US 5397882

I'll search for information on this specific patent number, including any litigation activity.

Let me run additional targeted searches to confirm litigation status and check authoritative sources.

US Patent 5,397,882 — Verification Summary

Search scope and methodology (and its limits)

I searched for the exact number "5397882" and related strings (title, assignee, inventor) plus a targeted query for CAFC 2026 dockets. I did not access US PatentCenter/Patent Public Search or the CAFC docket system via an authenticated API — I worked from Google Patents (full authoritative text provided), uspto.report, and Justia, and ran open web searches for litigation. Where I could not verify something, I say so below rather than inferring.

Litigation / CAFC 2026 finding: My searches returned no CAFC 2026 docket (or any litigation) involving patent 5,397,882. This is consistent with the patent's status. Google Patents records its anticipated expiration as 2013‑05‑14 ("Expired – Lifetime"), and the last recorded fee payment was the 12‑year maintenance fee (2006). A patent that expired in 2013 would not normally support live 2026 infringement appeals. I can only say my searches found nothing; I cannot affirmatively certify that no 2026 docket exists.

---

Bibliographic data (from the authoritative patent text)

Assignee anomalies to flag (do not auto-correct):

• The Google Patents "Current Assignee" field also lists Stoebich Brandschutz GmbH, a German fire‑protection company. The underlying assignment record ("STOBICH BRANDSCHUTZ GMBH … Assignors: LUTHER, JOACHIM; STOBICH, JOCHEN," 1993‑08‑19) appears to be an unrelated assignment folded into this page. The chain that actually traces to this invention is Albert Heijn B.V. → Ahold Retail Services AG → AHOLD LICENSING SA. I treat the Stoebich listing as a probable database artifact, not a genuine co‑owner.
• The "Ahold Delhaize Licensing SARL" name is the modern successor label; the recorded 2006 event was a change of name from AHOLD RETAIL SERVICES AG to AHOLD LICENSING SA.

---

Abstract

A customer pass/card registered in the client's name is issued to each client using the self‑registration system. A check parameter, e.g. an integer, is assigned to the pass and recorded either on the pass or in a computer memory. After the client self‑scans selected articles with a portable scanner, the scanner is coupled to a readout station and simultaneously the client's identity and check parameter are determined. An automatic device (e.g. a random number generator) uses the parameter to decide whether a spot check occurs. The store layout includes shelves 3, cash desks 4, and a check selection station 22 that reads the pass and routes the client either to a check cash desk 24 (spot check by a storeworker) or directly to the readout station 6 for automatic scanner readout.

---

Independent claims — plain language

There are four independent claims: 1, 10, 21, and 30. Claims 2–9, 11–20, and 22–35 all depend, directly or indirectly, on one of these four.

Claim 1 — Method (the core concept).
A method for checking articles that a customer registered himself with a scanner in a self‑service store, comprising these steps:

1. Issue each participating customer a customer pass registered in the client's name;
2. Assign a check parameter to that pass;
3. After the customer scans his articles, couple the scanner to a readout station;
4. Simultaneously determine, via the pass, (a) the client's identity and (b) the associated check parameter; and
5. Use an automatic device to decide, starting from the check parameter, whether a check should take place.

The novelty hook relative to the cited prior art (e.g. GB 2068132, NL 8800907, and Schwartz's US 3,878,365 cashier spot‑check) is tying the spot‑check decision to a per‑customer check parameter retrieved from that customer's pass.

Claim 10 — Store arrangement (apparatus counterpart to claim 1).
An apparatus/system for practicing claim 1, comprising: (i) a readout station that reads out articles registered in a scanner; (ii) a device for recognizing a customer pass; (iii) means for automatically determining the associated check parameter; and (iv) an automatically operating device that, depending on the determined parameter, generates a signal indicating whether a check should occur.

Claim 21 — Method (expanded recitation; effectively a fuller restatement of claim 1).
A method where article information is registered in a scanner by the customer, and:

• a customer pass with first client information identifying the client is issued to each participating client, with the registration system's information processing equipment handling the pass information;
• either the processing equipment or the pass stores a check parameter (second client information) governing whether that client's self‑registration is to be checked for accuracy;
• the method then comprises: (a) coupling the scanner to the processing equipment; (b) reading the registered article information; (c) coupling the customer pass to the processing equipment; (d) determining the identity and check parameter; (e) determining, from the check parameter, whether a check should take place; and (f) carrying out the check if so determined.

Claim 30 — Self‑service article selection/registration system (system counterpart to claim 21).
A system comprising: (a) a customer pass for each participating client carrying first client information; (b) information processing means to identify the client from the pass; (c) a check parameter stored in either the pass or the processing means indicating whether that client's use of the system should be checked for accuracy; (d) readout means to couple the scanner to the processing means, read the registered article information, and issue a receipt for payment; and (e) check selection means to couple the pass to the processing means, determine the client's identity, analyze the check parameter, and generate a signal specifying whether a check should take place.

---

Notable dependent-claim features (context, not exhaustive)

• Claims 2–3: Parameter is an integer; a random integer not greater than the parameter is generated, and a check occurs if it equals a predetermined number (e.g. parameter 3 → 1/3 chance). The check is done by a storeworker who manually re‑registers the articles; the results are compared without the cashier being able to see the comparison, and one single receipt issues.
• Claim 4: The check parameter is modified based on check results (the description: lowered from 3→2 after discrepancies, or raised for honest customers — an early adaptive/self‑learning audit probability).
• Claims 5–8: Scanner dispensed only after pass identification; pass may be a contactlessly readable card with modifiable information; parameter stored on card or in a computer.
• Claim 9 / 14–16: Pre‑readout check selection station routes checked customers to a check cash desk, whose registration result is fed to the readout station.
• Claims 11–20, 31–35: Integer/random‑generator implementation, dispensing station, receipt issuance, computer/random generator as the automatic device, and check‑cash‑desk‑to‑readout‑station coupling.

---

Cited prior art of note

GB 1267630; US 3,748,191; US 3,836,755; US 3,878,365 (Schwartz — spot checking a checkout cashier); US 4,157,738; EP 0003557; US 4,180,204; GB 2064184; GB 2068132 (weight‑based check), US 4,373,133 (Clyne); EP 0086059; GB 2117709; GB 2178576; WO 1989006406 (NCR); NL 8800907 (light‑barrier article count); EP 0390448; US 5,013,896 (NCR — cashierless checkout using customer passcard).

---

Uncertainty / caveats

• No verified CAFC 2026 docket activity was found for 5,397,882; treat this as "not found in open search," not as a definitive negative.
• The Stoebich Brandschutz GmbH co‑assignee listing is almost certainly a Google Patents data artifact; I flag it but did not "correct" it.
• Inventor name appears as "Ronald J. Van Solt" (patent) vs. "Ronald Johan van Solt" (assignment microfilm) — a spelling/format divergence, not corrected.
• All bibliographic details above are drawn from the authoritative patent text you supplied; any live‑source divergence is noted rather than silently merged.

I'll research litigation involving US patent 5397882 across patent litigation databases.

Let me refine the search with more specific litigation-focused queries.

Let me query litigation-specific databases and databases that track patent cases.

Litigation Search Report — U.S. Patent No. 5,397,882

Patent identified: U.S. Patent No. 5,397,882 — "Method for spot checking a client in a self-service store." Application No. 08/063,581; filed 1993-05-14; granted 1995-03-14; original assignee Ahold Retail Services AG (later Ahold Licensing SA / current listing includes Ahold Delhaize Licensing SARL and Stoebich Brandschutz GmbH). Status: Expired – Lifetime (anticipated expiration 2013-05-14).

Bottom line

I found no evidence of any litigation involving U.S. Patent No. 5,397,882. No district court case, ITC investigation, PTAB/PTO post-grant proceeding, or CAFC appeal asserting or challenging this specific patent was surfaced.

What I searched and what I found

Important caution on false matches (do not conflate)

Because you asked me to treat identifiers literally and not auto-correct them, I flag these non-matches that appeared in search results:

1. US 5,395,788 A (Abe et al.), 7 March 1995 — surfaced via a PCT search report. This is US 5,395,788, a different patent. It is not US 5,397,882.
2. The various " ’788" litigations (e.g., IPR2025-01004 on U.S. 7,557,788; U.S. 7,203,788 / 7,946,788 cases; the AUO '788 Patent dispute) — these involve seven-digit/million-series patents ending in 788, unrelated to 5,397,882.
3. A Brazilian JUCE C document listing the number "5397882" — this is a corporate registration (NIRE) number in a Ceará state business filing, not a patent litigation reference.

None of these are relevant to U.S. 5,397,882.

Confidence and limitations

• Confidence: moderate-to-reasonable that no litigation exists, but I cannot rule it out.
• Limits: The authoritative full text you provided (Google Patents) shows no "Litigation" or "Litigation History" block, only citation, family, and legal-status events — consistent with no recorded assertions. However, Google Patents' litigation coverage is incomplete, and my searches were cut off by tool limits before I could directly query Unified Patents litigation caselist, CourtListener/RECAP, Justia Dockets, or the PTAB Patent Trial & Appeal Case Tracking System by this specific number.
• The patent is a 1995-granted, expired (2013) patent directed to supermarket self-scanning spot-check logic. It is the kind of pre-Internet-era retail patent that could have been asserted, but nothing surfaced.

Recommended next steps to confirm

To close the gap with certainty, the following should be checked directly (I was unable to complete these due to search-step limits):

• Unified Patents litigation caselist — https://portal.unifiedpatents.com/litigation/caselist (filter by patent 5,397,882)
• CourtListener / RECAP — https://www.courtlistener.com (search "5,397,882")
• Justia Federal District Court dockets
• USPTO PTAB Patent Trial & Appeal Case Tracking System (search by patent number)
• CAFC docket search

If you'd like, I can run those four targeted database queries in a follow-up pass.

I'll verify the structured-data default against live sources before writing anything.

Let me run two more targeted checks — CBM review is the most likely vehicle for a 1995 business-method patent, and I want to confirm no post-issuance challenge exists.

PTAB Proceedings — U.S. Patent No. 5,397,882

Proceedings overview

Total AIA trial proceedings on file: ZERO. No inter partes review, no post-grant review, and no covered business method review has ever been instituted, and I could find no record of a petition ever being filed against U.S. 5,397,882 — the USPTO Open Data Portal structured block returns an empty set, and independent web searches for IPR/PGR/CBM petitions naming this patent surface nothing. There is accordingly no breakdown to report: 0 active, 0 claims invalidated, 0 claims sustained, 0 settled, 0 institution denied. The bottom-line defensive posture is unusual and is driven less by the patent's litigation history than by its legal status: the patent expired on 2013-05-14 on payment of the 12-year maintenance fee, and the term has run. A demand letter citing this patent today is not a demand letter at all — there is no live exclusionary right to enforce, and pre-suit damages are time-barred under 35 U.S.C. § 286. The absence of PTAB activity is not the interesting signal here; the expiration is.

I want to be explicit about the limits of my search, per the constraint against fabrication: I am reporting the absence of proceedings, which is inherently harder to prove than a presence. I searched for IPR/PGR/CBM petitions by patent number, by the spot-checking subject matter, and by the assignee chain (Albert Heijn B.V. → Ahold Retail Services AG → Ahold Licensing SA → Ahold Delhaize Licensing SARL), plus the Stöbich Brandschutz entity that appears in the assignment data. Nothing surfaced. If a proceeding exists in a form the ODP has not ingested and that is not indexed by public web search, I did not find it, and I am not going to invent a docket number to fill the format.

---

No proceedings in existence

There is no proceeding to place at the top of this section. Rather than pad the entry with a fabricated docket, here is what a defendant actually needs to know about the file:

• Type: N/A — none filed
• Filed: N/A
• Status: N/A
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A
• Settlement / termination: N/A
• Appeal: N/A
• Defensive value: A defendant never needs to reach the PTAB merits questions, because the patent's term has expired. The defensive play is a Rule 12(b)(6) or summary-judgment motion premised on expiration and the § 286 damages bar, not an IPR petition. Filing an IPR against an expired patent is still technically permissible — the Board does entertain petitions against expired claims, applying the Phillips claim-construction standard rather than the Phillips-vs-BRI live-claim framework — but it would be burning six figures to invalidate claims nobody can assert.

---

Strategic summary

Claim status: ALL 35 CLAIMS ARE UNTESTED AND NONE ARE CANCELED — and all of them are expired. No PTAB panel has ever construed a single claim of this patent, no FWD has ever issued, and no claim has been canceled via the trial side of the Office. The claim set as published runs to 35 claims: claims 1–9 and 10–20 are the original method and apparatus claims in the 1995 grant; claims 21–35 appear in a materially different drafting style, with the independent method claim 21 and independent system claim 30 reciting "first client information," "processing equipment," and "check parameter" in a layered a/b/c/d format. That stylistic break often signals a post-issuance amendment — a reexamination certificate or a reissue — but I want to be careful: the legal-events table in the record shows no ex parte reexamination, no reissue, and no certificate of correction. I cannot reconcile the two from the record in front of me, and I am flagging it as an open question rather than asserting a reexamination occurred. Anyone relying on the claim set should pull the paper-file wrapper from Patent Center before treating claims 21–35 as part of the original grant.

Estoppel landscape: there is none, because § 315(e)(2) estoppel only attaches after an FWD in a proceeding that was instituted, and no proceeding was ever instituted. § 315(e)(1) estoppel before the Office is likewise inapplicable. This cuts in a defendant's favor — the entire prior-art universe remains available in any forum — but it is largely academic given expiration. For completeness, the art cited on the face of the patent that a challenger would have built on includes several references squarely on point for a § 103 attack on the spot-checking concept: US 3,878,365 (Schwartz, "Method and apparatus for spot checking a checkout cashier operation in a store"), GB 2,064,184 (Krone, Random Number Generating System), EP 0,086,059 (Dabey, "computers for use in sampling procedures"), GB 2,178,576 (Brown, "Random selection systems"), and US 5,013,896 (NCR, "Cashierless checkout system using customer passcard"). That is a remarkably dense cluster of anticipatory-adjacent art on the very idea of random sampling by a processing device keyed to a customer identifier — the exact subject matter of independent claim 21. Had the patent been asserted, a § 103 challenge would have had strong runway. It never got the chance.

Pattern signals: there is no pattern to report. No petitioner has ever filed against this patent, so there is no serial-filer signal, no repeat-petitioner dynamic, and no defensive aggregator (Unified Patents, RPX, or similar) in the chain. The patent owner has never had to defend claims before the Board and has therefore never pursued a PTAB appeal to the Federal Circuit on this patent — meaning no CAFC docket number exists to cite and no FWD link exists to quote. One family note worth carrying: US 5,637,846 ("Method and apparatus for electronic payment by a client in a self-service store," granted 1997-06-10) is a continuation-in-part of the application that issued as the '882 patent, claiming the same 1993-05-14 priority. A defendant evaluating a portfolio rather than a single patent should check the '846 patent's own status separately rather than assume this analysis transfers.

One data caveat I want to flag rather than silently pass over. The Google Patents "current assignee" field lists both Ahold Delhaize Licensing SARL and Stöbich Brandschutz GmbH. The Stöbich entry traces to an assignment recorded 1993-08-19 of inventors Stöbich and Luther — a German fire-protection concern with no apparent relationship to a Dutch grocery self-scanning system. This is almost certainly a mis-keyed reassignment record in the source data, not a genuine co-ownership. If you are doing diligence on chain of title, verify through the USPTO Assignment database rather than relying on the aggregator field.

---

Recommended next steps

• If you have received a demand letter citing US 5,397,882: the letter is defective on its face. The patent expired 2013-05-14 ("Expired - Lifetime" in the ODP record, "Anticipated expiration 2013-05-14"). Confirm the maintenance-fee history and the expiration in Patent Center, then respond on that basis. There is no claim to license and no injunction available.
• Do not file an IPR. Even setting aside cost, an expired patent cannot support exclusionary relief, so there is no damages or injunction exposure left for an IPR to neutralize. The Board does institute against expired claims in some circumstances, but there is no strategic payoff here.
• If you are being threatened on a related patent — most plausibly US 5,637,846, the CIP in the same family — do not carry this conclusion across. That is a different patent with a different claim set and its own maintenance history, and it needs its own expiration and ODP check. The prior-art cluster identified above (Schwartz, Krone, Dabey, Brown, NCR) is nonetheless relevant to the whole family, since the '846 CIP expressly incorporates the '882 disclosure by reference.
• If you want affirmative confirmation of the empty PTAB set, the authoritative check is USPTO PTAB E2E (https://ptacts.uspto.gov/ptabweb/) searched by patent number, cross-referenced against the Patent Trial and Appeal Board's published decisions page. CourtListener (https://www.courtlistener.com/) can confirm the absence of a Federal Circuit appeal, which follows automatically from the absence of any FWD.
• The absence of PTAB activity is not, on these facts, the usual "unasserted patent" signal. Normally, a patent that has never drawn an IPR after thirty years is a patent nobody bothered to assert. Here the more likely explanation is that the patent lapsed at the end of its statutory term in 2013 before the self-scanning checkout wars generated the kind of assertion pressure that produces IPR activity — and by the time the industry's self-checkout litigation wave arrived, the '882 patent was already in the public domain.

Confidence note: I am highly confident that no AIA trial proceeding exists for this patent, based on the empty ODP structured block plus corroborating searches. I am not asserting the absence of district court litigation, ITC investigations, or ex parte reexamination activity — I did not find any, but I did not conduct an exhaustive litigation-docket search, and the claim-numbering anomaly in claims 21–35 in particular suggests something happened post-grant that the legal-events table does not record.

I'll verify the assignment chain against live sources before reconstructing it.

Let me dig for the recorded correspondent data and any assertion history.

Assignment & Ownership Analysis — US 5,397,882

Verification status: I retrieved the patent's own Google Patents legal-events feed (which carries reel/frame for four post-issuance-style records) and cross-checked Ahold corporate history via third-party sources. The USPTO Assignment Center / assignment.uspto.gov query interface did not return machine-readable records to my searches, so correspondent-of-record data could not be retrieved for any link in this chain. I flag that explicitly rather than guess — no correspondent name is asserted below. Verify correspondents directly at USPTO Assignment Center (search by patent number 5397882).

---

Inventors

Unusual patterns: None. There is a single inventor, no co-inventor fragmentation, no evidence of inventors departing the original assignee within 12 months, and no inventor-to-NPE assignment anywhere in the chain.

---

Original assignee

On the issued patent (1995-03-14): Ahold Retail Services AG — but this is the second owner. The first owner of the application was Albert Heijn B.V. (the operating Dutch grocery chain; assignment reel 006632/0127, executed 1993-06-24), which transferred the application to its Swiss group IP vehicle before grant (reel 006874/0919).

• Primary line of business: food retail / supermarket operation. Albert Heijn B.V. is the flagship Dutch supermarket operating company of the Ahold group; Ahold Retail Services AG (Zug, Switzerland) is a group IP-holding/retail-services vehicle, not an operating retail chain itself.
• Product embodying the claims: Yes, in the sense that matters here. The patent's own disclosure is an implementation of self-scanning / self-registration checkout — portable scanner dispensed against a customer loyalty pass, read out at the exit, with parameterized spot checks. This is the Albert Heijn "self-scanner" store system, and Ahold's descendant Ahold Delhaize continues to file self-checkout/mobile-checkout patents (e.g. WO2022123451A1 "Store checkout verification system"; US20210090056A1; EP4000051A1 "Integrated autonomous checkout system") — a self-consistent, still-investing operating-retailer portfolio.
• Current status: The Ahold entity is an operating, publicly traded group — Ahold Delhaize N.V. (Euronext Amsterdam / AEX: AD; formed 2016 from the Ahold–Delhaize merger). No insolvency, no reorganisation, no Chapter 7/11. The patent itself expired — Lifetime, anticipated expiration 2013-05-14, per Google Patents legal status.

Anomaly to flag (not auto-corrected): Google's legal-events feed also lists an assignment recorded 1993-08-19 at reel 006744/0137, assignors Stobich, Jochen and Luther, Joachim, assignee STOBICH BRANDSCHUTZ GMBH (Germany). Those inventors and that assignee have nothing to do with this patent (Stöbich Brandschutz is a fire-protection company; this patent's only inventor is Van Solt). This is almost certainly a feed cross-link artifact — the same record appears under unrelated German fire-protection filings, and it has propagated into the "Current Assignee" field as "Stoebich Brandschutz GmbH." I record it literally, per instruction, but treat it as not part of this patent's true chain and exclude it from the ownership diagram.

---

Assignment timeline

1. 1993-06-24 (executed) / recorded 1993-06-30 — Reel 006632/0127

   • Conveyance: Assignment
   • Assignor: Ronald Johan Van Solt (sole inventor)
   • Assignee: Albert Heijn B.V.
   • Correspondent: Not retrievable from the sources available to me (Google Patents legal events carry reel/frame only; Assignment Center query returned no data). No correspondent asserted.
   • Context: Initial inventor→employer assignment; standard employed-inventor filing, not a fire-sale.

2. [1993-08-03] / recorded 1993-08-19 — Reel 006744/0137

   • Conveyance: Assignment
   • Assignor: Stobich, Jochen; Luther, Joachim
   • Assignee: STOBICH BRANDSCHUTZ GMBH
   • Correspondent: Not retrievable.
   • Context: Anomalous — likely mis-associated record. Inventor and assignee do not match this patent. Documented for completeness, excluded from the chain and diagram.

3. 1994-01-26 (executed) / recorded 1994-02-22 — Reel 006874/0919

   • Conveyance: Assignment
   • Assignor: Albert Heijn B.V.
   • Assignee: Ahold Retail Services AG (Switzerland)
   • Correspondent: Not retrievable.
   • Context: Internal group reorganisation — operating subsidiary transfers the application into the Ahold group's Swiss IP-holding company before grant. No arm's-length third party.

4. 2006-08-23 (effective) / recorded 2006-10-06 — Reel 018362/0117

   • Conveyance: Change of Name (only)
   • Assignor: Ahold Retail Services AG
   • Assignee: Ahold Licensing SA (Switzerland)
   • Correspondent: Not retrievable.
   • Context: Change of name only — same legal entity re-badged; no transfer of title to a new party. Consistent with the 2006 Ahold restructuring into a dedicated "Licensing" IP vehicle.

5. 2017-09-13 (executed) — name change to Ahold Delhaize Licensing SARL, Rue Jean-François Bartholoni 4-6, Geneva, Switzerland

   • Source: USPTO TTAB opposition record 91270443 (Registrant ownership section), which traces Ahold Licensing SA → Ahold Licensing SARL (executed 2009-06-26) → Ahold Delhaize Licensing SARL (executed 2017-09-13). This step is not itemised as a separate assignment record in the Google Patents legal-events feed for US 5,397,882 — it appears only via the "Current Assignee" field and the parallel TTAB filing. Treat the exact reel/frame for this step as unverified.
   • Context: Change of name only, reflecting the Ahold–Delhaize merger.

Fee-payment events (not assignments): 1998-08-24 (4-yr), 2002-08-22 (8-yr), 2006-08-18 (12-yr). Maintenance was paid to full term; anticipated expiration 2013-05-14.

Bottom line: There is no post-2006 transfer out of the Ahold corporate family, and no assignment to any entity outside the operating retailer's own IP-holding structure at any point.

---

Timeline diagram

timeline
    title Ownership of US 5397882
    1993 : Filed by Albert Heijn B.V.
         : Van Solt assigns to Albert Heijn
         : Recorded reel 006632 frame 0127
    1994 : Assigned to Ahold Retail Services AG
         : Recorded reel 006874 frame 0919
    1995 : Patent issued 14 March
    2006 : Change of name to Ahold Licensing SA
         : Recorded reel 018362 frame 0117
    2013 : Patent expired 14 May
    2017 : Name change to Ahold Delhaize Licensing SARL

---

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. Every successive holder is part of one corporate family: Albert Heijn B.V. → Ahold Retail Services AG → Ahold Licensing SA → Ahold Delhaize Licensing SARL. The "Licensing" suffix here is a captive group IP vehicle of an operating grocery retailer, not a single-purpose Delaware/Texas assertion LLC. No registered-agent-service address, no single-member LLC structure, no products-free shell. Reels 006874/0919 and 018362/0117 (the latter is a change of name only, not a transfer).

2. Known asserter in the chain — Not present. No assignee or assignor matches Acacia, Marathon, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, or any Spangenberg entity, and no Unified Patents / RPX high-frequency plaintiff appears. The chain never leaves Ahold.

3. Repeat correspondent across the chain — Unclear (data unavailable). Google Patents legal events carry reel/frame but no correspondent-of-record, and my Assignment Center queries returned no records. I therefore cannot assess recurrence of a recording attorney or firm across reels 006632/0127, 006874/0919, and 018362/0117. This is a genuine gap, not a negative finding — the correspondent field is the single highest-value field for this analysis and should be pulled manually from Assignment Center.

4. Cascading transfers — Not present. Only two transfers occurred, both within ~9 months of filing (1993-06-24 and 1994-01-26) and both intra-group; the ownership then sat static for ~19 years until expiry. No chained LLC-to-LLC laddering, no shared correspondent addresses, no principals in common with any assertion entity.

5. Pre-litigation transfer — Not present. No infringement suit naming US 5,397,882 was identified. The most recent transfer (2006 change of name) predates expiry by seven years and is unrelated to any litigation timing.

6. Bankruptcy fire-sale — Not present. Ahold/Albert Heijn never filed Chapter 7/11; the group is an operating, listed retailer. No proceedings-based sale of this patent.

7. Privateering — Not present. No transfer to a third-party NPE asserting on Ahold's behalf, and no SEC 10-K/8-K or Patent Progress / EFF coverage connecting this patent to an assertion campaign.

8. Defensive aggregator (anti-NPE) — Not present, but note the neighbouring fact. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at a captive operating-company IP holder. The patent has nevertheless been neutralised by expiry (2013-05-14), so it is non-assertable today regardless.

---

Verdict

Defensive / non-asserting.

Justification: the entire recorded chain stays inside one operating retailer's corporate family — inventor Van Solt → Albert Heijn B.V. (reel 006632/0127, 1993-06-24) → Ahold Retail Services AG (reel 006874/0919, 1994-01-26) → Ahold Licensing SA, a change of name only (reel 018362/0117, effective 2006-08-23) → Ahold Delhaize Licensing SARL (2017 name change per TTAB record 91270443). There is zero sign of a licensing-only NPE: no external transferee, no assertion entity, no cascading LLC laddering, no bankruptcy, and no identified suit, and the patent expired 2013-05-14 and is unenforceable today. Two caveats to keep the call honest: (a) the taxonomy's "defensive" bucket is defined as termination at an aggregator such as RPX/AST, which is not literally the case — strictly this is a corporate IP-holding reorganisation with no assertion history, i.e. non-NPE and non-asserting; and (b) the correspondent-of-record was not retrievable, so signal 3 (repeat correspondent) remains genuinely unverified and should be confirmed at USPTO Assignment Center.

Data-quality warning carried forward: the "Stoebich Brandschutz GmbH" assignee appearing in the patent's current-assignee list and the 1993-08-19 reel 006744/0137 record are mismatched to this patent's inventor and subject matter and should be treated as a feed artifact pending confirmation at Assignment Center.

I'll verify key prior-art references before finalizing. Let me run targeted searches on the most relevant citations.

Good verification on the core references. Let me verify the remaining citations.

I've reached the search-step limit, so I'll finalize based on the authoritative full text (Google Patents mirror of USPTO data) plus the four verification searches that completed. Below is the prior-art analysis.

---

Prior-Art Analysis — U.S. Patent No. 5,397,882

Patent under analysis: U.S. 5,397,882 A, "Method for spot checking a client in a self-service store." Application 08/063,581; filed 1993‑05‑14; granted 1995‑03‑14; inventor Ronald J. Van Solt; original assignee Ahold Retail Services AG (assignment chain: Albert Heijn B.V. → Ahold Retail Services AG → Ahold Licensing SA; current listing also shows Ahold Delhaize Licensing SARL and Stoebich Brandschutz GmbH). Status: Expired – Lifetime (anticipated expiration 2013‑05‑14).

Number-integrity note (per your strict rule): I searched only for 5397882 and the specific citation set tied to it. I did not conflate this patent with similar numbers that appeared incidentally in search results — e.g., US 5,395,788 A (Abe et al., 1995), the various "'788" litigations (U.S. 7,557,788 / 7,203,788 / 7,946,788), or the Brazilian JUCE "5397882" corporate-filing number flagged in the earlier litigation section. Those are different identifiers and are excluded here.

---

A. The prior art at issue

The authoritative record lists 17 references cited by the examiner ("Patent Citations (17)"), all marked as examiner citations. Every one of them was published before 1992‑05‑14, i.e., more than one year before the 1993‑05‑14 U.S. filing date. Accordingly, all 17 qualify as prior art under both 35 U.S.C. § 102(b) (printed publication/patent more than one year before filing) and § 102(a) (published before the applicant's invention). No reference falls in the 1992‑05‑14 to 1993‑05‑14 window, so no § 102(a)/(e) "in-between" analysis is needed.

Summary table

---

B. Preliminary: an effective-date flag on claims 21–35 (do not skip)

The authoritative text prints "Claims (35)" for US 5,397,882. Claims 1–20 are the original set (method claims 1–9; apparatus claims 10–20). Claims 21–35, however, are drafted in the "information processing equipment / processing means" idiom and are structurally identical to the claim set that issued on the continuation‑in‑part US 08/278,872 → US 5,637,846 A ("Method and apparatus for electronic payment by a client in a self-service store"), filed 1994‑07‑21.

This matters for § 102 because a CIP's new matter is entitled only to the CIP filing date (1994‑07‑21), not the parent's 1993‑05‑14 date. I flag this as a record ambiguity to verify against the file history: if claims 21–35 contain matter not supported by the 1993‑05‑14 disclosure, their critical date for § 102 would be 1994‑07‑21. Practically, this changes nothing here, because every one of the 17 cited references published by 1991 — well before either critical date. But it is the kind of discrepancy worth stating rather than silently resolving. (Consistent with your instruction, I have not "auto-corrected" the claim count; I report the conflict between the printed 35 claims and the CIP family relationship.)

---

C. Reference-by-reference analysis

For each reference I give full citation, dates, a brief description, and the claim(s) it potentially implicates under § 102. Because a § 102 anticipation requires a single reference to disclose every element of a claim, I distinguish between references that disclose elements of the independent claims and any that could attack a claim as a whole.

1. GB 1267630 A — "Automatic marketing system"

• Assignee: FMC Corp. Priority: 1969‑05‑16. Published: 1972‑03‑22. § 102(b): yes.
• Description: A very early automated retail/checkout merchandising system — an "automatic marketing" architecture predating barcode self-scanning. (Description inferred from title and era; full text not retrieved.)
• § 102 impact: Background art only. No customer pass, no per-client check parameter, no spot-check determination. Does not anticipate any claim. Relevant only as evidence of the long-standing, crowded field of automated checkout.

2. US 3836755 A — "Self-service shop"

• Assignee: Gretag AG. Priority: 1972‑02‑14. Published: 1974‑09‑17. § 102(b): yes.
• Description: Self-service store with article identification and automated registration/settlement. Gretag was an identification-systems vendor; the reference concerns machine-readable article identification in a self-service environment.
• § 102 impact: Potentially relevant to the general "self-registration in a self-service store" environment (claim 1 preamble) and to card-based identification. No check parameter / no spot-check → no anticipation of any independent claim.

3. US 3749191 A — "Apparatus and method to facilitate shopping"

• Inventor: R. Ashen. Priority: 1972‑03‑23. Published: 1973‑07‑31. § 102(b): yes.
• Description: A shopping-aid apparatus (portable/self-help shopping device) of the early 1970s.
• § 102 impact: Perimeter art on portable shopping assistance. Does not anticipate any claim.

4. US 3878365 A — "Method and apparatus for spot checking a checkout cashier operation in a store" ⭐ (most relevant to the spot-check concept)

• Inventor: Hermann Schwartz. Priority: 1972‑11‑03 (CH 16059/72). Published/patented: 1975‑04‑15. § 102(b): yes. (Verified full text.)
• Description (verified): Customers feed information plates to an automatic checkout register. A high-frequency clock pulse train is compared for coincidence with a customer-initiated pulse train; a coincidence randomly selects a customer for a spot check. The spot check blocks the automatic register's normal completion and requires the customer to go to a manual checkout counter, where a cashier re-registers/prices the articles manually. Unknown to the customer and the cashier, the manual and automatic totals are compared; discrepancies are recorded and reported to a central location (with repeat-comparison logic, an indicator, a recorder, and optionally a camera). The passages are explicit that selection should be random so as not to embarrass the customer, and that discrepancies are not made known to the persons involved.
• § 102 impact — this is the single most pertinent reference to the invention's purpose:
  • Claim 1: Discloses an automatic random determination of whether to check, the manual re-registration by a cashier, and the covert comparison — but it does not disclose (i) a client customer pass issued in the client's name, (ii) a check parameter assigned to that pass, or (iii) determination of the client's identity and the associated check parameter by means of the pass at readout. It also presupposes a cashier-fed register, not client self-scanning. → Does not anticipate claim 1 as a whole, but it is the strongest § 103 obviousness reference against claim 1 when combined with a customer-pass self-checkout reference (e.g., US 5013896 A or GB 2068132 A / US 4373133 A).
  • Claim 3 (and family claim 24): Discloses nearly every element — the check is carried out by a cash desk storeworker who manually registers the articles; the manual result is compared (in the register/control equipment) with the automatic result; the comparison is not observable by the cashier/customer; and a single sales slip is produced by one of the registers. → Potentially anticipatory as to the manual-check/covert-comparison/single-receipt features of claims 3 and 24, subject to those claims' dependency on the check-parameter/random-integer limitations of claims 1–2 and 22–23.
  • Claims 2 / 22 (random selection): Discloses random selection, but by pulse coincidence, not by generating a random integer bounded by a per-client check parameter. → Element-level relevance only.

5. US 4157738 A — "Method for counting the number of articles using a weighing machine"

• Assignee: Shinko Denshi Company Limited. Priority: 1977‑05‑15. Published: 1979‑06‑12. § 102(b): yes.
• Description: Weight-based counting/verification of articles.
• § 102 impact: Cited as background to the weight-verification approach that the specification criticizes (the GB 2068132 method). Does not anticipate any claim of the '882 patent — the '882 patent deliberately departs from weight checking.

6. EP 0003557 A2 — "Payment and check point for self-service stores"

• Inventor: Hartwig Dr.-Ing. Benzler. Priority: 1978‑02‑09. Published: 1979‑08‑22. § 102(b): yes.
• Description: A combined payment and check point arrangement for self-service stores — the closest cited art conceptually to the '882 "check selection station / check cash desk" architecture.
• § 102 impact: Potentially relevant to claims 9, 14, 15, 29, 34 (a check point that routes a client for verification before/at settlement). Full-text comparison needed; on its title/subject it lacks the customer-pass-encoded check parameter and the probability-based spot-check determination, so it does not anticipate the independent claims.

7. US 4180204 A — "Automatic inventorying system"

• Assignee: The J. C. Penney Corporation, Inc. Priority: 1978‑11‑08. Published: 1979‑12‑25. § 102(b): yes.
• Description: Automated inventory data-capture system (portable data entry and machine-readable codes).
• § 102 impact: Background on portable/non-cashier article-data capture. Does not anticipate any claim.

8. GB 2064184 A — "Random Number Generating System" ⭐ (random-selection family)

• Assignee: Krone GmbH. Priority: 1979‑11‑10. Published: 1981‑06‑10. § 102(b): yes.
• Description: A random-number generating system (appears in the same citation lineage as checkout/register art).
• § 102 impact: Directly relevant to the random-generator limitation of claims 2, 11, 18, 22, 31, 32 ("generating a random integer"). Element-level only — it does not disclose bounding the random integer by a stored check parameter nor the customer-pass context. Not a standalone anticipation.

9. GB 2068132 A — "Method for producing a bill, during the collection of items in a self-service shop" ⭐ (self-scanning family)

• Inventor: L. H. Wright. Priority: 1980‑01‑03. Published: 1981‑08‑05. § 102(b): yes. (Verified via the US counterpart's text.)
• Description: Self-service shop in which the customer collects items and a portable device registers each item; at the sales terminal the device's data are read, the item collection is weighed, the measured weight is compared with the computed weight of the registered items (within tolerances), and a bill is produced if they match; if not, the collection is re-priced item by item in conventional fashion. An item-count and price/weight printout can be produced; the unit can be linked to a cash register and central data source.
• § 102 impact: This is one of the two references the '882 specification itself identifies as the closest prior art. It discloses the self-registration + readout-at-terminal + verification skeleton (claims 1, 3, 9 elements) but achieves verification by weight comparison, not by a per-client probabilistic check parameter. Does not anticipate the independent claims; strong § 103 material on the "verify at the sales terminal" concept.

10. US 4373133 A — "Method for producing a bill, apparatus for collecting items, and a self-service shop" (U.S. counterpart of GB 2068132 A)

• Inventor: Nicholas Clyne. Priority: 1980‑01‑03. Published: 1983‑02‑08. § 102(b): yes. (Verified full text.)
• Description: As above, with additional detail: the portable collecting unit stores item identities; at the terminal a central data source supplies price/gross weight; the print-out gives each item against its price and weight plus aggregate totals; the measured vs. computed weight comparison is automated with a red/green signal; an item count helps "detect fraud."
• § 102 impact: Same as GB 2068132 A — discloses self-registration, readout, comparison and receipt production, but not the customer-pass-encoded check parameter nor its use to decide whether to check. No anticipation of claims 1–2 / 21–23; strong § 103 background. (Note: because GB 2068132 A and US 4373133 A share the 1980‑01‑03 priority, they are effectively the same disclosure in two jurisdictions — do not treat them as two independent teachings.)

11. EP 0086059 A2 — "Improvements in or relating to computers for use in sampling procedures" ⭐ (random-sampling family)

• Assignee: Dabey Ltd. Priority: 1982‑02‑02. Published: 1983‑08‑17. § 102(b): yes.
• Description: Computer apparatus for sampling procedures — i.e., automated/statistical selection of items or cases to inspect.
• § 102 impact: Relevant to the "automatic device determines whether a check should take place" limitation (claims 1, 2, 10, 11, 21, 22, 30, 31). Element-level: it teaches automated sampling but not the per-customer check parameter stored on/in a pass or account that scales the selection probability. Not a standalone anticipation.

12. GB 2117709 A — "Random discount rate machine"

• Inventor: Sebastiao Dilson Drumond. Priority: 1982‑03‑17. Published: 1983‑10‑19. § 102(b): yes.
• Description: A machine applying a random rate (discount) — an automated random-selection device in a retail context.
• § 102 impact: Marginal; supports the general proposition that random selection devices were known and applied in retail (claims 2/11/22/31 element). Not an anticipation.

13. GB 2178576 A — "Random selection systems"

• Inventor: Lawrence David Brown. Priority: 1985‑08‑03. Published: 1987‑02‑11. § 102(b): yes.
• Description: Random selection systems (generic random-selection technology).
• § 102 impact: Same category as GB 2064184 A / GB 2117709 A — element-level support for the random-generator/random-selection limitations. Not an anticipation.

14. WO 1989006406 A1 — "Checkout system and method" ⭐ (cashierless/passcard family)

• Assignee: NCR Corporation. Priority: 1987‑12‑28. Published: 1989‑07‑13. § 102(b): yes. (Verified full text.)
• Description (verified): Cashierless checkout in which each customer is issued an identification item/identification code at an entry terminal (which scans the merchandise codes), pays at a separate payment terminal that reads/writes the identification item with "payment received" data, and an exit gate control unit reads the identification item and permits egress only if payment is complete. The entry terminals are physically separated from the payment terminals; a data-processing means checks full payment. This is the PCT parent of US 5013896 A.
• § 102 impact: Discloses a customer-identification item issued per customer and its use at a readout/decision station that automatically gates the client — i.e., elements of claims 1, 9, 14, 15, 30, 34. It does not disclose a check parameter governing whether an accuracy spot check occurs. → Element-level; not an anticipation of the independent claims. (Again: WO '406 and US 5,013,896 are the same NCR disclosure — one teaching, not two.)

15. US 5013896 A — "Cashierless checkout system using customer passcard" ⭐ (customer-pass element)

• Inventors/Assignee: Ono et al. / NCR Corporation. Priority: 1987‑12‑28. Patented: 1991‑05‑07. § 102(b): yes. (Verified full text.)
• Description (verified): Entry terminals scan goods; a passcard is issued magnetically recording the customer's ID number and total amount owed; the customer pays at a money-receiving terminal whose passcard read/write section records payment; an exit-gate passcard checker and gate controller release an exit gate only after correct payment; a control and processing unit (CPU) coordinates and stores sold/unpaid/paid data. Claims cover both the system and a method of checking out goods including "reading the passcard … checking the passcard … enabling the exit gate."
• § 102 impact — the best cited reference for the customer-pass/identity element:
  • Claim 1: Anticipates the elements of a store customer pass recording the customer ID, issuance per customer, and identification of the client by means of the pass at readout/decision. It does not disclose (i) assignment of a check parameter to the pass/account nor (ii) use of that parameter to decide whether an accuracy check should occur. → Does not anticipate claim 1, but is a strong § 103 partner for claim 1 (combined with US 3878365 A for the spot-check selection and a random-generation reference).
  • Claims 7 / 8 (card information modifiable; parameter stored on card or in computer): Discloses a read/write passcard whose stored data can be modified and a computer file associated with the ID → element-level relevance to claims 7 and 8 (and claim 27/28/30) — but the data modified is payment data, not a check parameter.
  • Claims 10, 17, 30 (readout station; identification of a customer pass; processing means): Discloses apparatus equivalents (scanning entry terminals, passcard checker, CPU) → element-level relevance, no anticipation of the check-parameter signal.
  • Claim 5 / 26 (supplying a scanner to a client after pass identification): Note a disconnect — NCR issues a passcard after scanning; it does not dispense a portable scanner upon pass recognition. → Claim 5/26 is not anticipated.

16. NL 8800907 A — "Supermarket shopping trolley with inbuilt bar code reader and memory; memory contents compared with laser scan at checkout without removing articles" ⭐ (self-scanning family; applicant's own art)

• Assignee: Ahold N.V. Priority: 1988‑04‑08. Published: 1989‑11‑01. § 102(b): yes.
• Description (per the Google Patents English abstract, verified): A shopping trolley with an inbuilt bar-code reader and memory; the memory contents are compared with a laser scan at checkout without removing the articles from the trolley. This is the reference the '882 specification discusses (checking article number by a light barrier above the trolley and reading a number of codes in a lock, then comparing with the registration).
• § 102 impact: Discloses client self-registration with a scanner/memory and comparison of the registration at checkout (claims 1, 3, 9 elements). It does not disclose the customer-pass check parameter or probabilistic spot-check decision. → No anticipation of the independent claims; this is the pre-art the patent was drafted against, and it is from the assignee's own corporate family (Ahold), so it appears here as disclosure, not as third-party art.

17. EP 0390448 A1 — "Portable checkout system"

• Assignee: NCR Corporation. Priority: 1989‑03‑27. Published: 1990‑10‑03. § 102(b): yes.
• Description: A checkout system built around a portable (hand-held) scanning terminal for self-registration.
• § 102 impact: Discloses the portable scanner self-checkout environment (claim 1 preamble; claim 12 element of a dispensing/portable scanner aspect). Does not disclose the check parameter or the check-selection decision. → Element-level; no anticipation of the independent claims.

---

D. Most relevant prior art — ranked

1. US 3878365 A (Schwartz, 1975) — the closest art on the core inventive concept: automatic, randomized selection of a customer for a spot check, a manual re-registration of the articles as the check, a covert comparison against the machine record, and production of a single slip. It is the reference the specification's own background analysis arguably tracks (the patent distinguishes weight-count and trolley-lock checks, not Schwartz), which makes it the most likely predicate for any § 103 challenge to claims 1–4 / 21–25.
2. US 5013896 A (Ono/NCR, 1991) — the closest art on the customer-pass / client-identity element (passcard recording ID; use of the pass at a readout/gate station to decide passage). Strong § 103 partner with #1 for claim 1.
3. GB 2068132 A (Wright, 1981) / US 4373133 A (Clyne, 1983) — the recognized "self-scan + verify-at-the-terminal" art (weight, not probability, based). Same disclosure, two jurisdictions.
4. NL 8800907 A (Ahold, 1989) — applicant-family self-scanning trolley with registration comparison; the art the spec explicitly distinguishes.
5. WO 1989006406 A1 (NCR, 1989) + EP 0390448 A1 (NCR, 1990) — cashierless/portable checkout with identification item and gated exit (same NCR family as #2).
6. GB 2064184 A (Krone, 1981), EP 0086059 A2 (Dabey, 1983), GB 2117709 A (Drumond, 1983), GB 2178576 A (Brown, 1987) — the random-number / random-sampling cluster supporting the "random integer" limitations of claims 2, 11, 18, 22, 31, 32.
7. US 4157738 A (Shinko Denshi, 1979) and US 4180204 A (J.C. Penney, 1979) — weight-counting and automated inventory-capture background.
8. GB 1267630 A (FMC, 1972), US 3836755 A (Gretag, 1974), US 3749191 A (Ashen, 1973), EP 0003557 A2 (Benzler, 1979) — early automated-retail / self-service / check-point background.

---

E. What the cited art does NOT disclose (why the independent claims issued)

Reading all 17 citations together, no single reference discloses the combination that defines claim 1 (and 21 / 30):

1. a customer pass registered in the client's name issued to each self-registration user AND
2. a check parameter assigned to that pass/account AND
3. determination, via the pass at readout, of both the client's identity and the associated check parameter AND
4. use of that parameter by an automatic device to decide whether an accuracy check takes place.

The two "legs" of the invention are split across the art — US 3878365 A supplies the randomized spot-check leg (without a pass/parameter), and US 5013896 A / GB 2068132 A / US 4373133 A supply the self-checkout-at-readout leg (without a spot-check decision). The claimed per-client parameter that tunes the check probability (claim 2's integer N → 1/N check chance) and its adaptive modification based on check results (claims 4/25) appear to be absent from every cited reference — which is consistent with claims 4/25 having no anticipation candidate in this set.

---

F. Limitations and confidence

• Source integrity: Analysis is grounded in the authoritative patent text (citation list, claim set, specification) you supplied, corroborated by four completed verification searches for US 3878365 A, US 5013896 A, GB 2068132 A / US 4373133 A, and NL 8800907 A (full or near-full text retrieved) and a partial result for GB 2064184 A. Descriptions for GB 1267630 A, US 3749191 A, US 3836755 A, US 4157738 A, EP 0003557 A2, US 4180204 A, EP 0086059 A2, GB 2117709 A, GB 2178576 A, and EP 0390448 A1 are based on their titles/subjects and field knowledge, not on retrieved full text (lower confidence) — I was cut off by search-step limits before verifying them.
• § 102 conclusions are "potential," not final. A definitive anticipation finding requires element-by-element comparison against each reference's full disclosure. What I can state with high confidence is the absence, across the entire cited set, of the check-parameter-based spot-check decision and its adaptive modification; what I cannot fully exclude without full texts is whether an individual foreign reference (e.g., EP 0003557 A2, EP 0086059 A2) contains incidental support for one of the other limitations.
• No litigation section restated (per your instruction); the earlier "no litigation found" conclusion is unaffected by this citation analysis.
• Recommendation: to close the § 102 gaps, pull the full texts of the ten unverified references (start with EP 0003557 A2 and EP 0086059 A2, the two most likely to bear on the check-point / sampling limitations), and check the '882 file history (USPTO PatentCenter) to confirm (i) how claims 21–35 came to be in this record and (ii) which references the examiner actually applied versus merely listed.

I'll ground the obviousness analysis by pulling details on the key prior-art references cited on this patent page.

Let me pull the remaining key references (the random/sampling-selection art and the weight-based self-scan check art).

Obviousness Analysis — U.S. Patent No. 5,397,882 under 35 U.S.C. § 103

Patent: US 5,397,882 A — "Method for spot checking a client in a self-service store" (App. 08/063,581; filed 1993‑05‑14; granted 1995‑03‑14)
Governing law: Pre‑AIA § 103 (application filed before 2013‑03‑16). All seventeen cited references pre‑date the filing date by more than one year and are therefore § 102(b) prior art.
Framing: Graham v. John Deere Co., 383 U.S. 1 (1966) (scope/content of prior art; differences; PHOSITA level; objective indicia), as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) (predictable variations, "obvious to try," market demand/design incentive, teaching‑suggestion‑motivation as one of several rationales).

---

0. Methodological caveat (read this first)

The prior‑art listing I was given supplies numbers, dates, titles, and assignees only. I retrieved substantive text for three references — US 3,878,365 (Schwartz), US 5,013,896 (NCR/Ono et al.), and WO 1989/006406 (NCR) (Google Patents / Espacenet / patentimages). For the remaining references — GB 2 064 184 (Krone), GB 2 178 576 (Brown), EP 0 086 059 (Dabey), GB 2 068 132 (Wright), US 4,373,133 (Clyne), NL 8800907 (Ahold), US 4,157,738 (Shinko Denshi), GB 2 117 709 (Drumond), EP 0 003 557 (Benzler), US 3,748,191, US 3,836,755, US 4,180,204, GB 1 267 630, EP 0 390 448 — I have only the bibliographic record. My search steps were exhausted before I could pull their full texts. I therefore state what these references appear to teach from their titles/classification and from the '882 specification's own characterization of them, and I flag every place where my mapping rests on the title rather than the disclosure. This is not a substitute for a full‑text FTO/validity chart.

One more threshold point: this is a hypothetical validity analysis. Per the earlier sections of this report, no litigation, PTAB proceeding, or opposition involving 5,397,882 was found, and the patent expired 2013‑05‑14. Its validity has never been adjudicated.

---

1. Level of ordinary skill in the art (PHOSITA)

Circa May 1993, a PHOSITA would be a retail‑systems engineer with a B.S. in electrical engineering or computer science (or equivalent experience) and roughly 2–3 years working with point‑of‑sale terminals, bar‑code scanning, and supermarket front‑end operations and shrink‑control procedures. Critically, the skill set includes familiarity with store security/loss‑prevention practice, not merely electronics. The invention adds no new hardware and no new algorithm of any depth — it is a systems‑integration and data‑flow invention. This is a modest skill level, which cuts toward obviousness.

---

2. What the '882 claims actually add over the admitted prior art

The specification itself concedes the state of the art. It states that self‑service stores "where clients themselves, by means of a scanning device, scan a code of each article … and register same in the memory of the scanning device" are "well known," citing GB 2 068 132 and NL 8800907. It further concedes weight‑comparison verification (GB 2 068 132) and light‑barrier/code‑lock verification (NL 8800907) as known, and criticizes both as costly.

Stripped to its essentials, the only point of novelty relative to that admitted art is:

A persistent, per‑client "check parameter" — carried on the customer pass or in the store computer — that modulates that individual client's probability of being spot‑checked, and (in claim 4) that changes based on the outcome of prior checks.

Every other element of independent claim 1 appears individually in the cited art:

So the § 103 question is narrow: was it obvious, as of 1993, to replace Schwartz's fixed‑probability random spot check with a per‑customer, stored, adjustable probability keyed to the passcard that NCR and the '882 specification's own admitted art already required the customer to present?

---

3. Reference‑by‑reference scope and content

US 3,878,365 — Schwartz, "Method and apparatus for spot checking a checkout cashier operation in a store" (1975). Retrieved full text. This is the single most important reference. It discloses: generating a random control pulse by coincidence of a customer‑arrival pulse with a free‑running clock pulse; using that pulse to force a selected customer through a manual checkout; comparing the manual total against the automatic register's total; suppressing the automatic receipt so the customer "attaches no significance" to the check; and referring discrepancies "either immediately or on a record to appropriate administration personnel." Schwartz expressly frames the goal as making spot checks of "about five percent" sufficient, and as unobtrusive and non‑invidious ("selection in this form is completely at random [so it] can have no invidious connotations and the privacy and personality rights of the customers are fully respected"). (https://patents.google.com/patent/[US3878365A](/patent/US3878365A)/en ; https://patentimages.storage.googleapis.com/49/e2/39/d4fc30218c0b7e/US3878365.pdf)

US 5,013,896 — Ono et al., NCR, "Cashierless checkout system using customer passcard" (filed 1987‑12‑28, granted 1991‑05‑07). Retrieved full text. Discloses a passcard magnetically recorded with the customer's ID number and the transaction amount; a passcard read/write section; a control and processing unit (CPU) that stores sold‑goods, unpaid‑goods, and settlement data keyed to the customer ID; and an exit gate control unit that reads the passcard and opens a gate conditioned on the stored data. This is the "customer pass + central computer with per‑customer records + gate" architecture read onto claims 10/14/15/30/33/34. (https://patents.google.com/patent/US5013896)

WO 1989/006406 — NCR, "Checkout system and method" (filed 1987‑12‑28, published 1989‑07‑13). Retrieved full text. Substantially the same architecture as the '896 patent: entry terminals issue an "identification item" bearing an identification code; payment terminals read/write it; a data processing means checks full payment; an exit gate control unit permits passage. Directly supports the claims that recite a check‑selection station that gates the customer down one of two routes.

GB 2 064 184 — Krone GmbH, "Random Number Generating System" (1979/1981). Title/date only. The title and the '882 classification (G07G1/0054, "supplementary check parameters") indicate a store/checkout‑related random‑number generator. I rely on it only as generic evidence that random number generation for retail selection decisions was known and conventional — a weak point in my mapping that should be verified against full text.

EP 0 086 059 — Dabey Ltd., "Improvements in or relating to computers for use in sampling procedures" (1982/1983). Title/date only. A reference whose express subject matter is computer‑implemented sampling. This is the reference I would most want full text for: if it discloses a selectable or variable sampling rate, it is the strongest bridge between Schwartz's fixed‑5% selection and the claimed per‑customer parameter. My mapping here rests on the title.

GB 2 178 576 — Brown, "Random selection systems" (1985/1987) and GB 2 117 709 — Drumond, "Random discount rate machine" (1982/1983). Title/date only. Both confirm that randomized selection/outcome generation tied to a customer transaction was a familiar technique in the retail context by the mid‑1980s.

GB 2 068 132 — Wright (1980/1981) and US 4,373,133 — Clyne (1980/1983). Cited by the '882 specification as the weight‑comparison self‑scan check. They establish the self‑registration + POS verification baseline that the '882 specification admits is old.

NL 8800907 — Ahold N.V. (1988/1989). Cited and characterized by the specification itself as the light‑barrier/lock code‑verification system. Notably, this is the patentee's own earlier art, and the specification admits it is workable but "laborious" and expensive — an admission that verification of self‑scan accuracy was a recognized, long‑standing problem in the field.

Remaining references (US 3,748,191; US 3,836,755; EP 0 003 557; US 4,180,204; US 4,157,738; GB 1 267 630; EP 0 390 448) are background — self‑service shop and article‑counting/weighing systems. They are cumulative context, not primary obviousness references.

---

4. The primary combinations

Combination A (strongest): Wright/Clyne self‑scan art + Schwartz '365 + NCR '896 [+ Dabey '059]

This is the combination I would expect an examiner or a defendant to run, and it renders independent claims 1, 10, 21, and 30 obvious.

• Self‑scan by the customer (the preamble of claims 1 and 21): admitted prior art (GB 2 068 132; US 4,373,133; NL 8800907).
• Customer pass identifying the client + computer with per‑customer records (claims 1, 21, 30, 33): NCR '896 / WO 1989/006406, verbatim.
• Readout/coupling of stored registration data to a POS device that issues a receipt (claims 10, 13, 30, 35): NCR '896's entry terminal + printing section; also the admitted self‑scan art.
• Automatic random spot‑check decision, manual re‑check, concealed comparison, single receipt (claims 1, 2, 3, 24): Schwartz '365.
• Sampling/random‑number selection as a computer‑implemented, parameterized function (claims 2, 11, 18, 22, 23, 31, 32): GB 2 064 184; GB 2 178 576; EP 0 086 059.

The only element not found in that combination is the limitation that the selection probability is stored per customer and derived from the pass. And that limitation is supplied by the intersection of the two primary references: NCR '896 already stores per‑customer ID‑keyed data in the CPU and already reads and writes the passcard at multiple stations. The '882 specification itself makes the point that this is a mere choice of data location: "The parameter is preferably written magnetically or electronically in the customer pass, but may also be stored in the memory of the computer with the name of the client."

Combination B: WO 1989/006406 + Schwartz '365

NCR's WO '406 supplies the passcard‑plus‑gate architecture (claims 10, 14, 15, 16, 30, 33, 34, 35); Schwartz supplies the random spot‑check engine (claims 1, 2, 3, 21). Because WO '406's exit gate control unit already gates a customer based on data read from that customer's identification item, reciting that same gate to route a customer to a check cash desk versus a readout station (claims 9, 14, 15, 34) is the predictable use of a disclosed element for its established function — precisely the KSR inquiry.

Combination C: GB 2 068 132 / NL 8800907 + Schwartz '365 + NCR '896

Useful if the examiner frames claims 3, 9, 24, and 29 (manual re‑registration, comparison, one receipt, two‑route routing) against a self‑scan baseline. The '882 specification's own admission that these systems verify self‑scan accuracy supplies the problem statement, and Schwartz supplies the solution structure.

---

5. Motivation to combine — the articulated rationale

Courts require a reason why a PHOSITA would have combined these teachings. Several independent, non‑hindsight rationales are available:

1. Same field, same problem, same solution architecture. All three primary references address self‑service store checkout. Schwartz's own text articulates the exact problem the '882 patent claims to solve: "As the checkout operation is improved … it becomes more and more necessary to provide checks that will avoid mistakes and will defeat attempts to cheat. A system of spot checking the operation is usually a sufficient safeguard." The '882 patent states the identical object ("a simpler and yet efficient method for checking articles selected and registered for payment by a client"). This is direct, express motivation — not imputed hindsight.

2. The self‑scan system's own economics demand selective verification. The '882 specification concedes that full verification (weight, light barrier, code lock) is "sizeable and costly." A PHOSITA implementing self‑scan therefore must sample, not verify everything — the very premise of Schwartz's "five percent is normally regarded as sufficient."

3. The data needed to vary the rate already existed and was already stored. NCR '896's control and processing unit stores per‑customer transaction records; Schwartz already records and reports discrepancies to administration. Once discrepancy history is stored against a customer ID, using that history to set that customer's sampling probability is an improvement in one of the prior art elements according to known methods — the classic KSR formulation, "the predictable use of prior art elements according to their established functions."

4. Statistical sampling was a known, named computer application. EP 0 086 059's very title is "computers for use in sampling procedures." A PHOSITA seeking to allocate a finite number of check‑cash‑desk resources would look to the sampling‑plan art, where varying sampling rates by stratum (here, by customer) is elementary. (Caveat: I could not verify Dabey's disclosure; this rationale is strong only if the full text confirms variable sampling rates.)

5. Obvious to try / finite predictable solutions. Given the identified need, the design space was small and wholly predictable: (a) where to store the parameter — on the card or in the computer; and (b) how to convert it to a decision — generate a random integer in a range determined by the parameter and compare to a constant. Claims 7, 8, 27, and 28 recite exactly the two storage options as alternatives.

6. No teaching away, and no unexpected result. Nothing in Schwartz, NCR, or the admitted art teaches that all customers must be sampled at the same rate; nothing suggests that varying the rate would fail. The '882 patent alleges no surprising result — the 1/N probability arithmetic in the specification is ordinary probability theory.

---

6. Claim‑by‑claim

---

7. The patentee's best non‑obviousness arguments, and responses

A competent challenger should expect these and be ready:

(a) "Schwartz teaches away." Schwartz emphasizes that selection must be "completely at random" so it "can have no invidious connotations," and that its object is to protect customer "privacy and personality rights." One can argue this counsels against treating customers differently. Response: Schwartz's concern is that store personnel not hand‑pick customers; it is directed at human discretion, not at statistical differentiation. Schwartz itself records discrepancies "to appropriate administration personnel" — the very data a per‑customer parameter would consume. And the '882 patent's stated purpose is efficiency, not stigma.

(b) "No reference suggests storing a check parameter on a pass." True as a literal matter for the references I retrieved. Response: NCR '896 stores customer data keyed to the passcard ID, and the '882 specification concedes the on‑card vs. in‑computer location is a design choice. KSR permits this kind of predictable substitution. This is the argument most likely to survive summary judgment, so the Dabey '059 full text and any contemporaneous loss‑prevention literature should be developed to strengthen it.

(c) Objective indicia. I found no evidence of the classic Graham secondary considerations with a nexus to the claimed feature: no litigation (per the earlier sections of this report), no established commercial‑success nexus, and no long‑felt‑need evidence tied to the per‑customer parameter specifically. The long‑felt problem of checkout congestion was solved by the earlier self‑scan art, not by this patent. Absent such evidence, the secondary‑consideration factor is neutral or favors obviousness. I state this as "not found," not as a verified negative.

---

8. Conclusion

Independent claims 1, 10, 21, and 30, and the overwhelming majority of the dependent claims, would have been obvious under § 103 over the art cited on the face of the patent. The strongest single combination is:

GB 2 068 132 / US 4,373,133 / NL 8800907 (admitted self‑scan self‑registration) + US 5,013,896 and/or WO 1989/006406 (customer passcard, per‑customer CPU records, gating) + US 3,878,365 (automatic random spot check, concealed manual comparison, single receipt) + GB 2 064 184 / GB 2 178 576 / EP 0 086 059 (random‑number and computer‑implemented sampling selection).

The motivation is not merely imputable — it is express in Schwartz's own specification, which identifies the self‑scan accuracy problem, prescribes spot checking as the solution, and states that a low fixed sample rate suffices. Converting that fixed rate into a stored, per‑customer, adjustable rate is a predictable variation employing data (per‑customer transaction and discrepancy records) the NCR references already stored.

Confidence: Moderate‑to‑high as to claims 1, 2, 10, 11, 21–23, 30–32 and the routing/gating claims (9, 14–16, 29, 34, 35), where the retrieved full texts of Schwartz '365 and NCR '896 map cleanly. Moderate as to claims 4 and 25 (adaptive parameter modification) and qualified as to any rationale that depends on EP 0 086 059 (Dabey), GB 2 064 184 (Krone), or GB 2 178 576 (Brown), for which I had titles and dates only. Before any final conclusion, I would pull the full texts of Dabey '059, Krone '184, and Brown '576 — those three, and especially Dabey, determine how strongly "variable sampling rate" can be attributed to the prior art rather than to hindsight.

Caveat repeated: This is a hypothetical validity analysis of an expired, never‑litigated patent. It is not a legal opinion and should not be relied upon as such.