by Brian G. McHenry, Jan 1997
June 1997 update Jan
1998 update Feb 1998 update
December 1998 update
Terry Day and Wes Grimes set up a review process for the Accident Reconstruction (AR) sessions at SAE. They should be commended for their efforts at adding a review process. A review process is truly needed and a valuable part of preparing a technical paper. However, the process as currently implemented by Messrs. Day and Grimes contains serious flaws which need to be addressed and corrected. Among the problems we encountered which need immediate correction are:
As part of the 'review' process for the AR session two reviewers are assigned by some unknown and arbitrary process to each paper. The 'chosen' reviewers are supposed to be assigned based on familiarity with the subject matter. We were asked to review two papers. Neither paper we reviewed was up to reasonable standards for a scientific publication. One of the papers was an outline rather than a 'draft paper'. An outline is insufficient to gage a paper's scientific integrity, however qualified the paper may eventually be in it's completed draft form.
The review process is meant to be an approximately two month long procedure to provide constructive comments that will assist the author(s) in final clarification of the paper. At the completion of the 'review process' the reviewers are allowed to write a "one to two paragraph review to accompany the papers" (letter from Terry Day to authors on 12/13/96). Our 4 papers were all completed on schedule, which included a draft by mid-September 1996. On the basis of discussions with other 'reviewers' some of the papers by other authors for the session were never even delivered in draft form for review until December 1996!
We dealt with 8 reviewers as a part of our preparation of 4 papers for the 1997 session and we thank 7 of the 8 reviewers for taking the time and effort to review and provide constructive comments on our papers.
One of the 8 reviewers, Mr. James Neptune, made the review process an unreasonable and offensive ordeal.
With regard to Mr. Neptune's lengthy, misguided, and erroneous Reviewer's Discussion attached to our paper SAE 970960, the following three questions are appropriate:
Equation (16) from SAE 97-0960 (page 321) is as follows: .
Obviously, from elementary mathematics,
Note that the "impossible" calculation results cited by Mr. Neptune are the direct result of his incomprehensible rejection of equation (16) for the case of = 0.
Mr. Neptune also states on page 325, paragraph 5 that "The restitution model needs to be reconfigured such that restitution is less than, or equal to, the theoretical limit of 1.0 at zero dynamic crush and decreases from that point as the dynamic crush increases" which is clearly already an integral part of the model as defined in SAE 970960 (i.e., in the form of equation (16)).
There is no rule of mathematics applied to support his comments. Mr. Neptune made a fundamental mathematical error in his Reviewer's Discussion.
An extensive file of related correspondence exists. The Session Organizers and SAE were all provided the complete file. They were also notified several times of the errors and misstatements that Mr. Neptune made in his related correspondence and Reviewer's Discussion.
Session Co-Organizer Grimes is a co-author with Neptune on another paper on damage analysis published at the 1997 SAE Congress (SAE paper 97-0942), Mr Grimes also published a paper using the ATB in the HVE (SAE 97-0967) and Mr. Neptune published a paper last year on the HVE vehicle database (SAE paper 96-0896) (note that HVE is Mr. Day's SGI graphics interface package for SMAC and CRASH). Each of these individuals should have been aware of the errors and misstatements that Mr. Neptune made in his comments and 'Reviewer's Discussion'.
After no response from several notices to SAE and the Session Organizers of the errors in the Reviewer's Discussion in an attempt to get them to correct or remove the Reviewer's Discussion, we were forced to add Appendix 3: Clarifications to attempt to correct some of the errors and misstatements of the "Reviewer's Discussion". Mr. Neptune, after reviewing Appendix 3, then 'refined' his 'Reviewer's Discussion' by adding two additional paragraphs with even more errors and misstatements. We again notified the Session Organizers and SAE and their only response to our many requests that they intercede was to threaten not to publish our SAE Paper 970960 unless we allowed the unchallenged and erroneous 10 paragraph 'Reviewer's Discussion' to be published with the paper. The result, contained in SP-1237, as Reviewer's Discussion, reflects poorly on Mr. Neptune, the Session Organizers and SAE.
We requested several times to be able to add an Authors' Reply to permit a correction of the errors and misstatements made by Mr. Neptune. The SAE Session Organizers would not allow an Authors' Reply.
Authors' Reply should be permitted, when appropriate. An Authors' Reply permits a response and clarification of questions and comments contained in a Reviewer's Discussion. Many other technical societies include comments and responses and these have provided for further clarification of the subject material of technical papers. Decisions to permit an Authors' Reply should be made by a Review Referee.
For a more complete response to the erroneous review, please see our Authors' Reply for SAE 97-0960
We received notice a month or so ago that SAE was initiating a review of the Accident Reconstruction Review Process. However, the recent mailing and Call For Papers: 1998 SAE Congress by the Session Organizers indicates that no changes have been implemented in the review process set up by Terry Day and Wes Grimes. In particular, the autocracy of the session organizers and reviewers they 'choose' for particular papers can and will allow lengthy, misguided and/or intentionally erroneous "Reviewer's Discussions" to be attached to papers with no recourse for authors other than to withdraw the paper or allow drivel to be attached to the paper.
The review process has no review and/or appeals process to monitor the 'chosen' reviewers and the Reviewer's Discussion they prepare to accompany the papers. The Session Organizers demonstrated at the 1997 SAE Accident Reconstruction session that they are either unwilling or unable to maintain the professional integrity and objectivity of the review process.
Obviously, we will not subject ourselves to this flawed review process and therefore will not submit any papers for publications to SAE until the situation is rectified. In the meantime, we will seek other more objective scientific avenues for publication.
The cronyism of the SAE Accident Reconstruction pseudo "Peer Review" process continues.
We have received no response from SAE or the session organizers with regard to our questions and concerns about our "Peer Review" experience at the 1997 SAE Accident Reconstruction session. We recently received a copy of the Final Program for the 1998 SAE International Congress and Exposition. As part of the program we were surprised to find that a paper to be published is entitled "Crush Stiffness Coefficients for M-CRASH and M-SMAC" by Mr. James Neptune. We assume that that paper must be based on a continued "critique" of our recent research paper "Effects of Restitution in the Application of Crush Coefficients", (SAE 97-0960), (or else it must be that Mr. Neptune finally realized the errors of his ways and decided to rectify his errors and mis-statements in an SAE paper form).
SAE Paper 97-0960 presented proposed analytical approaches to (a) the simulation of restitution properties in the SMAC/EDSMAC programs and (b) the introduction of restitution properties in the CRASH/EDCRASH programs. There was no mention of M-SMAC or M-CRASH in 97-0960 or any of our SAE papers. First because we don't believe an SAE technical paper and/or Accident Reconstruction "scientific session" should be used to plug commercial software programs and services. Secondly, because the proposed SMAC/EDSMAC approach has not yet been implemented in the M-SMAC program in view of the exploratory nature of the proposal and related limitations in the available crash test data. And also the M-CRASH program does not yet exist in a fully defined configuration and it is currently not available commercially.
A critique of the M-SMAC and M-CRASH programs, which are trademarks and products of McHenry Software. Inc., obviously cannot be based on proposed analytical approaches that are clearly not yet a part of either M-SMAC or M-CRASH.
On January 7th we faxed and mailed to SAE, the Session Organizers and Mr. Neptune notification that the title and content of the paper should be carefully evaluated from the viewpoint of the potential liability associated with false and potentially libelous statements about the commercial products of McHenry Software.
We stand behind the contents of SAE paper 97-0960 and will be pleased to respond to legitimate comments (See Questions/Answers related to SAE 97-0960). However, SAE and the AR Session Organizers allowance of Mr. Neptune to publish a pseudo "peer reviewed" critique of our research paper (97-0960) cannot serve as the basis for an attack on a currently unrelated commercial product.
Given our unpleasant experience with Mr. Neptune during the 1997 SAE review process (see above) which included his many errors and mis-statements as part of both the review process and his Reviewer's Discussion and given the 1998 Session Organizers apparent intentional decision to drop both Ray and myself from the review committee for the 1998 session (we were on the review panel for the 1997 session and should have been likely candidates to review Mr. Neptune's paper), we are skeptical as to the scientific nature and merit of the as yet unseen paper.
February 1998 update:
At the 1998 SAE Accident Reconstruction session, a paper 980029 was published in SP-1319 by James Neptune entitled "Crush Stiffness Coefficients, Restitution Constants and a Revision of CRASH3 & SMAC" which is a continuation of his critique of our paper 970960 published at the 1997 SAE session. The following are some comments related to SAE 980029:
In his Reviewer's Discussion published with our SAE paper 970960, in SAE SP-1237, Mr. Neptune stated (erroneously) on page 325, lines 9-11, that our equation 16 can not be used when is equal to zero. In SAE 980029, he states (correctly) that his equation 2 (which is identical with our equation 16 in SAE 970960 ) reduces to a constant when is equal to zero.
Clearly, Mr. Neptune made a fundamental mathematical error in his Reviewer's Discussion of SAE 970960 .
In the last paragraph of his Reviewer's Discussion of SAE 970960 , Mr. Neptune stated that "the restitution model needs to be reconfigured". Yet in SAE 980029 he indicates without any reconfiguration that "properly determined coefficients/constants used in conjunction with the revised CRASH3/SMAC force-deflection model should result in improved accuracy over the original CRASH3 and SMAC programs".
In other words, the mathematical model, which he originally trashed, has now been found to be OK. The "problem" has been transformed into a potential misuse of the unchanged model by means of improperly determined coefficients.
We took the cautious position in our research paper (SAE 970960) that "progress towards a rigorous and complete validation study is data-limited at the present time". In other words we hope to see substantially more real-world test data before determining what constitutes "sound engineering judgment" about restitution behavior. However, Neptune offers a means to estimate the coefficients/constants in the absence of comprehensive testing, that is essentially identical to the data-fitting procedure that we defined in SAE 970960 with added warnings about the use of "sound engineering judgment".
Note that in the next to the last paragraph of his Reviewer's Discussion of SAE 970960, his "sound engineering judgment" questioned our indication of typical values of 10% to 20% restitution at 30 MPH.
The major focus of Mr. Neptune's attention in SAE 980029 is at the lower threshold of residual deformation (i.e., impacts with very nearly full dimensional recovery) which is substantially below the damage range of most injury producing accidents.
It is the same severity range in which the effects of bumper isolators may complicate the determination of effective restitution properties.
The value of restitution for the suggested revised CRASH3 model is not a single value. The revised CRASH3 model (like the original CRASH3) allows 2, 4 or 6 values of deformation to be input. The program integrates across the damage interface to initially determine the approximate absorbed energy and effective restitution properties. A secondary step of the analysis is performed in conjunction with the collision partner to then determine the effective restitution of the system (in consideration of the fact that the collision partners share restitution properties during the collision).
This was demonstrated in SAE 970960 for the revised SMAC (see figure 16 of SAE 970960 ).
The only value of SAE 980029 (since it offers nothing new except a reiteration of our paper SAE 970960) appears to be to Mr. Neptune. It is an attempt to reconstruct his reputation after the many errors and misstatements he made in his Reviewer's Discussion attached to our paper SAE 970960 in SAE SP-1237. The fact that SAE 980029 was allowed to be published demonstrates that Mr. Neptune has better connections with the SAE session organizers than we do.
December 1998 update:
The following is a letter sent to Mone Asensio and SAE via email with respect to this Accident Reconstruction session peer review. As is stated in the letter, all 4 papers we published at the 1997 session made 1997 SAE Transactions. However, they also now apparently include the erroneous Reviewer's Discussion with the papers and provide no avenue for an Authors Reply or Response (generally authors of papers in SAE Transactions are provided the opportunity to add a response to any discussions published with the paper).
To: Mone Asensio, SAE
It has been 2 years since our numerous discussions, faxes and emails concerning the flawed SAE review process for the Accident Reconstruction session. You have never provided us with any support or response to our concerns with the flawed pseudo peer review for the Accident Reconstruction Session. As you may recall there was a totally erroneous 2 page Reviewer's Discussion attached to our SAE paper 970960 which you assured me in December 1996 would not be published with the paper. Then, in early January 1997, after you rescinded on that assertion, you secondarily assured me that it would not be included in the 1997 SAE Transactions (if the paper made SAE Transactions).
Now that 1997 SAE Transactions has been released I find that the discussion is included! Please explain to me when SAE decided to start including these pseudo scientific 'peer' reviews with the papers? Are you that out of touch of what is going on at SAE that you misinformed me on two major issues or was you misinformation intentional? You will note that all 4 papers we published at the 1997 session made SAE Transactions. (970961,970947,970960, 970949) Having a paper published in SAE Transactions used to be considered an honor. But if hastily put together hack job 'pseudo-peer' reviews can be included in the transactions with the added insult of no provision for an authors response, it is obvious that there is no longer any honor or scientific integrity with SAE. You can be assured that all our future scientific papers will be published at objective scientific societies. SAE no longer qualifies as an objective scientific society.
Brian McHenry, McHenry Consultants
A 2008 book published by SAE Crash Reconstruction Research: 20 Years of Progress (1988-2007) includes our paper "Effects of Restitution in the Application of Crush Coefficients" and has affirmed the integrity of our 1997 SAE research paper.
A description of the book is as follows:
"The science of crash reconstruction enables engineers to determine the most probable scenario for how and why traffic collisions occur. Over the past 20 years, ongoing research has continually enhanced crash reconstructionists' knowledge of the application of physical laws in this field.
Crash Reconstruction Research: 20 Years of Progress (1988-2007) features 47 papers that have presented significant steps forward, focusing on the following areas within the field of crash reconstruction that have experienced major advances."
- ISBN Number: 978-0-7680-2052-6
- Date Published: March 2008
- 756 Pages 45 Papers
- Product Code: PT-138
Included in the 45 papers is SAE Paper 970960 by McHenry & McHenry, "Effects of Restitution in the Application of Crush Coefficients".
We hope you also read the 4 SAE papers we prepared with McHenry Consultants, Inc. for the 1997 SAE congress.
Also see SAE 2001 paper Reviews
We welcome any questions, comments, corrections or requests for further information. We can be reached by e-mail at McHenry@mchenrysoftware.com
For additional information, please see our website and the many technical papers referenced on the website | www.mchenrysoftware.com |
| Homesite | News! |