NOTE: COMMENTS REGARDING ANY FEDERAL REGISTER NOTICE MUST BE SENT TO THE ADDRESS INDICATED IN THE DOCUMENT. ANY COMMENTS ON THE RAPID INFORMATION BULLETIN BOARD SYSTEM (RIBBS) ABOUT ANY FEDERAL REGISTER NOTICES WILL NOT BE USED OR CONSIDERED IN THE COURSE OF ANY RULE MAKING. ----------------------------------------------------------------------- POSTAL SERVICE Presort Accuracy Validation and Evaluation (PAVE) AGENCY: Postal Service. ACTION: Notice of establishment of program. ----------------------------------------------------------------------- SUMMARY: This notice adopts standards for the Presort Accuracy Validation and Evaluation (PAVE) Program. EFFECTIVE DATE: August 24, 1994. FOR FURTHER INFORMATION CONTACT: George T. Hurst, (202) 268-5232, or Lynn Martin, (202) 268-5176. SUPPLEMENTARY INFORMATION: On May 23, 1994, the Postal Service published in the Federal Register proposed standards for the Presort Accuracy Validation and Evaluation (PAVE) Program (59 FR 26609-26614). PAVE was proposed as a voluntary program in which the Postal Service would, upon request, provide testing for certain categories of presort software and hardware products to determine their accuracy in sorting address information according to the mailing standards of the Domestic Mail Manual (DMM). This program would assure those using presort software packages that a PAVE-certified product, if used properly, would have the capability of performing its intended function according to the current mailing standards of the Postal Service. The deadline for submitting comments on the proposed program was June 15, 1994. All comments received or mailed by that date have been considered. The Postal Service received comments on the proposed program from five different commenters. On the basis of the comments received, the Postal Service has decided to adopt the PAVE Program as proposed. Evaluation of Comments Received One commenter commented that optical character readers (OCRs) should be tested by electronic media rather than the proposed physical test deck because manufacturers of OCR equipment and software cannot control the additional manual preparation that must accompany their process to prepare actual mailings. Most OCR software has parameter settings allowing users to assign certain groupings of mail to specific sort or stacker bins on the machine. Once mail is processed through the OCR, operators must manually remove the groupings from these bins and further arrange them into a properly packaged and/or trayed mailing. An electronic test file could be used to evaluate an OCR's software logic for sorting addresses to bin-type designations. However, this test would not be a true test of the ability of the presort software to follow accurately all the presort standards of a specific category; moreover, the test would not indicate whether the hardware could follow the sortation logic being tested. For these reasons, the Postal Service has determined that the use of a physical test deck of actual mailpieces for PAVE testing of OCRs is a more relevant measurement of an OCR's overall capabilities. Although such a test will require additional manual intervention to complete, this too is a useful indicator of the software/hardware developer's understanding of presort standards. One commenter stated that there should be mailer or user PAVE certification in addition to certification at the developer or manufacturer level. The Postal Service has determined that too many users of presort software and hardware products exist to attempt to certify them all under this type of program. Although it is understood that individuals can misuse even the best of software or hardware, the resources needed to evaluate the volume of potential end users of such products would not be cost effective. Because misuse of a PAVE-certified presort product could result in presort errors, the Postal Service does not propose to grant unique rate eligibility to users of PAVE-certified products. The Postal Service will continue to verify presort rate eligibility as it is done today and to pursue other options for evaluation of presort product end users. Referring to the physical OCR test deck, one commenter inquired about the availability of labels and tray tags. In addition, this commenter inquired about the information that would be required on packages and tray labels. PAVE Program standards will require that an OCR examinee presort, package (if appropriate), tray, and return the physical test deck to the Postal Service National Customer Support Center as though the examinee were submitting an actual mailing. Thus the information printed on packages and tray labels must conform to the information required by the DMM standards for the specific presort category being evaluated. PAVE Program participants will be able to obtain the same labels, tray tags, and other mail preparation supplies as all mailers are authorized from post offices. Two commenters stated that PAVE participants would likely have difficulty determining when a change to a presort product was significant enough to warrant recertification. The commenters stated that the proposed guidelines do not cover the variety of relatively insignificant changes that might be made to presort products in any specific year. One commenter stated that the use of a toll-free telephone number to determine the significance of presort product changes would be a good idea. PAVE Program standards note that a change significant enough to necessitate recertification would be a key alteration of the basic sortation logic of a presort product; a major change in the content, layout, format, or availability of computer-generated documentation or facsimiles; or a modification that caused significant differences in software operator use. Changes of less significance would not require recertification. The significance of some changes may require additional evaluation to determine whether recertification is warranted. The Postal Service will establish a toll-free telephone number to share information with presort product developers about the types of changes that would require a product to be recertified under PAVE. In addition, the Postal Service will begin to assemble and maintain a list of presort product alterations previously ruled on for their significance toward requiring recertification. Once developed, this list will be made available to customers on request. Two commenters requested that tests be included for first- and third-class flat-size barcoded categories, and one commenter inquired about a physical test deck for flats-processing OCRs. The Postal Service has decided initially to test the categories noted in the original proposal as follows: (1) Presorted first-class letter-size; (2) First- and third-class barcoded letter-size; (3) Second-class presort flat-size (carrier route, 3/5 digit, and basic); (4) Third-class presort letter-size (3/5 digit and basic); and (5) Third-class carrier route presort flat-size. The Postal Service has determined that these categories provide a wide yet manageable range of presort categories most often used for the initial test cycle. Though not able to test all presort categories every year, the Postal Service intends to include new testing categories in future years. The inclusion of more testing categories of flats will be examined at a later date. The future selection of a physical test deck category for flats barcoding OCRs will also be deferred to a later date and will then be evaluated based on the number of such machines in use by the mailing industry at that time. Also inquiring about the physical test deck, one commenter wanted to know where the barcode would be printed on the test piece to determine whether equipment adjustments would be necessary. In addition, this same commenter wanted to know how equipment that could not read barcodes would be certified. The physical test deck for OCRs used in the initial test cycle will have addresses and barcodes printed on inserts to appear through window envelopes in the lower right corner. The characters of the recipient's address will be of Gothic Text font in 10-point type; the same Helvetica-type, OCR-readable font used in the physical test deck for Postal Service Multiline Accuracy Support System (MASS) certification. Barcodes will be printed below the address on the insert in the lower right barcode clear zone as noted in Domestic Mail Manual C840.2.7. The addresses and barcodes will use black ink on a white background to establish the best readability conditions practicable. Participants will be notified if any change to this format is adopted for subsequent test cycles. One commenter stated that advance knowledge of ZIP Code ranges used in the physical test deck would be beneficial for individual users of OCRs. The commenter also stated that many OCR users develop local sort schemes to perform finer sortations of local mail in fewer passes on their equipment. These sort schemes are based on mail volume for specific geographical areas and cannot efficiently handle a wide variety of ZIP Codes. The commenter stated that advanced knowledge of ZIP Code ranges used in the physical test deck would allow such users time to reprogram their sort schemes to handle the ZIP Codes contained in the test. As noted, the Postal Service does not plan to offer PAVE testing to the end users of presort products. However, manufacturers of presort products may share this same concern. Because the PAVE test cycle is relatively lengthy, running from August to the end of December, participants can obtain information on ZIP Code ranges essentially 5 months in advance of the certification deadline by ordering the tests as soon as they become available. The Postal Service has determined that this is sufficient advance notice of testing information to accommodate all participants. One commenter wanted to know how the Postal Service will grade physical test decks. This same commenter wanted to know what would be considered a passing grade. Physical test decks will be graded the same as electronic test files except that proper tray preparation will also be evaluated. Examinees will be responsible for ensuring the accuracy of the presort, the accuracy of the supporting documentation required to accompany the mailing statement, and the accuracy of hard copy documentation to support the accurate sortation and rate application of the mailpieces in the test deck. The entire test deck must be presorted, trayed, labeled, and appropriately documented, without error, according to the standards of the DMM, to obtain PAVE certification. One commenter wanted to know whether PAVE certification obtained on one model of OCR equipment would be transferable to similar OCR models offered by the same developer or manufacturer. Just as different versions of presort software will have to be individually PAVE-certified, the Postal Service has determined that different models of hardware will also require individual certification. Although different hardware models from a single manufacturer may share the same software and similar hardware sorting mechanisms, if the manufacturer of a piece of equipment determines that enough difference exists between it and another product to assign the equipment a unique model number or name, the Postal Service will treat the product as a different piece of equipment requiring its own PAVE certification. This is consistent with Postal Service certification programs for address matching accuracy and barcode quality. One commenter inquired whether postal personnel will be required to observe the processing of the physical test deck. The Postal Service has no plans at this time to conduct on-site observation of the processing of the test decks but may re-evaluate its position on this issue in the future if it appears that on-site observation is needed to ensure the integrity of the test. One commenter stated that OCR equipment cannot document overflow trays because the operator must decide how much mail will flow to the final tray and physically place that mail in the tray. This undocumented overflow makes some of the PAVE tests irrelevant. As previously noted, the Postal Service recognizes that some manual intervention will be required to complete the PAVE test deck for OCRs. Thus, individuals, rather than the specific equipment or software being tested, may in some instances determine the final placement of mailpieces in trays including overflow trays. Although non-OCR presort software can adhere to tray volume parameters established in the electronic PAVE test files and develop overflow trays of predictable volume, OCRs have difficulty determining exact tray volume because the operator makes this decision independent of the equipment. For this reason, the PAVE certification process does not require that OCR examinees provide computer-generated volume reports for such trays. However, manual documentation of overflow trays will be required to meet the DMM documentation requirements. One commenter stated that because the address matching process of the MLOCR is supposed to be turned off for PAVE processing of the physical test deck, further clarification is needed to understand how to report 5-digit, 9-digit, and delivery point barcodes and numeric ZIP Codes in the results. In addition, this commenter wanted to know why the test will require the reporting of 9-digit barcodes, which are not acceptable except for barcode rates for flats. Electronic test files for PAVE will contain some addresses with 5- digit ZIP Codes and some addresses with 11-digit numeric ZIP Code information (5-digit ZIP Code, 4-digit add-on, and 2-digit delivery point code). Address matching mechanisms are not required to process these files and should not be used to determine the completeness of the address information. Examinees should treat the 5-digit addresses as noncodable and sort them as pieces that do not qualify for barcode rates, while treating the 11-digit addresses as correctly coded qualifying pieces. Similarly, the physical test decks for PAVE testing of OCRs contain pieces that have 5-digit barcodes and pieces that have complete delivery point barcodes. The 5-digit barcoded pieces in the OCR test decks have corresponding 5-digit ZIP Codes in the address. The delivery point barcoded pieces in these test decks have corresponding 11-digit numerics printed in the address (5-digit ZIP Code, 4-digit add-on, and the 2-digit delivery point code--the correction character is not given numerically, but is included in the actual barcode on the piece). The 5-digit numeric/barcoded pieces are to be treated as nonqualifying, whereas the 11-digit (delivery point barcoded) pieces are to be treated as qualifying for the barcode rate. Although addresses in the PAVE test files have been selected so that matching or ``cleansing'' processes should not corrupt test results, it is recommended that they not be used. PAVE certification does not require the identification and reporting of 9-digit barcodes. Two commenters expressed concern about the time frame established for returning test results, stating that more time may be required to reprocess PAVE tests if the first test failed. Even though the Postal Service has established November 15 through December 15 as the period for official evaluation and response of PAVE test results, it will strive to evaluate tests on receipt and provide results as quickly as possible to participants. The Postal Service will pay particular attention to examinees whose tests cannot be certified so that those examinees can be provided as much time as possible for retesting. Two commenters suggested that the timing of the PAVE cycle may conflict with a relatively busy season in the mailing industry. Both commenters recommended that the Postal Service adhere to the current suggested time frames but remain flexible to re-examine this issue in the future. The Postal Service has determined that for 1994, the PAVE cycle will begin immediately, and end with the December 31 retesting deadline. However, the Postal Service will remain open to future recommendations to shift the PAVE cycle to meet industry needs. After considering these comments, the Postal Service has determined to implement the PAVE Program, effective immediately, as described in the Federal Register notice published on May 23, 1994 (59 FR 26609- 26614). To obtain detailed information on participation in PAVE, presort product developers may request the PAVE Program Technical Guide from the Postal Srvice National Customer Support Center by calling 1- 800-331-5746, extension 651 or 454. Participants may use the PAVE order form, included in that guide, to order PAVE tests. Stanley F. Mires, Chief Counsel, Legislative. [FR Doc. 94-20781 Filed 8-23-94; 8:45 am]