99 occupants at the time, out of a total seating capacity of 544. The DVT suffered major damage, which included the cab being separated from the underframe. This was due to it being more lightweight, and having a lower ride height than the freight locomotive it collided with.{{sfn|HSE|2002|p=22}} 45 of the 52 seriously injured passengers, and all eight fatalities, were travelling in the first five coaches. These coaches were noted for having the most damage, with their survival space reduced due to several impacts. Overall, the report noted that the coaches performed well in the crash, considering the high speeds involved.{{sfn|HSE|2002|p=24}}

The front and right side of the freight locomotive suffered heavy damage as a result of the impact with the DVT. Though the locomotive's body had a robust construction, which meant it only suffered relatively minor structural damage compared to the other vehicles involved in the accident, its design meant that it did not have any energy absorbing features. The report added "Its design does not appear to complement the crashworthy considerations that guide the construction of passenger vehicles that use the same rail network." Some freight wagons impacted the passenger train's coaches, causing extensive damage. This was partially due to the angular design of the wagons, which was also noted in the [[Southall rail crash]] investigation.{{sfn|HSE|2002|pp=28–29}}

Overall, the HSE report recommended that further research be conducted on how derailments occur and how the vehicles behave during them. This research should consider several factors, such as vehicle construction and safety features such as obstacle deflectors and bogie retention. It raised concerns relating to the design of freight trains as a whole, and mentioned that future designs should include features such as energy absorption functionality and bogie retention. It also stated that freight vehicles should avoid aggressive corners at possible interfaces with passenger trains in the event of a crash. [[Tripwires]] similar to those already used close to airport runways were mentioned as potentially warranting further research. These could detect the incursion of a road vehicle onto railway property, and interface with different train control systems. Although this may not help when a train is already close to the incident, it could provide an early warning to the incursion in other cases.{{sfn|HSE|2002|pp=37–38}}

===Health and Safety Commission and Highways Agency reports===
In response to the crash, the [[Health and Safety Commission]] (HSC) and [[Highways Agency]] (HA) created working groups to look into the risk of incursions onto railway property.{{sfn|DfT|2003|p=1}} The HSC reviewed previous incidents of road vehicles entering railway property other than at level crossings, and the HA reviewed the use of safety barriers on the nearside of major roads.{{sfn|HSC|2002|pp=5-6}}{{sfn|HA|2002|loc=chpt. 1}} Both groups published their reports on 25 February 2002.{{sfn|DfT|2003|p=1}}

The HSC identified that of the ~50 vehicles that incur onto railway property each year, ~5 are hit by a train. The working group stated that they expected a derailment to occur every one to two years, and that around one in a hundred of these derailments would then involve a collision with another train. They calculated that a similar event on the scale of the Selby rail crash would occur every 300–400 years. This resulted in a risk of ~0.1 deaths per year for occupants in trains, and ~0.4 deaths per year for occupants in road vehicles.{{sfn|HSC|2002|pp=8-11}}

The HSC working group compared the ~0.5 deaths per year previously calculated to the ~600 deaths per year in accidents where road vehicles leave the carriageway in general, and the ~20 deaths per year in general rail accidents. The working group concluded that the risk was tiny in comparison to other road risks, and small in comparison to other rail risks. They stated that considerations should still be made to reduce the rail risk, but that these should not outweigh other projects tackling higher risks to road and rail.{{sfn|HSC|2002|pp=12-14}}

Several methods of reducing risk were considered by the HSC working group. These included publicity campaigns to educate drivers, and tackling vandalism and careless behaviour. The report also included methods for reducing the severity of incursions onto railway property. These included measures in preventing incursions from reaching a point where a line is obstructed, preventing trains from hitting vehicles on the line, and reducing the severity of train collisions if they did occur.{{sfn|HSC|2002|pp=24-29,35}} Overall, the report made recommendations including enabling and carrying out risk assessments; appropriately responding to those risk assessments; and implementing improvements where necessary.{{sfn|HSC|2002|pp=33-36}}

The HA working group reviewed historical data for road vehicles leaving the carriageway towards the nearside, using police-submitted accident reports. These reports were only completed where injuries occurred, and the working group found that non-fatal accidents were under-reported to the police. From the data, they found that such accidents only accounted for 5% of all road accidents, but for 11% of annual road fatalities.{{sfn|HA|2002|loc=chpt. 2}} They calculated that around 0.4 vehicles a year would leave a major road to the nearside, and reach a rail line without being stopped by either a safety barrier or hazard. They found that unprotected nearside road hazards carry a higher economic risk than the hazard associated with a vehicle reaching a rail line, mainly due to the much lower likelihood of the latter occurring.{{sfn|HA|2002|loc=chpt. 3}}

The HA working group found no risk assessment process in place for providing safety barriers at specific sites. A standard did exist for where to place safety barriers. This standard was risk assessed at the design stage.{{sfn|HA|2002|loc=chpt. 3}} Overall, the HA working group found that the existing standards for the provision of safety barriers were sufficient, but made recommendations including further research to protect against collisions with nearside hazards; further developing risk assessments for providing safety barriers; and improving the collection of accident data, including accidents without any injuries.{{sfn|HA|2002|loc=chpt. 6}}

===Department for Transport report===
The [[Department for Transport]] (DfT) released a report in February 2003 outlining steps to manage the risk of future railway incursions by road vehicles, in response to the recommendations in the HSC and HA reports.{{sfn|DfT|2003|p=1}} It stated that all sites where vehicles could enter railway property, such as at bridges and parallel roads (but excluding level crossings) should be given a risk score. Forms and detailed criteria for different types of carriageways, and example cases for specific sites were included, and a mitigation spreadsheet was provided to choose cost-effective methods of reducing risks at identified sites. The spreadsheet calculates the cost-effectiveness of each option, but the report noted that engineers should still use their judgement on the suitability of the options at each site.{{sfn|DfT|2003|pp=3–5,10}} The DfT has since updated the report in September 2020 with improved guidance for risk assessing a road vehicle striking a bridge, and debris subsequently landing on the railway.<ref name="dft_report_updates">{{cite web |date=11 September 2020 |title=Managing accidental rail obstructions by road vehicles (TAL 06/03) |url=https://www.gov.uk/government/publications/managing-accidental-rail-obstructions-by-road-vehicles-tal-0603#full-publication-update-history |access-date=23 March 2025 |website=gov.uk }}</ref>

An increase in the score by two points doubles the risk. For example, a score of 90 has around a million times greater risk compared with a score of 50.{{sfn|DfT|2003|p=17}} Mitigations should be considered for sites with a score above 90, and the practicality of mitigations should be considered for scores between 70 and 90.{{sfn|RAIB|2014|p=14}}

{| class="wikitable mw-collapsible" style="margin: auto; border: none;"
|+{{nowrap|An excerpt from ''Form 1a Single carriageway road vehicle incursion risk ranking''{{sfn|DfT|2003|p=26}}}}
! Factor !! Options !! Score
|-
|f1{{efn|Score f1 on the basis of the corner of the bridge with the least containment.|group=dft_table}}
|'''Road Approach Containment'''
*Score 1 for acceptable (safety fence and/or heavily wooded side approaches, buildings or brick wall thicker than 450mm)
*Score 12 for inadequate (imperfect fencing and/or medium/lightly wooded approaches, 225mm thick brick wall)
*Score 24 for non-existent (No fencing, or only post and rail/wire, no significant vegetation)
|{{sdash}}
|-
|f2
|'''Road Alignment (Horizontal)'''
*Score 1 for straight road with at least 7.3m carriageway
*Score 3 for straight less than 7.3m carriageway or curved at least 7.3m carriageway
*Score 7 for curved road less than 7.3m carriageway
*Score 10 for reverse curves less than 7.3m carriageway
|{{sdash}}
|-
|style="border: none;"|
|style="border: none;"|
!Total
|-
|style="border: none;"|
|style="border: none;"|
|{{sdash}}
|}
{{notelist|group=dft_table}}

{| class="wikitable mw-collapsible" style="margin: auto; border: none;"
|+{{nowrap|An excerpt from ''Form 3 Example mitigation spreadsheet''{{sfn|DfT|2003|p=65}}}}
!Measure
!Approx cost per item (2002 prices)
!Approx max cost (£ per site)
!Estimated average effectiveness
!Suitable for site (Y/N)
!Approx. cost of measure for this site
!Estimated effectiveness of measure for this site
!cost-effectiveness (b/1000e) measure alone
!Selection order
!cost-effectiveness of measure given others selected
!Cost of selected measures
|-
|style="background-color: #a7b3d6;"|Low level safety barrier
|£100+/linear m
|40,000
|style="background-color: #bbd8d2;"| -95%
|Y
|15,000
| −95%
| −16
|style="background-color: #a7b3d6;"|1
|style="background-color: #a7b3d6;"|
|style="background-color: #a7b3d6;"|£15,000
|-
|High level safety barrier
|£500+/linear m
|200,000
|style="background-color: #bbd8d2;"| -95%
|Y
|15,000
| −50%
| −30
|
| −1,319
|
|-
|style="background-color: #a7b3d6;"|Rumble strips
|£5/lin. m or £25/strip
|250
| −35%
|Y
|250
| −35%
| −1
|style="background-color: #a7b3d6;"|2
|style="background-color: #a7b3d6;"|
|style="background-color: #a7b3d6;"|£250
|-
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|style="border: none;"|
|Total cost of selected measures
|style="background-color: #a7b3d6;"|£15,250
|}

The [[Rail Accident Investigation Branch]] (RAIB) referenced the DfT report in its report into a road vehicle incursion onto railway property in [[Aspatria, Cumbria]] in October 2013. In this incident, an unattended vehicle from a side road rolled down a hill, crossed a main road, and broke through a wooden fence surrounding the railway. The vehicle then rolled down a cutting and stopped across two tracks.{{sfn|RAIB|2014|p=7}} The RAIB report stated that the DfT report did not take into account how the risk of an incursion from a runaway vehicle travelling downhill from a side road would be combined with the risk from the neighbouring main road. This could lead to the risk at some sites being underestimated.{{sfn|RAIB|2014|p=17}} It made a recommendation to the DfT to review and amend its report to make sure that this risk is considered.{{sfn|RAIB|2014|p=23}} The DfT reported that it had implemented actions in response to this recommendation.{{sfn|RAIB|2018|p=2}} It updated the report in July 2017 to include guidance for risk assessing side roads joining neighbouring roads to the railway.<ref name="dft_report_updates"/>