Other I/M Technology.

In addition to its other activities, Gordon-Darby continues to work on developing new vehicle inspection related products that are designed to complement our testing and program management services, including:

TransMass™, a low-cost software-based testing approach, was developed by Gordon-Darby to predict mass emissions during a transient drive cycle (e.g., the IM240). This patented approach to transient emissions testing uses BAR97-type equipment to determine mass emissions scores without the need to measure total exhaust flow. States can use this solution to generate near-IM240 SIP credits.

The TransMass™ methodology bridges the gap between less effective concentration-based emissions testing and conventional, higher cost transient mass emissions inspections. Because if does not require the use of expensive or high maintenance flow measurement hardware such as CVS or VMAS™ equipment, it can be cost-effectively implemented in either centralized or decentralized test networks.

Gordon-Darby developed TransMass™ to address this shortcoming in vehicle exhaust (tailpipe) emissions testing. Because it involves use of industry standard transient drive cycles, this solution generates near-IM240 credits without suffering from the limitations of ASM testing. When combined with centralized or decentralized BAR97 caliber equipment, a transient drive trace and descriptive vehicle data, TransMass™ accurately predicts transient mass emissions. It is a customizable, software-based solution that allows a user to select the desired stringency (failure rate) level based on the magnitude of the required air quality improvements or SIP reductions, without resulting in excessive false failures. Preliminary studies show TransMass™ identifying 80%-95% of IM240-identified excess emissions while keeping errors of commission (false failures) at low levels.

The customizable nature of TransMass™ means that errors of commission and omission (false failure and false pass) thresholds can be set on a community-by-community basis. Communities that can tolerate higher levels of false failures can reduce their false passes, thereby generating increased emissions credits. Conversely, communities that might not need the same level of air quality benefit can reduce their false failures by increasing their false pass threshold.
In addition, the use of low cost BAR97 type equipment (gas analyzers and dynamometer) with no need for flow measurement means that TransMass™ can be implemented as either a centralized or decentralized solution. There is no costly or complex flow measurement equipment to install, operate and maintain, making it ideal for a decentralized application involving licensed repair shops.

The bottom line is that TransMass™ provides a very attractive alternative to ASM and conventional METT testing that combines the best features of both.

Historically, when states and communities needed more emissions credits than they could obtain from Idle or Loaded Mode testing, their testing options were the ASM test, or a Mass Emissions Transient Test (METT) like the IM240 or IM147. When properly implemented, the conventional METT is proven to correlate to the Federal Test Procedure (FTP) better than any other form of tailpipe emissions test. As such, it offers the greatest emissions credits. For communities that need maximum air quality benefits, METT testing is an attractive option.

But for all its benefits, a METT program has some shortcomings. It is inherently complex. That complexity, along with the equipment involved in a proper implementation, makes METT testing costly, both in terms of initial and ongoing operational costs. This is particularly true in decentralized programs. Even VMAS™, which is marketed by Sensors as a low-cost approach to transient testing, has been found in actual program implementations to involve additional complexities in equipment installation, operation, maintenance and auditing.

The ASM concentration test is positioned as a lower cost, less complex alternative to METT testing that lends itself to both centralized and decentralized testing. By design, ASM testing generates more emissions credits than Idle or Loaded Mode testing, but fewer than the IM240. However, ASM has a serious shortcoming in that it artificially loads vehicles with its heavy-load fixed-speed design and as such likely results in greater numbers of false failures. The US EPA has responded to this design deficiency by releasing modified final vehicle cutpoints (standards) that are designed to minimize such false failures. However, states such as Virginia that have implemented the new US EPA standards have found them to result in significantly fewer total test failures, thus reducing the overall effectiveness of their vehicle inspection programs.

Entry of vehicle identification information has always been a source of frustration with inspectors in vehicle test programs, since it can be a somewhat lengthy and unforgiving process (if you make a mistake entering this information you may have to start all over). Unless tightly controlled through proper display prompts and entry logic, it can also result in significant entry errors that distort subsequent analysis of the data. I/M programs have been able to reduce vehicle information entry time and errors by sending information from the VID in response to inspector entry of specified vehicle parameters (usually the Vehicle Identification Number and / or license plate number). However, problems still exist with this process. The VID information may have errors (e.g., if they exist in the vehicle registration database that was the source of the information). This requires inspectors to overwrite and re-enter the “correct” information, which is only as accurate as the care taken in entering the modified information.

We developed the Gordon-Darby VIN decoder by analyzing millions of records from our I/M programs as well as large datasets of additional vehicle identification information obtained from other sources. The resulting database system has been deployed in Gordon-Darby programs to fill in almost all required vehicle information at the beginning of each test based on the VIN entered by the inspector, regardless of whether an online or offline test is being conducted. VIN decoder content is also being expanded on an ongoing basis as additional information is collected during its use in programs such as New Hampshire where vehicles not included in traditional emissions inspection programs (e.g., farm vehicles and trailers) are now being inspected and adding to our decoded VIN database system.

Off-line inspections are also problematic. Since no vehicle information is available from the VID, inspectors must enter it all, sometimes incorrectly. Some (mainly centralized) vehicle inspection programs have therefore also incorporated VIN decoding at the test unit to aid inspectors in properly identifying test vehicles and minimize vehicle identification errors.

Various VIN decoding products have been marketed to the vehicle inspection industry for use at both the VID and test unit end of the vehicle inspection process. However, these other decoders are adaptations from other applications and suffer from a variety of shortcomings. In response to the previous lack of an I/M-specific VIN decoder, Gordon-Darby developed a customized VIN decoding-based database system specifically designed for vehicle inspection purposes that is designed to reside onboard vehicle test units and populate display screens and test records with correct vehicle information. This system is also suitable for VID use in checking VINs arriving both from the test units and from download of state vehicle registration data.

Overall, the Gordon-Darby VIN decoder provides a substantial step forward in improving the speed and ease of inspections while also increasing the accuracy of the resulting data. We continually work on and consider further enhancements to the VIN decoder. This type of improvement will further increase the utility of the decoder, making it an ideal product for use at either the VID or test unit end of the vehicle inspection process. Besides being used in Gordon-Darby-managed inspection programs, it is also available to other programs under a licensing arrangement.

The implementation of OBD testing on 1996 and newer model year passenger cars and light trucks by most U.S. vehicle inspection programs has been accompanied by concerns about possible “clean scanning.” This term refers to an illegal practice in which a known clean vehicle is subjected to an electronic OBDII scan in place of the vehicle that is supposed to actually be tested. All new cars and light trucks beginning with model year 2008 are required to have the VIN embedded in the OBDII data stream to allow such fraudulent behavior to be detected. Other measures are needed, however, to detect and deter clean scanning on 1996-2007 models that do not have an embedded OBDII VIN (this is available on some but not all pre-2008 models).

To better address this issue, Gordon-Darby developed and implemented a custom OBDII “fingerprinting” technique that enables advanced clean scanning identification. This proprietary methodology analyzes many more parameters than proposed by others in order to accurately identify (fingerprint) actual test vehicles down to the sub-model level. The resulting fingerprint is then compared to reference fingerprints contained in a database assembled from millions of OBDII test records in Gordon-Darby programs, all of which have involved the collection of the parametric information needed to build the OBDII fingerprint database. The end result is the most complete method available, short of the collection of an electronic OBDII VIN, for detecting incidences of illegal clean scanning.

Various methods, involving such OBDII parameters as PID count, PCM module ID and individual communications protocol, have been proposed for clean scanning detection on pre-2008 OBDII equipped vehicles. These approaches require the development of reference databases (truth tables) containing the appropriate parametric values for specific makes and models, against which values collected for an individual vehicle can be compared to verify that at least the proper type of vehicle is being tested. However, such truth tables must first be developed. In addition, some of the parameters involve the same non-unique values for multiple vehicle models, making the matching of actual test parameters to the benchmark values less than ideal for use in identifying test fraud.

While our fingerprinting methodology and database are an integral part of Gordon-Darby’s OBDII inspection program implementations, they are also available for licensing to other programs along with technical support in ensuring that all required parametric data are collected during the vehicle test process in order to enable proper fingerprint matching.

Gordon-Darby developed and trademarked the RIMS™ test technology, which involves remote oversight of OBDII kiosk inspections conducted in either decentralized or centralized inspection networks, and data exchange between the remote test kiosks and a centralized database. Using the RIMS™ inspection process / technology, inspection results (including a digital video display of the inspection as it is being performed) are transmitted via any of several possible communications networks (e.g., the Internet, leased phone line, dedicated phone line, satellite connection, etc.) to a central database and oversight location.

The inspector or motorist follows the prompts displayed on the hand-held unit to conduct the RIMS™ inspection, including the following inspection elements:


  • Initial entry of any required vehicle information. The inspector/motorist is prompted to scan and/or train the video camera incorporated into the unit on both the vehicle license plate and barcoded VIN. (Barcode scanners are typically used for this purpose, but recent technological advances have made it now feasible to capture the barcode on video and then decode it using available software.)
  • Connection of the OBDII connection cable to the vehicle diagnostic link connector (DLC). The mobile computing unit then performs the required on-board diagnostic system.
  • Disconnection of the OBDII cable, which would be followed by display of the inspection results.

A series of “close-up” digital photographs is taken during the inspection. Each close-up photo also triggers an overhead photograph. The entire dataset (OBDII results and digital photographs) is then transmitted to the central database and remote overview location. Test supervisors monitor, review, and approve in-progress testing in real-time prior to the completion of the inspection, thereby providing an effective means to detect and deter so-called clean scanning. The data is also retained for audit purposes.

KioskA relatively small number of supervisors can observe a large number of tests. In addition to observing and reviewing vehicle tests, they can provide assistance in conducting inspections. This latter element would parallel the “lead inspector” concept that is typically used to improve inspection performance in conventional centralized programs. An enhanced audit mode built into the RIMS™ software also allows the use of streaming video.

The RIMS™ approach combines the motorist convenience of decentralized inspections with a level of oversight similar to that found in centralized inspection networks, thus combining the best features of both networks. It also has considerable flexibility and can be readily adapted to other test environments. For example, the same approach could be used in centralized inspection facilities to eliminate possible test fraud.