Transcription
Information on the implimentation of RCAR crash standards in theGerman insurance vehicle rating system and information on AEB systemsAs an RCAR member in Germany, AZT ist often contacted for information on German type classrating and the implementation of RCAR tests in this rating procedure.Herewith AZT wants to provide this information in a concentrated form at a central access point.BackgroundThe German type class rating is used to evaluate the insurer’s claims costs in order to estimatethe specific risk of a particular passenger vehicle in the German market. It is not a specification ofpremiums. However, most insurers rely on the type class as a basis of their individual calculationof premiums.The process is prepared by the German Insurers Association (GDV) and finalized by a commission under the lead of an independent trustee. The preparation includes a technical evaluation ofthe vehicle’s repair friendliness and its ability to interact with an opponent. GDV has publishedthe respective information for the procedure, the implementation of the bumper test and additional information on lower load pathes over the last years. An additional paper describes the consideration of AEB systems at the initial classification procedure.With courtesy of GDV the named documents are provided in the following.1.2.3.4.5.Description of Group Rating System, Issue April 2011Role of the RCAR Bumpertest, Issue Nov 09Documentation scheme of Bumper Test resultsStatement „Unterer Lastpfad“ (Lower Loadpath), Issue Januar 2011Consideration of AEB systems at the initial classification procedureContact:AZT Automotive GmbHAllianz Zentrum fuer TechnikGesamtverband der DeutschenVersicherungswirtschaft e. V.Kraftfahrtversicherung, Kfz-Technik und StatistikCarsten ReinkemeyerMuenchener Strasse 8985737 IsmaningTel 49.89.3800.6393Fax ning, 2013-8-29Dr. Jürgen RedlichLeiter Kfz-TechnikWilhelmstraße 43 / 43 G10117 BerlinFon 49 30 2020 5321Mobile 49 172 399 5835Fax 49 30 2020 6611
Information on the implimentation of RCAR crash standards in theGerman insurance vehicle rating system and information on AEB systemsDocument 1Description of Group Rating Systemissue April 2011 by GDV
Rating system of passenger carsin German motor insurance(Full and partial comprehensive coverage,third party liability)(Status: 04/11)1.Type class lists, trustee and classification committeeThe type class lists for third party liability, full and partial car comprehensiveinsurance are kept by an independent trustee, PricewaterhouseCoopers AGWirtschaftsprüfungsgesellschaft. The trustee is advised by a classification committeeto which delegates of the German federal office for motor transport (KBA), theGerman insurance association (GDV) as well as technical experts and consultantsfrom the insurance industry belong. Suggestions for the classification of a new orreclassification of an existing auto model are put forward by GDV (i.e. the Germaninsurance association), submitted to the trustee for examination and subsequentlydiscussed with the classification committee. On the basis of this consultation processthe trustee implements these classifications and/or reclassifications of car models.The type class lists are drawn up by the trustee on the 1st of October of every year.The German insurance association (GDV) sends a circular containing the type classlists to the insurance companies and at the same time places the updated typeclasses in the Internet (www.gdv.de). In the preceding annual examination the modelcategories of all vehicles are checked on the basis of the newest GDV statistics tosee whether the current classifications based on the cost of claims are still warrantedor whether reclassifications ought to take place as a result of changes in the claimscosts.In full and partial comprehensive coverage the classifications of new vehicle models,which take place several times every year, are also immediately passed on to theinsurance companies by the GDV in the form of supplements to the latest type classlists. This information is also placed in the Internet.
-2-2.Type classification of the full and partial comprehensive coverageThe existing structure in motor insurance was initially recommended to insurancecompanies on 18.9.2002 in the form of studies. Among other things the studiesdescribe the procedures for establishing the risk-premium index of each vehicle type,for creating type classes as well as for the initial classification of new vehicles types.2.1 Type classificationThe model categories or classes as well as their index limits were established bystatistical means (iterative minimum variance analysis) in both the new full and newpartial comprehensive coverage. This resulted in 25 model categories in fullcomprehensive coverage (Appendix 1) and 24 model categories in partialcomprehensive coverage (Appendix 2). The index limits as well as the range of modelcategories are different in full and partial coverage.The classification into the individual model categories is based on the risk premium(RP) of the relevant vehicle type. This RP value is expressed in index points. A rangeof index points is allocated to each model category. If the upper or lower limits of thisrange is transgressed a reclassification of the model category takes place.2.2 Formation of design groupsAll models which are categorized by the European Type approval procedure(2007/46/EG) as M1 or M1G (passenger car / off-road passenger car) are describedby a combination of manufacturer and model identification numbers (HSN/TSN)issued by the federal office for motor transport (KBA).Among the numerous vehicle types there are many models similar in design, but theymay differ in their performance figures. In full and partial comprehensive coveragethese vehicle types can be combined into various design groups. The composition ofthe design groups in full and partial comprehensive may be different.
-3-2.3 Classification procedureIf a new vehicle type can be allocated to an existing design group, e.g. a variant of anexisting model series, the variant assumes the risk premium index of this designgroup. For vehicle types which cannot be allocated to a design group and for vehicletypes of a completely new model series, by contrast, one or more new design groupsare established. The index values for claims frequency, claims average or riskpremium will be ascertained anew.2.3.1 Full comprehensive coverageIn full comprehensive coverage the classification of new models or vehicle typeswhich cannot be allocated to a design group takes place on the basis of the riskpremium index (RPIndex) which is derived at by multiplying the claims frequency index(CFIndex) by the claims average index (CAIndex), divided by 100.The index CA is derived from the ratio of the claims average of the model to beclassified (CAIndividual) to the claims average of all market models (CATotal), multipliedby 100. The index CF is derived from the ratio of the claims frequency of theindividual model (CFIndividual) to the claims frequency of all market models (CFTotal),again multiplied by 100.Thus the basic formula for the calculation is as follows:RPIndex IndexCAIndividual 100 IndexCATotalCFIndividual 100100CFTotalThrough the reference to the average of all market models the difference between thecalculated index figure for claims average and claims frequency and the average of100 shows by what percentage the model being classified is above or under theoverall average of all models in terms of claims average and claims frequency.The CF value is constituted by the relevant index value of the previous model and/orthe CF medium value of relevant comparable models from other manufacturers fromthe GDV model statistics. The lower claims frequency of new vehicles, which hasbeen statistically proven, is taken into account through a percentage reduction,calculated annually, of the claims frequency of the model to be classified.
-4-Since no CA value is available for new models, it is determined in accordance withthe so-called model damage calculation (TSB). To be able to also take into accountthe influence of the driving power (P) on the average repair costs within a particularmodel series, the claims index CATSB of the model series will be extended for eachvehicle type by one component as a function of the driving power (CAkW).Index CA Index CATSB Index CAkWwhere Index CAkW 0,1 * P (kW) 5,7This consists of the weighted repair costs for typified damage to the front, side andrear which, with a share of 54 % for front damage, 30 % for rear damage and 16 % forside damage, forms the basis of the initial classification in the model damage method.The claims settlement costs incurred as a result of the actual loss adjustment aretaken into account by means of an annually adjusted percentage surcharge on theclaim average CATSB of the model to be classified.In this calculation method not only the price of spare parts, but also the working timesare important for repairing the damage. The shorter the repair times and the lessmaterial (spare parts and paints) necessary, i.e. the more repair-friendly the design ofthe model, the lower the model damage calculation, the claims average index of themodel and thus also the initial model category classification. This method ofcalculation allows the manufacturer to achieve a favourable classification for fullcomprehensive coverage by designing the vehicle as repair-friendly as possible.A crash repair test (RCAR 15 km/h 10 1) to be carried out by the manufacturer shouldform the basis of the front and rear damage (Appendices 3 and 4). If there is no crashrepair test available, the calculation is made on theoretical basis. For the sidedamage a theoretical damage is assumed whose essential components are asectional repair of the driver’s door sill as well as the renewal of the driver's door.Since January 1st 2010, the results of the RCAR Bumper Test Procedure (Appendix5) are considered within the classification formula in order to evaluate theperformance of the bumper system closer to reality. The front and rear bumper testsare based on the RCAR bumper test paper2 and supplementary the comments of1The Procedure for Conducting a Low Speed 15 km/h Offset Insurance Crash test to Determine the Damageability andRepairability Features of Motor Vehicle, Issue 2.1 :September 2006(http://www.rcar.org/Papers/Procedures/rcar test protocol angled barrier.pdf)2RCAR Bumper Test;Issue 2.00 September 2010 including Appendix 1 „Dimensions and Specifications of the RCAR BumperBarrier System“ rocedure.pdf)
-5-GDV3. If the car is complying with the requirements of the bumper test, the result ofthe type damage calculation, based on the RCAR 10 structural test, remainsunchanged. If not, the result of the type damage calculation will be increased by onetype class each for the front and rear test independently. For this reason, the indexrisk premium (RP) and claims average (CA) will be recalculated in order to fit with thenew resulting type class. Therefore the maximum influence of the bumper test islimited by 2 type classes.In the case of special high performance variants of a model series or variants fittedwith high price parts, the CA value is ascertained according to the so-called highprice/high-performance method. A surcharge allowing for the increased risk premiumof such models is placed on the CA value of the basic model commensurate with theratio of the sales price of the high-price/high-performance model to be classified tothe sales price of the basic model.The model category calculation is commissioned by the trustee and carried outseveral months prior to the new model coming onto the market by representatives ofthe German insurance association (GDV) in close collaboration with the vehiclemanufacturers or importers.2.3.2 Partial comprehensive coverageSince no comparable model claims calculation is possible, the initial classification inpartial comprehensive coverage insurance is based on a similar previous model's riskpremium index borne out by statistics. If no comparable previous model withstatistical data exists, the risk premium index figures of one or more comparablecompetitor models of the same or a different manufacturer are normally used - eitheralone or in addition - to establish a medium value.3Integration of the RCAR Bumper Tests into the initial rating procedure used by the German motor insurance industry for fullycomprehensive vehicle insurance
-6-2.4 Reclassification procedureClassifications are reviewed annually based on the GDV model statistics of theprevious year. To widen the statistical basis and to be able to counter coincidentalinfluences, the review includes the last three statistical years. The result is publishedin an up-dated model category index on 1 October each year.2.4.1 Full comprehensive coverageReclassification in full comprehensive coverage takes place if the current statisticalvalues of the design group's CF Index and CA Index have increased or decreased inrelation to the initial or the previous year's classification and the RP Index of thedesign group has consequently exceeded or dropped below the points limit for thenext model category. The single model's CA Index, which is derived at through themodel damage calculation method, is examined only if the relevant design group hasa minimum of 7300 annual units4. At least 2800 annual units are necessary for theexamination of the CF Index.If there is no statistical certainty for the CF or CA index values due to the fact that2800 and 7300 annual units respectively have not yet been reached or have againdropped below the limit, a credibility procedure will be used to establish the claimsfrequency index value CF or the claims average index value CA.In this credibility procedure the index value CF or CA is a weighted average which iscalculated from the index value of the initial classification on the one hand and thestatistical index value on the other in accordance with the formula below (for the CFIndex).AUCFcred CFstat AUtotalKey: CFcredCFstatCFicAUAUtotalAU CFic 1AUtotal: credibility claims frequency index: statistical claims frequency index: claims frequency index of initial classification: number of annual units of vehicle type: statistically ascertained limit of annual units of respective index value
-7-2.4.2 Partial comprehensive coverageThe risk premium values of initial classifications in partial comprehensive coverageare examined if a minimum of 8100 annual units exist. Here too a credibilityprocedure is used if the number of units has not yet reached the annual limit or hasfallen below it again (cf. 2.4.1).3.The model category classification of Third Party Liability insurance3.1 Type classificationSince July 1996, on the basis of the non-binding risk premium rate of the GDV, thethird party liability premium of a vehicle type is no longer determined according to 11kW (Kilowatt) categories etc., but, similar to auto insurance, according to the riskpremium index of the relevant vehicle type. The 16 model categories of third partyliability insurance as well as their index limits (Appendix 6) were determined throughstatistical methods (iterative minimum variance analysis).3.2 Establishment of design groupsIn third party liability insurance too, similarly constructed vehicle models can becombined in design groups (refer also to item 2.2). The combination of design groupsin third party liability insurance can differ from those in full or in partial comprehensive.3.3 Initial classification procedureIf a new vehicle type can be allocated to an existing design group, it adopts the riskpremium index of this design group. If, on the other hand, a new model or a newvehicle type forms its own design group, risk premium index values borne out bystatistics of previous and/or comparable models will be adopted.3.4 Reclassification procedureThe allocation of design groups to the individual model categories is reviewedannually, based on the updated GDV model statistics. If any changes appear in the4An annual unit (AU) is equivalent to a vehicle insured for the period of one year. (1st of January until31st of December).
-8-risk premium index so that the points limit to the next highest or lowest modelcategory is transgressed, appropriate reclassifications will be made. To widen thestatistical basis and to be able to counter random factors, the last three statisticalyears are used as a basis.In third party liability insurance the risk premium index values can be assumed tohave a sound statistical basis if 11000 annual units have been reached in a designgroup. If no sound statistical basis for the risk premium index values exists becauseof annual units having not yet reached 11000 or having again dropped below thatnumber, a credibility method will be used to determine the risk premium index value(cf. 2.4.1).The classification results are published in a model category index that is updatedannually by 1 October.Hamburg, 13.04.2011Trustee and classification committee
-9-Appendix 1: Index points of the type classes for the new full coverageType classType class limits1011121314up 2276,3330,03031323334330,1377,5438,7516,6more than377,4438,6516,5696,6696,7
- 10 -Appendix 2: Index points of the type classes for the new partial coverageType classType class limits1011121314up 4,8416,4486,930313233487,0628,8763,9more than628,7763,8975,4975,5
- 11 -Appendix 3: Front crash test for the calculation of the type classes(AZT Automotive GmbH, Allianz Zentrum für Technik)cTest arrangementReferent widthOverlapVehicleConditionLoadMeasurementsThe front of the vehicle is driven against a rigid angular barrierwith an overlap of 40% on the drive s side.Test houses may decide to test the other side of the car, ifappropriateB Vehicle width without mirrorsÜ 0,40 x BF VehicleFree of any exterior or propelling force at the moment of impactServiceable, battery connected, ignition „on“.Saftey equipment (belt pretensioner, airbags) in function.Air conditioning system drained and pressure checked. Brakesreleased, gear lever in neutral position1 Dummy, 50% male, driving position, secured with seat beltFuel tank filled with petrol or diesel, water also permissibleAxle alignment and body gaps checked before and after impact.Body alignment checked before and after impact. Actual vehicle smass in test configuration.Vehicle acceleration on left and right sill (base of the “B” post)Barrier dimensionsBarrier AngleHeightDepthRoundingα 10 (also permissible at short barrier surface -c-)Barrier clearly higher than the vehicle frontNo contact of the vehicle front with the wall adjacent to the barrierR 150 mmImpact speedVF 15,0 km/h ( 1/-0 km/h )Remark:The front door on the impacted side shall be opened completelyafter the crash test.
- 12 -Appendix 4: Rear crash test for the calculation of type classes(AZT Automotive GmbH, Allianz Zentrum für Technik)Test arrangementThe rear is impacted by a mobile barrier on the driver s side withan overlap of 40%.Test houses may decide to test the other side of the car, ifappropriateReferent widthOverlapB Vehicle width without mirrorsÜ 0,40 x BConditionF VehicleFree of any exterior or propelling force at the moment of impactServiceable, battery connected, ignition „on“.Saftey equipment (belt pretensioner, airbags) in function.Stationary, brakes released, gear lever in neutral positionVehicleLoad1 Dummy, 50% male, driving position, secured with seat belt.Fuel tank filled with petrol or diesel, water also permissibleMeasurementsAxle alignment and body gaps checked before and after impact.Body alignment checked before and after impact Actual vehicle smass in test configurationVehicle acceleration on left and right sill (base of the “B” post)Mobile barrierStW Mobile barrierFree of any exterior or propelling force at the moment of impactmST 1.400 kg, center of gravity “S” in the middle planeWheel baseY 1,5 mWidthA 1,2 mMassSled dimensionsBarrier dimensionsHeightLower edgeRoundingImpact speedH h R r 700 mm200 mm150 mm50 mmVStW 15,0 km/h ( 1/-0 km/h )
- 13 -Anlage 5:RCAR BumpertestAZT (RCAR) Bumper Test Front/ RearMain test parameter:Test arrangement:Vehicle impact on a rigid bumper barrier with energy absorbingelementCenterline of the car aligned with centerline of the barrier ( /- 50mm)Vehicle ride height:at construction level as defined by the manufacturer ( /- 10 mm)(Test houses may decide to test the vehicle „as delivered“)Condition and load:Tire pressure: standard, vehicle serviceable, battery connectedTank filled with fuel or water ( 90%)1 Dummy 75 5 kg (e.g. 50% male) in driver position, securedwith seat beltBarrier ground clearance:Front crash H 455 mm ( /- 3 mm)Rear crash H 405 or 455 mm ( /- 3 mm) (depending onlocal market)Impact speed:V 10.0 km/h ( /-0.5 km/h )
- 14 -Appendix 6: Index points of the type classes for the third party liabilityType classType class limits1011121314up 5,925more than196,0
Information on the implimentation of RCAR crash standards in theGerman insurance vehicle rating system and information on AEB systemsDocument 2Role of the RCAR BumpertestIssue Nov 09, by GDV
Integration of theRCAR bumper testsinto the initial rating procedureused by the German motor insurance industryfor fully comprehensive vehicle insuranceIssue November 20090.IntroductionThe classification commission of the German motor insurers sees the RCAR bumper test as asuitable method for assessing how well a vehicles bumper system protects the vehicle fromdamage in low-speed impacts in real-world crashes. The goal is to encourage vehiclemanufacturers to further improve the level of self-protection of passenger cars and to increasethe vehicles' geometric compatibility in collisions.The RCAR bumper test is a necessary complement to the current RCAR 10 structural testnecessary to avoid undesirable developments in the bumper system, such as too small or tooweak bumper beam or even no bumper beam at all.The amended rating procedure is based on the RCAR paper "RCAR Bumper Test" ofNovember 2008, Issue 1.02 and Appendix 1 "Dimensions and Specifications of the RCARBumper Barrier System" of September 2007, Issue 1, with the following additions andexplanations.1.ApplicationThe RCAR bumper test will be used in addition to the RCAR 10 structural test in determiningthe initial classification of new cars into the comprehensive insurance type classification systemas of 01 January 2010. Vehicles, which will be rated in the November 2009 meeting (last ratingmeeting without RCAR bumper test), will have to be launched on the market by 31 March 2010.This means that these vehicles must be available for delivery to the end user and ready forregistration and use by the end user.During the transition period, variants of model series (e.g. body variants or engine variants) thatwere rated before 01 January 2010 without the bumper test will likewise be rated without thebumper test.The RCAR bumper test will be used for all vehicles, which are homologated as passenger carsin the European type approval procedure (M1 or M1G) and listed in the type classification list.GDV Papier RCAR Bumpertest Engl Stand Nov 09.doc1/3
2.RequirementsIn order to pass the bumper test, a vehicle must fulfill the requirements in 2.1 through 2.3 belowfor both the front and the rear bumper test:2.1 Geometric requirementsa) The relevant bumper engagement of the bumper beam with the barrier must be 75 mm(see section 5.7, fig. 5, and sections 5.3 and 5.4 of the RCAR paper). For the Germanmarket, the relevant position of the barrier is that described in section 6.1 of the RCARpaper, with a barrier ground clearance of 455 mm for frontal impacts, and 405 mm forrear impacts.Alternatively:b) In cases where the relevant bumper engagement is 75 mm, but a positive test result isexpected, the qualifying bumper beam height shall be 100 mm (see section 5.1 of theRCAR paper).2.2 Dynamic testWhen the vehicle collides with the barrier, the vehicle/bumper beam must not underride oroverride the barrier, with the barrier subsequently intruding into the vehicle above or belowthe bumper beam. A purely vertical movement of the test vehicle along the barrier withoutsubsequent forward movement is acceptable.2.3 Damages to vehicle following the testa) Front impact test:- No damage to the vehicle structure: The vehicle structure includes, in particular, themain chassis members as well as any parts or panels welded to the vehicle (e.g.radiator grill/front grill, crash boxes). Any change in the position of the main chassismembers by more than /- 3 mm (tolerance, measurement before and after the test)is considered damage. Components/panels that, for assembly-related reasons, arewelded on at a few welding points but are installed with screws/bolts in case of arepair, are not considered welded parts.- No damage to cooling package, consisting of intercooler, A/C and radiator. Damageto fastening elements of the cooling packages or individual elements thereof isacceptable if it can be repaired using screw-on replacement fasteners.b) Rear impact test:- No damage to the vehicle structure: The vehicle structure includes, in particular, themain chassis members as well as any parts or panels welded to the vehicle (e.g. rearend panel, floor panel, etc.).- No damage to the boot lid or tailgate caused directly by the barrier. Indirect damage,caused for instance by wrinkling of the bumper cover, is tolerated.GDV Papier RCAR Bumpertest Engl Stand Nov 09.doc2/3
3.Effect on the initial rating of vehicles for fully comprehensive insuranceThe RCAR bumper test is introduced only in the initial rating procedure for fully comprehensiveinsurance, as a complement to the RCAR 10 structural test. If the vehicle fulfills the bumpertest requirements set forth in section 2 above, the result of the type damage calculation from theRCAR 10 structural test will be used without any changes. If the test requirements are notfulfilled, the result of the type damage calculation from the RCAR 10 structural test is increasedby one type class each for the front impact and the rear impact tests. Therefore, the impact ofthe bumper test is limited to a maximum of 2 type classes.4.Test vehicleThe test vehicle must be the variant with the largest projected sales volume of the model seriesto be rated. The ride height of the test vehicles differs from the nominal ride height of the vehicleby maximum /- 10 mm, as specified by the manufacturer (see section 4.0 of the RCAR paper).If different suspension setups are available, the standard suspension setup is to be used. Thetest vehicle must be fitted with air-conditioning. For vehicles with adjustable body height (e.g. incase of air suspension), the vehicle must be set to normal position at city speed (50 km/h).Where, in case of variants with charged engines, an intercooler must be taken into account,these variants shall be assessed separately based on the remaining clearances between thebumper beam and the AC condenser. Body variants within a model series, which have adifferent nominal ride height (i. e. Coupes) will be assessed separate also. The vehiclemanufacturer may perform simulations or additional tests to be on the safe side.5.Presentation of the test resultsThe GDV will inspect the test results either directly on the crash vehicle or through detailedphoto/video documentation of the test. In this case, the test vehicle can be repaired and reusedfor the RCAR 10 structural test. Specifically, the following points must be documented in anadequate manner:-Design height (nominal ride height) of the test vehicle under test conditions as specified bythe manufacturer.-Actual ride height of the test vehicle under test conditions (photo of the measurement)-Qualifying bumper beam height (see sections 5.2 through 5.4 of the RCAR paper)-Relevant bumper engagement with the barrier (see section 5.7 of the RCAR paper; photoshowing the test vehicle with removed bumper cover face-to-face with the barrier, includingphotographic documentation of the measurement)-Video footage of the bumper test in side view (y axis) and top view (z axis) allowing toassess override/underride.-Logs detailing the measurements of the vehicle structure (underbody) before and after thecrash test (e.g. crash boxes left and right, or other suitable points, as well as at least twoadditional points along the x axis of the main chassis members). If the same vehicle is usedfor the 10 structural test, the post-measurements of the bumper test can be used as premeasurements for the 10 structural test.-Gap measurements before and after the crash.-Documentation of damaged/undamaged vehicle parts (see section 2.3) before and after thecrash with meaningful photos.-----------------------------GDV Papier RCAR Bumpertest Engl Stand Nov 09.doc3/3
Information on the implimentation of RCAR crash standards in theGerman insurance vehicle rating system and information on AEB systemsDocument 3Documentation scheme of Bumper Test resultsIssue Nov 09, by GDV
Documentation of RCAR Bumper Test ResultReferences: - RCAR Bumper Test Procedure Issue 1.02 - November 2008- Appendix 1: Dimensions and Specifications of the RCAR Bumper BarrierSystem
German insurance vehicle rating system and information on AEB sys-tems . As an RCAR member in Germany, AZT ist often contacted for information on German type class rating and the implementation of RCAR tests in this rating procedure. Herewith AZT wants to provide this information in a concentrated form at a central access point. Background
