The agenda (PSI-02-04) circulated by the chairman immediately prior to the meeting was adopted.

The minutes (PSI-02-03) and progress report (PSI-02-02) from the first meeting held in Bonn, Germany and circulated by the chairman prior to the meeting were agreed.

The minutes (PSI-02-03) and progress report (PSI-02-02) from the first meeting held in Bonn, Germany and circulated by the chairman prior to the meeting were agreed.

The agenda (PSI-02-04) circulated by the chairman immediately prior to the meeting was adopted.

Mrs Meyerson made a presentation on the new US ejection mitigation rule – FMVSS 226 (PSI-02-05). The goal of the rule is to increase occupant protection in rollovers and side crashes. It is expected that the rule will lead to larger airbags, longer inflation times and improved sensors.

Mr Hogan noted that the NHTSA had assumed no benefits in relation to rollovers and ejections when it amended FMVSS 214 and queried whether any of the benefit data for FMVSS 226 might have relevance to a Pole Side Impact GTR and whether this could be extracted.

Mr Belcher presented a consolidated summary of the crash data provided to Australia by the various contracting parties (PSI-02-06).

Mr Loew asked why fatal pole/tree side impact crashes are so common in Australia. Mr Hogan explained that road conditions are different for each country, and in the case of Australia, there is a very large road network, including in rural areas.

Ms Dausse presented a summary of initial French national and LAB pole side impact data (PSI-02-07). Further data will be provided.

Mr Belcher presented a consolidated summary of the crash data provided to Australia by the various contracting parties (PSI-02-06).

Mr Loew asked why fatal pole/tree side impact crashes are so common in Australia. Mr Hogan explained that road conditions are different for each country, and in the case of Australia, there is a very large road network, including in rural areas.

Ms Dausse presented a summary of initial French national and LAB pole side impact data (PSI-02-07). Further data will be provided.

Mr Belcher presented a draft pole side impact GTR scope (based on the current scope of the FMVSS 214 pole test and the scopes of GTRs 7, 8 and 9) for discussion (PSI-02-08). Mr Pott suggested that we should continue analysis of safety need before moving to agree a scope. Mrs Meyerson stated that the US is ready to go forward and ready to start working on the text of the GTR. Mr Abraham suggested that the scope could simply be the same as GTR 7, which would remove the need for the last sentence of the proposed scope.

Mr Hynd asked how well the WorldSID seating procedure will work for some of the more upright seats in Category 2 vehicles. Ms Tylko advised that the WorldSID can be physically positioned / installed in upright Category 2 vehicle seats, but that the validity of the dummy responses obtained from testing in such a position is not clear.

Mr Broertjes expressed some support for the wording of the proposed scope, noting it currently had everything in it, but that there was a need to be able to account for the different ways vehicles of each category are used in different countries.

Mr Versmissen presented a summary of the APROSYS car-to-pole side impact research activities (PSI-02-09), which noted that there were around 10,000 car occupant fatalities in side impact crashes in Europe annually. APROSYS had compared full-scale oblique and perpendicular pole side impact test results and had also used computer simulation to investigate the influence of impact speed, impact angle, pole diameter and impact alignment.

Mr Versmissen stated that the APROSYS group had expressed a preference for a perpendicular impact angle, mainly because of oblique rib loading concerns, but that there was no major objection to an oblique test if necessary for harmonization.

The APROSYS oblique tests had generally resulted in slightly higher numbers overall than the perpendicular pole tests. There was good repeatability for the oblique tests, but there was a need to update some test facility equipment (e.g. a need for larger carrier sleds), but this was not considered an unreasonably high cost / impost for test facilities. It was found that moving the perpendicular impact location forward 100 mm (i.e. an offset perpendicular test method) would produce similar rib deflection values as a 75 degree oblique test method. Mr Versmissen suggested that impact location was more important than impact angle.

The APROSYS group also investigated the suitability of the pole diameter and had concluded that the current 254 mm (10 inch) pole diameter was an appropriate pole diameter to be using in a vehicle-to-pole side impact test.

The APROSYS group had concluded that a 100 mm offset perpendicular vehicle-to-pole test may be better than 75 degree oblique on the basis that the ES-2 and WorldSID dummies were both considered likely to provide more accurate measurements of injury risk in perpendicular impact. However, Mr Versmissen noted that recent multi-dimensional rib deflection measurement systems for WorldSID, such as RibEye and 2D-IRTRACC would be able to be used to overcome any oblique rib loading concern the APROSYS group had at the time the APROSYS study was concluded.

Mr Terrell commented that recent oblique pole tests undertaken by Australia and Transport Canada with the WorldSID 50th percentile male dummy (WorldSID 50th) fitted with RibEye had shown predominantly lateral rib deflection responses, while perpendicular pole tests had produced considerable forward movement of the WorldSID 50th ribs.

Mr Damm asked if the APROSYS group had considered the inclusion of rear seat occupants in the test procedure. Mr Versmissen advised that a struck-side front-row 50th percentile male dummy only was considered given that a separate test would be necessary for rear seat occupants. Mrs Meyerson then commented that the FMVSS 214 pole test requirements apply to a 50th percentile male and 5th percentile female dummy positioned in the front-row only. Placement of dummies in the second row had been considered but was not justified.

Mr Ridella presented a summary of recent US oblique pole side impact crash tests using a WorldSID 50th with IRTRACC (PSI-02-10). Mr Ridella advised that NHTSA intends to make the test data available online.

Mrs Meyerson presented an updated analysis of injuries in US pole side impact crashes (PSI-02-11). Mrs Meyerson advised that the FMVSS 214 pole test benefits analysis presented at the Bonn meeting (PSI-01-16) had conservatively not included any benefits for the abdomen and pelvis in vehicle to pole/tree side impact crashes as the MAIS had always occurred at the head or thorax.

There were serious abdomen and pelvis injuries in the US pole/tree side impact statistics, but the abdomen and pelvis had not been recorded as the location of the Maximum AIS (MAIS) injury in these crashes. The US should therefore be seeing some benefits by way of reduced abdomen and pelvis injuries in pole/tree side impact crashes, although no benefits had been claimed in the FMVSS 214 pole test benefits analysis.

Mr Langner presented an update of GIDAS and CCIS in-depth pole side impact data, by angle of impact, injured body region and casualty age (PSI-02-12). Mr Langner undertook to provide further analysis based on vehicles manufactured in the last 5 years only and including serious injuries.

Mr Belcher presented an analysis of vehicle structural deformation in oblique, perpendicular and offset perpendicular pole side impact crash tests conducted by Australia (PSI-02-13). All tests had been conducted at 32 km/h. Mr Belcher noted that the perpendicular and offset perpendicular tests produced marginally more structural deformation than the oblique test at 32 km/h, but the 32 km/h oblique test was theoretically predicted to produce more structural deformation than a perpendicular or offset perpendicular test conducted at 29 km/h.

Mr Belcher also noted that the oblique impact looked very much like a perpendicular impact and, in that respect, it would be easy for an oblique impact to be mistaken for a perpendicular impact in the statistical coding of field crashes.

Mrs Meyerson presented a summary of NHTSA’s motivations for including an oblique pole test in FMVSS 214 (PSI-02-14). Mrs Meyerson advised that NHTSA had analysed NASS/CDS field crash cases where the side airbags should have deployed, but did not. NHTSA had decided on the oblique impact angle as a result of side airbag system failures / limitations observed in lab tests and from analysis of field crash data.

Mr Terrell asked if NHTSA knew what sensors were being used in the case study vehicles and commented that he did not notice pressure sensors in the vehicles on display at the EuroNCAP exhibition in Brussels he had attended the day before.

Mr Wiacek advised that he had noticed that many manufacturers of US market vehicles were now moving towards multiple pressure sensors instead of single b-pillar acceleration type sensors.

Mr Limmer advised that two sensors are generally required to fire a side airbag, including one at the airbag control module near the vehicle centre of gravity. In an oblique impact the signal at the airbag control module can sometimes be a bit noisier than in a perpendicular impact. Mr Limmer stated that some cars give really good b-pillar acceleration signals while other cars do not and noted that impacts that occur at the end of the door can be difficult to detect with pressure sensors. It is therefore a matter of designing the most appropriate sensing system for each vehicle model.

Mrs Meyerson presented an updated analysis of injuries in US pole side impact crashes (PSI-02-11). Mrs Meyerson advised that the FMVSS 214 pole test benefits analysis presented at the Bonn meeting (PSI-01-16) had conservatively not included any benefits for the abdomen and pelvis in vehicle to pole/tree side impact crashes as the MAIS had always occurred at the head or thorax.

There were serious abdomen and pelvis injuries in the US pole/tree side impact statistics, but the abdomen and pelvis had not been recorded as the location of the Maximum AIS (MAIS) injury in these crashes. The US should therefore be seeing some benefits by way of reduced abdomen and pelvis injuries in pole/tree side impact crashes, although no benefits had been claimed in the FMVSS 214 pole test benefits analysis.

Mr Langner presented an update of GIDAS and CCIS in-depth pole side impact data, by angle of impact, injured body region and casualty age (PSI-02-12). Mr Langner undertook to provide further analysis based on vehicles manufactured in the last 5 years only and including serious injuries.

Mr Belcher presented an analysis of vehicle structural deformation in oblique, perpendicular and offset perpendicular pole side impact crash tests conducted by Australia (PSI-02-13). All tests had been conducted at 32 km/h. Mr Belcher noted that the perpendicular and offset perpendicular tests produced marginally more structural deformation than the oblique test at 32 km/h, but the 32 km/h oblique test was theoretically predicted to produce more structural deformation than a perpendicular or offset perpendicular test conducted at 29 km/h.

Mr Belcher also noted that the oblique impact looked very much like a perpendicular impact and, in that respect, it would be easy for an oblique impact to be mistaken for a perpendicular impact in the statistical coding of field crashes.

Mrs Meyerson presented a summary of NHTSA’s motivations for including an oblique pole test in FMVSS 214 (PSI-02-14). Mrs Meyerson advised that NHTSA had analysed NASS/CDS field crash cases where the side airbags should have deployed, but did not. NHTSA had decided on the oblique impact angle as a result of side airbag system failures / limitations observed in lab tests and from analysis of field crash data.

Mr Terrell asked if NHTSA knew what sensors were being used in the case study vehicles and commented that he did not notice pressure sensors in the vehicles on display at the EuroNCAP exhibition in Brussels he had attended the day before.

Mr Wiacek advised that he had noticed that many manufacturers of US market vehicles were now moving towards multiple pressure sensors instead of single b-pillar acceleration type sensors.

Mr Limmer advised that two sensors are generally required to fire a side airbag, including one at the airbag control module near the vehicle centre of gravity. In an oblique impact the signal at the airbag control module can sometimes be a bit noisier than in a perpendicular impact. Mr Limmer stated that some cars give really good b-pillar acceleration signals while other cars do not and noted that impacts that occur at the end of the door can be difficult to detect with pressure sensors. It is therefore a matter of designing the most appropriate sensing system for each vehicle model.

Mrs Meyerson presented an updated analysis of injuries in US pole side impact crashes (PSI-02-11). Mrs Meyerson advised that the FMVSS 214 pole test benefits analysis presented at the Bonn meeting (PSI-01-16) had conservatively not included any benefits for the abdomen and pelvis in vehicle to pole/tree side impact crashes as the MAIS had always occurred at the head or thorax.

There were serious abdomen and pelvis injuries in the US pole/tree side impact statistics, but the abdomen and pelvis had not been recorded as the location of the Maximum AIS (MAIS) injury in these crashes. The US should therefore be seeing some benefits by way of reduced abdomen and pelvis injuries in pole/tree side impact crashes, although no benefits had been claimed in the FMVSS 214 pole test benefits analysis.

Mr Langner presented an update of GIDAS and CCIS in-depth pole side impact data, by angle of impact, injured body region and casualty age (PSI-02-12). Mr Langner undertook to provide further analysis based on vehicles manufactured in the last 5 years only and including serious injuries.

Mr Belcher presented an analysis of vehicle structural deformation in oblique, perpendicular and offset perpendicular pole side impact crash tests conducted by Australia (PSI-02-13). All tests had been conducted at 32 km/h. Mr Belcher noted that the perpendicular and offset perpendicular tests produced marginally more structural deformation than the oblique test at 32 km/h, but the 32 km/h oblique test was theoretically predicted to produce more structural deformation than a perpendicular or offset perpendicular test conducted at 29 km/h.

Mr Belcher also noted that the oblique impact looked very much like a perpendicular impact and, in that respect, it would be easy for an oblique impact to be mistaken for a perpendicular impact in the statistical coding of field crashes.

Mrs Meyerson presented a summary of NHTSA’s motivations for including an oblique pole test in FMVSS 214 (PSI-02-14). Mrs Meyerson advised that NHTSA had analysed NASS/CDS field crash cases where the side airbags should have deployed, but did not. NHTSA had decided on the oblique impact angle as a result of side airbag system failures / limitations observed in lab tests and from analysis of field crash data.

Mr Terrell asked if NHTSA knew what sensors were being used in the case study vehicles and commented that he did not notice pressure sensors in the vehicles on display at the EuroNCAP exhibition in Brussels he had attended the day before.

Mr Wiacek advised that he had noticed that many manufacturers of US market vehicles were now moving towards multiple pressure sensors instead of single b-pillar acceleration type sensors.

Mr Limmer advised that two sensors are generally required to fire a side airbag, including one at the airbag control module near the vehicle centre of gravity. In an oblique impact the signal at the airbag control module can sometimes be a bit noisier than in a perpendicular impact. Mr Limmer stated that some cars give really good b-pillar acceleration signals while other cars do not and noted that impacts that occur at the end of the door can be difficult to detect with pressure sensors. It is therefore a matter of designing the most appropriate sensing system for each vehicle model.

Mrs Meyerson presented an updated analysis of injuries in US pole side impact crashes (PSI-02-11). Mrs Meyerson advised that the FMVSS 214 pole test benefits analysis presented at the Bonn meeting (PSI-01-16) had conservatively not included any benefits for the abdomen and pelvis in vehicle to pole/tree side impact crashes as the MAIS had always occurred at the head or thorax.

There were serious abdomen and pelvis injuries in the US pole/tree side impact statistics, but the abdomen and pelvis had not been recorded as the location of the Maximum AIS (MAIS) injury in these crashes. The US should therefore be seeing some benefits by way of reduced abdomen and pelvis injuries in pole/tree side impact crashes, although no benefits had been claimed in the FMVSS 214 pole test benefits analysis.

Mr Langner presented an update of GIDAS and CCIS in-depth pole side impact data, by angle of impact, injured body region and casualty age (PSI-02-12). Mr Langner undertook to provide further analysis based on vehicles manufactured in the last 5 years only and including serious injuries.

Mr Belcher presented an analysis of vehicle structural deformation in oblique, perpendicular and offset perpendicular pole side impact crash tests conducted by Australia (PSI-02-13). All tests had been conducted at 32 km/h. Mr Belcher noted that the perpendicular and offset perpendicular tests produced marginally more structural deformation than the oblique test at 32 km/h, but the 32 km/h oblique test was theoretically predicted to produce more structural deformation than a perpendicular or offset perpendicular test conducted at 29 km/h.

Mr Belcher also noted that the oblique impact looked very much like a perpendicular impact and, in that respect, it would be easy for an oblique impact to be mistaken for a perpendicular impact in the statistical coding of field crashes.

Mrs Meyerson presented a summary of NHTSA’s motivations for including an oblique pole test in FMVSS 214 (PSI-02-14). Mrs Meyerson advised that NHTSA had analysed NASS/CDS field crash cases where the side airbags should have deployed, but did not. NHTSA had decided on the oblique impact angle as a result of side airbag system failures / limitations observed in lab tests and from analysis of field crash data.

Mr Terrell asked if NHTSA knew what sensors were being used in the case study vehicles and commented that he did not notice pressure sensors in the vehicles on display at the EuroNCAP exhibition in Brussels he had attended the day before.

Mr Wiacek advised that he had noticed that many manufacturers of US market vehicles were now moving towards multiple pressure sensors instead of single b-pillar acceleration type sensors.

Mr Limmer advised that two sensors are generally required to fire a side airbag, including one at the airbag control module near the vehicle centre of gravity. In an oblique impact the signal at the airbag control module can sometimes be a bit noisier than in a perpendicular impact. Mr Limmer stated that some cars give really good b-pillar acceleration signals while other cars do not and noted that impacts that occur at the end of the door can be difficult to detect with pressure sensors. It is therefore a matter of designing the most appropriate sensing system for each vehicle model.

Mr Petit provided an update on the progress on the development of the ISO/WG6 WorldSID 50th injury risk curves (PSI-02-15). Mr Petit advised that the ISO/WG6 had agreed to use the survival analysis statistical method to construct the injury risk curves. To enable injury criteria limits to be finalised the ISO/WG6 will now need to establish the most suitable injury risk predictors for each body region (e.g. pelvis injury risk could be predicted by 3ms pelvis acceleration and/or pubic force) and the recommended thresholds. The same process would need to be followed for the WorldSID 5th. WG6 would meet again in May.

Mr Belcher presented data (2000-2009) on fatalities and serious injuries in Victorian (Australian state) side impact crashes by occupant age (PSI-02-16). It was noted that the 45 year age, which is currently used for the age scaling of current side impact dummy injury risk curves, exceeds (by age) around 85% of Victorian pole/tree side impact crash fatalities, but only around 50% of other side impact fatalities.

Mr Ridella presented a summary of an UMTRI investigation of the effects of occupant age on AIS 3+ injury outcomes (PSI-02-17). Mr Ridella undertook to investigate with UMTRI the possibility of separating pole side impacts from other side impacts in the data presented in this study. In response to a question from Ms Tylko, Mr Ridella indicated that while it would decrease the number of available cases, it was possible to identify contact point in the NASS data base.

Mr Petit provided an update on the progress on the development of the ISO/WG6 WorldSID 50th injury risk curves (PSI-02-15). Mr Petit advised that the ISO/WG6 had agreed to use the survival analysis statistical method to construct the injury risk curves. To enable injury criteria limits to be finalised the ISO/WG6 will now need to establish the most suitable injury risk predictors for each body region (e.g. pelvis injury risk could be predicted by 3ms pelvis acceleration and/or pubic force) and the recommended thresholds. The same process would need to be followed for the WorldSID 5th. WG6 would meet again in May.

Mr Belcher presented data (2000-2009) on fatalities and serious injuries in Victorian (Australian state) side impact crashes by occupant age (PSI-02-16). It was noted that the 45 year age, which is currently used for the age scaling of current side impact dummy injury risk curves, exceeds (by age) around 85% of Victorian pole/tree side impact crash fatalities, but only around 50% of other side impact fatalities.

Mr Ridella presented a summary of an UMTRI investigation of the effects of occupant age on AIS 3+ injury outcomes (PSI-02-17). Mr Ridella undertook to investigate with UMTRI the possibility of separating pole side impacts from other side impacts in the data presented in this study. In response to a question from Ms Tylko, Mr Ridella indicated that while it would decrease the number of available cases, it was possible to identify contact point in the NASS data base.

Mr Petit provided an update on the progress on the development of the ISO/WG6 WorldSID 50th injury risk curves (PSI-02-15). Mr Petit advised that the ISO/WG6 had agreed to use the survival analysis statistical method to construct the injury risk curves. To enable injury criteria limits to be finalised the ISO/WG6 will now need to establish the most suitable injury risk predictors for each body region (e.g. pelvis injury risk could be predicted by 3ms pelvis acceleration and/or pubic force) and the recommended thresholds. The same process would need to be followed for the WorldSID 5th. WG6 would meet again in May.

Mr Belcher presented data (2000-2009) on fatalities and serious injuries in Victorian (Australian state) side impact crashes by occupant age (PSI-02-16). It was noted that the 45 year age, which is currently used for the age scaling of current side impact dummy injury risk curves, exceeds (by age) around 85% of Victorian pole/tree side impact crash fatalities, but only around 50% of other side impact fatalities.

Mr Ridella presented a summary of an UMTRI investigation of the effects of occupant age on AIS 3+ injury outcomes (PSI-02-17). Mr Ridella undertook to investigate with UMTRI the possibility of separating pole side impacts from other side impacts in the data presented in this study. In response to a question from Ms Tylko, Mr Ridella indicated that while it would decrease the number of available cases, it was possible to identify contact point in the NASS data base.

Mr Slaba presented an analysis of GIDAS data and internal BMW oblique and perpendicular vehicle-to-pole side impact crash testing (PSI-02-18). This BMW GIDAS analysis was intentionally restricted to vehicles not registered before 2000 with damage to the occupant compartment.

Mr Slaba noted that vehicles fitted with ESC (as will be required by regulation) showed an increased tendency towards oblique impact as ESC acts to bring the vehicle more on track.

Mr Slaba also noted that the ES-2re dummy readings from internal BMW 32 km/h oblique tests were generally more severe than the ES-2 readings from 29 km/h perpendicular tests and that the oblique test had proven practicable to conduct.

Mr Terrell presented an Australian side impact field data research study proposal for comments / feedback from members of the informal group (PSI-02-19).

Mrs Meyerson recommended that the study should include an analysis of detailed case studies as these can provide a more accurate understanding of injuries than more broad statistical databases. Mrs Meyerson then suggested that NASS case studies could be considered.

Ms Tylko suggested that the study should focus on crashes where there was side impact damage to the occupant compartment.

Mr Francis suggested that the effects of occupant age should be included in the study.

Mr Hogan advised that Australia would also complete a further round of pole side impact crash tests in collaboration with Transport Canada prior to the next meeting of the informal group and invited any suggestions from the informal group regarding the research focus of these tests.