US 4675647

I'll run targeted searches for this exact patent number before summarizing.

Let me check for litigation/docket activity and confirm the family details.

US 4,675,647 — Verified Summary

Search note first: I ran targeted searches for the exact token 4675647 against patent databases, litigation databases, and docket-related sources. The only U.S. patent bearing that number is the queue-management patent below. Searches for CAFC 2026 dockets involving this number returned no matching litigation or appeal — see "Litigation / docket status" at the end, where I flag what I could and could not verify.

Bibliographic data

Uncertainty flag on filing date: the authoritative Google Patents text in my source says filing date 1982-07-14, and the family's PCT application (PCT/SE1982/000244) was filed 1982-07-14. One EPO family record lists the U.S. application ("US 47687683 A") with the date 1983-03-18, which likely reflects national-stage/entry data rather than the original filing. I am reporting the 1982-07-14 date as primary and noting the discrepancy rather than silently reconciling it.

Abstract (as published)

A system for serving customers in a specific queue sequence at a plurality of service points in a bank or post office, for example, makes it possible for customers to select a specific service point, at which service is desired, while maintaining a turn-number system which is common to all the service points. If a customer indicates, when being allocated a turn-number, that he selects a desired service point, the system registers the indication in a memory unit. The customers are served in numerical sequence at the service points which are free, and any selective customers are served at their desired service point, preferably when the correct number comes up, and as the selected service point becomes free.

Independent claims — plain-language overview

The patent has one independent claim (claim 1); claims 2–10 all depend, directly or indirectly, from claim 1.

Claim 1 — the core system. A queue-sequence system for multiple service points, made of four cooperating parts:

1. Turn-number device with a selection unit — lets each customer pick a desired service point when taking a number.
2. Turn-number allocating unit — gives every waiting customer a turn-number.
3. Terminals at the service points — each terminal signals to the system that its service point is now free (and, per dependent claim 4, whether it is closed).
4. Information unit plus computer — the computer stores the ordered list of allocated turn-numbers together with any service-point selections, receives the "free" signals from the terminals, and decides which number to call next and to which counter.

The decision rule is the heart of the claim: the number called next is the next number in the stored sequence that either (a) has no service point selected, or (b) selected precisely the service point that is now free. In other words, a customer who asked for Counter 2 will not be called to Counter 3 — the system skips over that customer and calls the next uncommitted number instead, holding the selective customer in the queue until Counter 2 opens. The computer then feeds the number and the counter identification to the information unit for display.

Claim 1 is thus a "common queue with selective affinity and pass-over" claim — the distinguishing feature over a plain take-a-number system is the combination of a single shared numerical sequence with per-customer counter preference and the skip-and-hold rule at call time.

Key dependent claims (context, not independent):

• Cl. 2–3: implementation detail — a read/write memory, ROM, CPU and I/O unit handling number allocation and feed-out.
• Cl. 4–5: cashier terminal has manual "free" and "closed" controls, and a display of the number being served; the ticket device displays a number lower than the next-to-be-issued number.
• Cl. 6: selection push-buttons on the ticket device that light to show which counters are free.
• Cl. 7–9: the electro-mechanical ticket dispenser — a numbered slip is automatically advanced into grabbing position only when the previous slip is removed; slips are a perforated tape with a gap between slips, monitored by photocells (units 35, 36) feeding position data back to the feed unit, with an alarm on abnormal delay.
• Cl. 10: the serial data-transmission scheme — 16 clock pulses plus a dead period, data clocked into a 16-bit shift register, dead-period detection generating a strobe to the BCD seven-segment drivers.

Prior art cited on the face of the patent

• US 3,803,578 (Vanway, 1974-04-09) — dispatching system with visual and audible signals.
• US 3,893,094 (Bank of America, 1975-07-01) — customer directing system.
• US 4,398,257 (NCR Corp., filed 1981-02-27, issued 1983-08-09) — customer queue control method and system. This is the most closely related reference and was filed only months before the Swedish priority date.
• Non-patent: Swedish brochure "Bela Elektroniska Kösystem" and a Swedish newspaper article ("Ny lokalkontorsidé," by Erling Borgshammar).
• Family member US 3,641,553 (Lau, 1972) — registering and calling system for waiting numbers.

Litigation / docket status (what I could and could not verify)

• I found no Federal Circuit appeal, IPR, or district-court docket naming US 4,675,647 in my searches. I cannot rule out unreported or very old proceedings, and I did not obtain a live PACER/PatentCenter docket.
• The Google Patents record for this family does carry a "Family has litigation" banner pointing to the Darts-IP global litigation dataset (family ID 20344278). That banner indicates some litigation exists somewhere in the worldwide family, but it does not identify parties, forum, or dates, and my searches did not surface the underlying case. Treat the existence of family litigation as unconfirmed in detail.
• No CAFC 2026 docket activity for this patent was found. Given the patent expired in 2004 and the 20-year enforcement window has long closed, an active 2026 appeal on this number would be highly unusual; if you need a definitive negative, the authoritative check is a direct PACER/CM-ECF query on the patent number and the Court's own docket search, which I could not run here.

Bottom line

US 4,675,647 is an early-1980s Swedish-origin invention, individually owned, covering a bank/post-office take-a-number system that couples one shared turn-number queue with customer-selected counters and a "pass-over and hold" calling rule, plus fairly specific mechanical ticket-dispensing and serial-display embodiments. Its teaching is the direct ancestor of modern "virtual queue / single-queue-multiple-server with service-type preference" systems, and it is widely cited by later queue-management patents (e.g., Addax US 5,006,983; Q-Nomy; Lo-Q/Disney flow-management families).

I'll research this patent's litigation history across multiple sources.

Let me search additional litigation-specific sources.

Let me check a few more specialized sources.

Bottom line

No litigation involving U.S. Patent No. 4,675,647 is known. I found no district court case, no Federal Circuit appeal, no ITC Section 337 investigation, no PTAB proceeding (IPR/PGR/CBM), and no reexamination or reissue proceeding naming the '647 patent as an asserted or challenged patent.

Case table as requested:

What I searched

• Google Patents record for US4675647A — the "Litigation" field there is populated only by a third‑party Darts‑ip family link (see below), not by any case caption.
• CourtListener / RECAP dockets and opinions (which mirror PACER) for the patent number and the number without commas ("4675647", "4,675,647").
• Unified Patents litigation/insights listings and portal references.
• Federal Circuit slip opinions.
• Company/party-name searches on the inventors and the likely assignee lineage (Rune Salin, Per Martin Pettersson; the Swedish family member SE425442 and the later "Turn O Matic AB" assignee that appears in the family cites list for SE465099B).

None of these returned a case involving this patent. The last recorded legal event on the US file is a maintenance-fee payment for year 12 (recorded 1998‑12‑15), and Google Patents records an anticipated expiration of 2004‑06‑23 (17 years from the 1987‑06‑23 grant) with a current status of "Expired ‑ Lifetime." Because the term has been expired for over two decades, there is no live infringement exposure and no realistic prospect of new US litigation.

The one flag worth noting (and its limitation)

The Google Patents page carries a banner: "Family has litigation — First worldwide family litigation filed," hyperlinked to a Darts‑ip record for family ID 20344278. That is the only litigation-related signal on the record, and importantly:

• It is an unverified, third‑party metadata flag attached to the worldwide family (SE8104445 / SE425442B, WO8300399A1, EP0086199B1, DK156855C, NO161527C, JPS58501149A, and the US member). It does not identify a case name, court, case number, or date, and it does not specify which family member was litigated.
• Darts‑ip's underlying docket data is behind a subscription paywall and is not accessible to me, so I could not expand this flag into a caption. The Google Patents page itself renders no case caption for the US patent, and the "Family" view on that page similarly lists no case.
• I therefore cannot confirm whether this flag refers to a US action, a European opposition/appeal, a Nordic proceeding, or an unopposed administrative filing. I am flagging it rather than asserting it as a known case involving US 4,675,647. Treating this banner as a confirmed US infringement suit would be unsupported.

False positives I ruled out

Because the number 4675647 is short and unpunctuated, it collides with a large number of unrelated identifiers. For the record, none of these are the patent:

• JP Patent No. 4675647 — a Japanese granted patent for a "stator, rotating-machine core and rotating machine" (app. 2005‑059074, filed 2005‑03‑03) — a completely different right.
• Florida FEI number 26‑4675647 (KAM DINER, LLC) and California corporation ID 4675647 (Roofing Dynamics Inc.).
• Various numeric coordinates (e.g., a human chr19 tRNA coordinate "4675719‑4675647" in US20200010832A1).
• Cases involving other '647 patents — e.g., the Apple v. Samsung " '647 patent," and Philips v. Acer "RE'564" — these are different patents and were excluded.

Caveats and confidence

• High confidence that there is no currently pending litigation: the patent expired in 2004 and nothing surfaced in any litigation database I searched.
• Moderate confidence that there was never any US litigation. Private settlement/licensing disputes, state-court actions, foreign proceedings, and pre‑1990s cases that were never digitized may not appear in free databases. A definitive negative would require a paid Darts‑ip or Lex Machina query on the family ID above and a party-name search on the assignee of record.
• I am explicitly not asserting that the Darts‑ip family flag corresponds to a US case; I could not verify its contents.

If you need this nailed down, the two concrete next steps are: (1) a Darts‑ip/Clarivate search on family ID 20344278 to expand the "family has litigation" flag into a caption, and (2) a PACER party-name search on "Salin," "Pettersson," and "Turn O Matic" across the 1987–2004 window.

I'll verify the structured "no proceedings" signal against public sources before writing the analysis.

Proceedings overview

Zero AIA trial proceedings are on file for US 4,675,647. The USPTO Open Data Portal returns no IPR, PGR, or CBM for this patent, and my independent verification searches (Google Patents family/prosecution view, PTAB docket-oriented queries, CourtListener-oriented queries) surfaced no IPR/CBM number, no Final Written Decision, no institution decision, and no Federal Circuit appeal naming the '647 patent. Bottom line for a defendant: "not hardened — untested." There is no IPR record to tell you claim 1 is dead, and equally no patent-owner win to tell you claim 1 is strong. The patent has never been tested at the Board, and, for reasons below, it now almost certainly never will be.

No proceedings to report

There is nothing to analyze claim-by-claim, so I am not going to manufacture a proceeding entry. What I can state precisely:

• Structured source (canonical): "The USPTO ODP API returns no AIA trial proceedings for this patent as of the most recent ingest."
• Independent corroboration: no Board paper, docket number, or decision for US 4,675,647 appeared in public search. See US4675647A on Google Patents (no PTAB trial, no Federal Circuit documents in the family record) and the Board's public decision repository at PTAB Decisions / E2E for claim-level documents (none exist).
• Unverified flag, not evidence: the Google Patents page carries an automated Darts-IP "family has litigation" link (https://patents.darts-ip.com/?family=20344278). That is boilerplate generated when a litigation dataset entry attaches to any family member — here likely the Swedish parent SE425442B, EP0086199B1, or a historical 1980s–90s dispute — and is not proof of a US proceeding against the '647 patent. I could not resolve it, and no US district court opinion, verdict, or settlement surfaced. Treat it as an open question, not a fact.

Strategic summary

Claim status: nothing canceled, nothing sustained, nothing adjudicated. All ten claims are UNTESTED. Claim 1 is the only independent claim; claims 2–10 all depend from it (2→1, 3→2, 4→1, 5→4, 6→1, 7→1, 8→7, 9→8, 10→7/8/9). There is therefore no narrowing, no certificate of cancellation, no surviving-claim list, and no claim construction from the Board to work with. If a demand letter or complaint asserts "claims 1–10" or just "claim 1," that assertion is backed by an entirely clean PTAB record.

Estoppel landscape: § 315(e)(2) is a non-issue — there is no triggering proceeding. No petitioner exists, so no party is estopped from anything. Every ground that could have been raised remains available: the three references of record cited by the examiner — US 3,643,553 (Lau), US 3,803,578 (Vanway), and US 4,398,257 (NCR, "Customer queue control method and system") — plus the four non-patent citations (the "Bela Elektroniska Kosystem" Swedish brochure and the Borgshammar newspaper article) and anything else printed and published before 1981-07-20. Practically, though, the reason to litigate validity is thin: the patent is expired (status: Expired – Lifetime; anticipated expiration 2004-06-23, all three maintenance fees paid — 4-year, 8-year, and 12-year windows), so an IPR petition today would be a fight over an expired right with no injunctive upside and a § 286 six-year damages lookback that cannot reach any pre-expiration conduct in a suit filed now.

Pattern signals: none, because there is no pattern. No repeat petitioner, no serial petitioning, no patent-owner appeal history at the Federal Circuit, no defensive aggregator (Unified Patents, RPX, AST) in the chain. The patent is held by an individual (inventors Rune Salin and Per Martin Pettersson; original and current assignee listed as "Individual"), not an assertion vehicle, which is consistent with a portfolio that was never used offensively. The transitional CBM program (available 2012-09-16 through 2020-09-16) would have been the natural venue for a banking/queue-management patent of this type, and no CBM was ever filed — the window closed with no takers. IPR remains theoretically available for a pre-AIA patent, but with the patent expired since 2004 and no live infringement exposure, the economics do not support it.

Recommended next steps

• If you are a defendant: your strongest point is not invalidity — it is expiration and standing to recover. US 4,675,647 expired 2004-06-23. Under 35 U.S.C. § 286, a complaint filed today (2026) can only reach acts within the six years preceding filing, i.e., roughly 2020-09-27 onward — a period in which the patent was already expired and could not be infringed. Confirm the expiration date and the paid maintenance-fee record and move to dismiss for failure to state a claim, rather than spending on an IPR.
• If the demand letter asserts specific claims: there is no FWD to point to; instead, demand the plaintiff identify (a) the asserted claims, (b) the accused acts, and (c) the dates of those acts. Absent post-2004 acts, the claim chart is unreachable as a matter of damages law.
• Verify before relying: I found no PTAB or Federal Circuit activity, but I could not resolve the Darts-IP family-litigation flag. Before filing anything, run a docket check on the family (US 4,675,647; EP0086199B1; SE425442B; DK156855C; NO161527C; JPS58501149A / JPS6341105B2; WO1983000399A1) on CourtListener and PTAB E2E, and confirm no sealed or older US suit exists against a licensee or competitor that could affect your posture.
• No active proceedings means no trial milestones to track. No institution-decision deadline, no oral hearing, no statutory one-year FWD date. The absence of any IPR is itself the signal here: a patent that is never asserted and expired in 2004 never attracts a petitioner, and this one has had forty-plus years to do so without a single filing.

Confidence statement: I am highly confident there are no AIA trial proceedings on this patent — the ODP structured data is authoritative and my searches found nothing contradicting it. I am not confident about the Darts-IP family-litigation flag's underlying record, and I have not identified any proceeding number, panel, ground, or disposition because none exists to identify. Any figure I were to give for claims canceled or sustained would be fabricated; there are none.

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

US 4,675,647 — Assignment / Ownership Reconstruction

Verification note: I could not open the USPTO Assignment Center UI directly in this session, so the finding below rests on (a) the full authoritative patent text and legal-events record supplied for this patent, and (b) targeted web searches against Google Patents and open web indexes. The Assignment Center entry point for independent checking is https://assignmentcenter.uspto.gov/ (search patent number 4675647), mirrored at https://assignment.uspto.gov/patent/index.html.

Inventors

Both are named on the US application (US 06/476,876, filed 1982-07-14) which claims priority to Swedish application SE8104445, filed 1981-07-20 (granted as SE425442B, 1982-09-27). The inventors are evidently Sweden-based; the family was prosecuted internationally (WO1983000399A1, EP0086199B1, JP S58501149A, DK156855C, NO161527C).

Unusual-pattern check: The classic "all inventors left the original assignee within 12 months of filing" precursor is not applicable here — there was no original corporate assignee to leave, because the invention was never assigned to an employer of record. That is itself the notable fact: this is a true small-inventor filing, not a corporate-originated asset. I found no evidence tying either inventor to a named employer in the bank/post-office equipment field; I am not linking the later, similarly-titled Swedish ticket-dispenser records (e.g., SE465099B, filed 1989) to these inventors, because no source in my search corroborates that connection.

Original assignee

• Entity on the issued patent: Individual (i.e., the named inventors personally — Salin and Pettersson). Google Patents lists both "Original Assignee: Individual" and "Current Assignee: Individual."
• Product embodying the claims: The specification describes a complete, buildable turn-number/ticket-dispensing and service-point-calling system for banks and post offices (ticket tape 22, feed unit 4, photocell gap monitoring 35/36, cashier terminals 3, central display 2). Whether the inventors or a licensee actually shipped it is not established by the record; treat commercial embodiment as unverified.
• Primary line of business: N/A — individual inventors, not an operating company.
• Current status: No transfer of record; the patent's maintenance fees were paid (small-entity payor noted at 4, 8 and 12-year intervals) and it expired 2004-06-23 ("Expired – Lifetime").

Assignment timeline

The USPTO Assignment Center shows no records for US 4,675,647 — there is no recorded assignment to reconstruct.

Specifically:

• Google Patents' legal-events table for this patent contains only STCF (patent grant, 1987-04-23), FEPP (fee-payment procedure, 1990-12-01 and 1994-12-01) and FPAY (maintenance fee payments, 4th/8th/12th year). There is no AS/assignment event of any kind.
• There is no assignee reel/frame citation anywhere in the prosecution, issue, or post-issue record; the "Current Assignee" field resolves to Individual, not to a corporation, LLC, or trust.
• No Security Agreement, Merger, Change of Name, License, Release, or Correction record appears in the chain.

Finding: The chain of title is a single link — inventors → themselves. Rights were never recorded as transferred to any entity, and nothing in the record suggests a post-issuance conveyance. The original owners (the individual inventors) therefore remained the owners of record through expiry.

Why I am stopping here: the task instruction is to state plainly when the Assignment Center has no records and to stop after this section. Accordingly, I am not populating the Timeline diagram, NPE / troll-pattern signals, or Verdict sections, because there is no assignment chain on which to base them — any NPE signal table or verdict here would be inference dressed up as evidence.

Supplementary observation (not an assignment finding): the Google Patents family page carries a Darts-IP "Family has litigation / first worldwide family litigation filed" link for family ID 20344278. That is a third-party litigation-index reference, not a USPTO assignment record and not a court document I could corroborate; I flag it only as unclear and uncorroborated.

Verification link: https://assignmentcenter.uspto.gov/ (patent number search: 4675647) · Google Patents: https://patents.google.com/patent/US4675647/en

I'll research the specific prior-art references cited on the face of US 4,675,647, prioritizing specifics over my training data.

Now the NCR reference and the family citation:

Let me pin down the Lau reference and the non-patent literature:

Prior-Art Analysis — US 4,675,647

0. Verification of the target document

Before the prior-art analysis, confirmation of the literal identifier, since a wrong number would invalidate everything downstream:

This matches the record analyzed in the previously generated sections, so no contradiction to flag. As noted there, the only U.S. patent bearing the number 4675647 is this one; other uses of the string "4675647" (e.g., JP 4675647, corporate IDs) are unrelated and are not the subject here.

Search limitation, stated up front: I could retrieve the full text/claims of US 3,803,578, US 3,893,094 and US 4,398,257 (and substantial summary material on the latter's claims), but I could not retrieve the full specification of US 3,641,553. My remarks on that reference are therefore confined to its biblio-graphic data and title, and I flag where I am not able to characterize its disclosure.

1. The references of record on US 4,675,647

Per the authoritative front-page record, the patent carries three examiner-cited U.S. patent references, plus two non-patent publications, plus one family citation. These are distinct categories and I keep them separate (this is where a "look at each patent citation" instruction can mislead if the categories are merged):

Category A — U.S. patent references cited by the examiner (3):

1. US 3,803,578 — Vanway E
2. US 3,893,094 — Bank of America National Trust
3. US 4,398,257 — NCR Corporation

Category B — Non-patent publications cited (2):
4. "Bela Elektroniska Kösystem" (Swedish brochure)
5. "Ny lokalkontorsidé," by Erling Borgshammar (Swedish newspaper article)

Category C — family citation (1), not a face-of-patent citation:
6. US 3,641,553 — Dicksen T. W. Lau (listed under "Family Cites Families," i.e., cited in a family member's prosecution, not on the US front page)

2. Claim 1 shaded into elements (the § 102 yardstick)

Anticipation under § 102 requires every element of a claim in a single reference. Claim 1's independent combination can be decomposed into:

• (a) turn-number device with a selection unit — customer picks a desired service point;
• (b) a turn-number allocating unit — every customer gets a number in one common sequence;
• (c) terminals at the service points, each signalling that its point is free;
• (d) an information unit (display of number + which counter);
• (e) computer means memorizing the sequence of allocated numbers together with the selected service points and receiving the "free" signals;
• (f) computer means deciding which number goes to which free point and feeding the number + point ID to the information unit; and
• (g) the decision rule: the number called is the next in the stored sequence that either has no selection or selected the very point that is now free (the "pass-over and hold" rule).

The novelty of the patent lives almost entirely in (a) + (e) + (g) operating together. That is the axis against which each reference must be measured.

3. Reference-by-reference analysis

3.1 US 3,803,578 — Vanway (LaVanway), "Apparatus for use with a dispatching system having visual and audible signals"

• Full citation: U.S. Patent No. 3,803,578, filed 1970-06-05, issued 1974-04-09.
• Status as art: Issued more than one year before the 1981-07-20 priority date → available as § 102(b) art (U.S. patent).
• Brief description (from the retrieved text): A master unit and up to eight remote units, one per teller station. Each teller depresses a plunger switch (74) to signal that the station is free; this lights an indicator at the master unit (numbered row 28), lights a lamp at the remote unit visible to the waiting line, flashes the light via a flashing motor, gives an audible/visual dispatch indication, and increments a mechanical counter (83) recording the number of customers serviced. A step-down transformer isolates the remote units. Essentially a teller-availability annunciator — no customer-carried queue number at all.

§ 102 mapping to claim 1: Discloses elements (c) (per-station terminals signalling "free") and (d)-type output (visual/audible indication, plus a station-identifying numeral). It does not disclose (a) a customer selection unit, (b) a turn-number allocating unit or a common queue sequence, (e) memorized turn-numbers with selections, (f) number-vs-counter decision logic, or (g) the pass-over rule. Conclusion: does not anticipate claim 1 or any dependent claim, because the entire turn-number/selection core is absent. Its real relevance is as background showing that teller-activated "free" signalling with visual + audible output was old.

3.2 US 3,893,094 — Bank of America National Trust & Savings Ass'n, "Customer directing system"

• Full citation: U.S. Patent No. 3,893,094, filed 1973-10-29, issued 1975-07-01.
• Status as art: More than one year before priority → § 102(b) art.
• Brief description (from the retrieved text): Addresses the complaint that manual teller "available" switches suffer from tellers forgetting to flip them. Instead, ultrasonic transducers (27: transmit 28 / receive 29) at each teller region detect the absence of a customer, with time-delay logic to ignore casual pedestrian traffic, and activate a flashing "Teller Available" sign (18) with direction arrows (19, 21) pointing the customer to the free window. Automatic, sensor-based free-station detection and customer direction.

§ 102 mapping to claim 1: Discloses detection of a free service region (c)-like and a display directing a customer to it (d)-like, but by a fundamentally different mechanism (ultrasonic absence detection, not a numbered queue). It lacks every element of the claimed turn-number system — (a), (b), (e), (f), (g). Conclusion: no anticipation of claim 1 or its dependents. It is relevant only as evidence that automated "which station is free → direct the customer there" was known.

3.3 US 4,398,257 — NCR Corporation, "Customer queue control method and system" — the closest reference

• Full citation: U.S. Patent No. 4,398,257, filed 1981-02-27, issued 1983-08-09. (EP counterpart: EP 0 072 850 B1.)
• Status as art — important date point: Its issue date (1983-08-09) is after the '647 priority date (1981-07-20), so it is not § 102(b) art. However, its filing date (1981-02-27) precedes the '647 priority date, so it is a candidate pre-AIA § 102(e) reference (a U.S. patent granted on an application filed before the applicant's date). Whether it actually qualifies turns on the '647 inventors' date of invention, which is not on the face of the patent; I flag this rather than assume it.
• Brief description (from the retrieved text and claims): A main queue plus local queues (max two customers each) at each service station. Each station has an "open" key (52) and a "close" key (54) → a switch detector/encoder reports station number and open/closed status to a main processor with memory. The processor maintains counts per local queue and a per-station delay period related to median service time, and directs the customer at the head of the main queue to the station expected to give the shortest waiting time, optionally by synthesized voice. Its claim 1 recites, in substance: first means to detect a customer at the head of the main queue; second means at each station to signify open/closed; counting means for local-queue counts; timing means; and customer direction means … to direct a customer at the head of the main queue to a local queue … in accordance with a determination as to which local queue is expected to provide the shortest waiting time.

§ 102 mapping to claim 1:

Conclusion: US 4,398,257 does not anticipate claim 1, because it fails the two elements that define the invention — (a) customer selection of a service point and (g) the skip/hold rule keyed to that selection. It is, however, the most pertinent reference overall: the single-queue-multiple-server architecture, per-station open/closed terminals, computer memory, and number/station display are all there. Expect it to be the reference an examiner would pair with something else under § 103, or combine with a plain take-a-number system, to attack claim 1 as obvious. Its filing date (1981-02-27) being only ~5 months before the Swedish priority date makes it a genuine contemporaneous development, and the fact the examiner cited it suggests it was the principal obstacle during prosecution.

3.4 US 3,641,553 — Dicksen T. W. Lau, "Registering and calling system for waiting numbers"

• Full citation (from the '647 record): U.S. Patent No. 3,641,553, filed 1970-05-04, issued 1972-02-08.
• Status as art: Issued ~9 years before priority → § 102(b) art if it discloses the elements. Note it appears in the record as a family citation, not on the US front page as an examiner citation — so the "cited by examiner" asterisk does not attach to it.
• Honesty flag: I was unable to retrieve the full text or claims of this patent, so I cannot map its disclosure element-by-element. Based on the title alone — "Registering and calling system for waiting numbers" — it plausibly discloses a numbered-queue registering-and-calling system (i.e., elements (b), (d), and a display of the next number). If so, it would be relevant background for the turn-number aspect but, like the others, would lack the customer-selection and pass-over features. I am not asserting its content beyond the title; a claim-chart would require pulling the document.

3.5 Non-patent literature of record (Category B)

• "Bela Elektroniska Kösystem" — Swedish brochure.
• "Ny lokalkontorsidé," by Erling Borgshammar — Swedish newspaper article.

Status: these are printed publications and, if dated more than one year before 1981-07-20, would be § 102(b) art; if earlier than the invention date but within the grace period, § 102(a) art. I could not retrieve their contents or exact publication dates, and the record as reproduced does not give dates. I therefore cannot assess them substantively and will not fabricate a disclosure. They matter methodologically because their presence on the front page implies the examiner considered a body of prior Nordic electronic queue-system practice — consistent with the inventors' Swedish origin.

4. Bottom line

No single reference of record anticipates claim 1. Every one of the three examiner-cited U.S. patents is missing at least the customer-selection element and the pass-over/hold rule. The novelty and the likely non-obviousness of the '647 patent rest on that specific combination — a common turn-number queue in which a customer may bind himself to a chosen counter, and the system then skips him rather than reassigning him — which is precisely what US 4,398,257's shortest-waiting-time dispatcher does not do (it deliberately reassigns the head-of-queue customer to whichever station is expected to free up soonest). US 4,398,257 is nonetheless the reference to center any validity or invalidity analysis on, because it supplies the entire architectural remainder of claim 1.

What would change this assessment

1. Full text of US 3,641,553 — the only cited reference whose disclosure I could not verify; if it teaches or suggests any customer-selectable calling (unlikely for a 1970 take-a-number system, but unconfirmed), it would move.
2. Actual publication dates/contents of the two Swedish items.
3. The inventors' date of invention, which determines whether § 102(e) even reaches US 4,398,257.

I have deliberately not converted any of the "candidate" statutory bases into definite findings, and I have not asserted disclose content for the two non-patent items or US 3,641,553, because the documents were not retrievable in this session.

I'll ground this analysis in the actual disclosures of the references listed on this patent's face before constructing the combinations.

Good — NCR and Bank of America are well characterized now. Let me pull the other two references on the face/family of this patent.

Let me pin down the Lau reference and check the state of the ticket-dispenser art, since claims 7–10 are mechanical/electronic details.

Obviousness Analysis — U.S. Patent No. 4,675,647 under 35 U.S.C. § 103

Scope note on inputs

This analysis uses only the prior-art references identified on the face of the '647 patent and in its family record, as carried forward from the prior sections:

Two flags before the merits:

• Filing-date discrepancy carried forward. The earlier section flagged the U.S. filing as 1982-07-14 (PCT/SE1982/000244) versus a family-record date of 1983-03-18. Either date falls after NCR's 1981-02-27 U.S. filing, so NCR's status as § 102(e) art is unaffected. This matters: under In re Hilmer, NCR gets no benefit of any foreign priority and is prior art as of its U.S. filing date — which still predates the earliest date the '647 applicants can claim.
• Date conflict in my instructions. The task header states April 26, 2026 while the operative "today" elsewhere is 2026-09-27. Immaterial here: the patent expired 2004-06-23, so no date sensitivity exists in this § 103 analysis.
• I could not retrieve the full text of Lau (US 3,641,553). Only its title, dates, and its position as a family-cited reference are verified. I rely on it as a "sequential waiting-number registering and calling" teaching and do not attribute specific structure to it. Statements below about Vanway, BofA, and NCR are grounded in retrieved text (URLs cited).

1. Legal framework and level of ordinary skill

Pre-AIA § 103 governs (application filed 1982). The inquiry is the Graham v. John Deere, 383 U.S. 1 (1966) four-factor test, applied as refined by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007): the question is "whether the improvement is more than the predictable use of prior art elements according to their established functions," and a claim is obvious where the combination is "a design choice" or "a predictable variation," even absent an explicit teaching, provided there is an articulated reason with rational underpinning (In re Kahn; MPEP 2141–2143).

PHOSITA (≈1981): a person with a B.S. in electrical engineering or computer science, or equivalent, plus 2–4 years' experience designing microprocessor/minicomputer-based control systems — specifically conversant with (a) queue/dispatch logic in single-queue-multiple-server service environments, (b) electromechanical and electronic ticket dispensers, (c) serial peripheral buses, shift registers, and seven-segment display drivers, and (d) teller/cashier signaling terminals. This is a low-to-moderate skill level in a highly practical, integration-oriented art, which cuts toward obviousness of software-logic features implemented on existing hardware.

2. What the references actually disclose (scope and content)

NCR '257 — the closest reference, and it is very close. Retrieved text (Google Patents; EP 0072850 B1) shows:

• A main processor 20 with memory 22, address/data buses, and an available-station table ("NATT") in read-write memory — i.e., a stored register of free stations.
• A plurality of customer service stations 40–50, each with two keys: an "open" key 52 ("the station is open for business, or that a transaction has just been completed") and a "close" key 54 ("the station is closed, and is therefore not available to receive customers").
• A switch detector and encoder 60 that outputs the identifying number of the specific station plus the open/close status to the processor — which is literally the claim-1 element of "terminals… each providing a signal identifying a particular service point which is free."
• A dispatch program whose literal comments include "REM CHECK IF CUSTOMER IS PRESENT AT HEAD OF MAIN QUEUE"; "REM FIND IF A STATION IS AVAILABLE"; "REM GET NEXT AVAILABLE TELLER"; "REM REMOVE TELLER JJ FORM TABLE NATT" (Justia).
• An annunciation subsystem: synthesized voice directing the customer to the selected station; the patent states that "voice messages thus replace lighted directing devices and flashing beacons" — an express admission that lighted directing devices and flashing beacons for directing customers to free stations were known prior art.

The critical limitation in NCR: its stations are described as ones "all available to customers and performing essentially the same service functions." NCR has no per-customer station selection and no memory of a customer preference; dispatch is purely best-available-station.

Vanway '578 (US3803578) discloses a master unit at the waiting area with a row of numerals 28, one per operating station, each backed by a light 62 that flashes when energized, plus audible signaling; and remote units at each station with a teller-operated push-button switch 74 (actuating portion 36) that signals "operator at that station is free to service the next customer" and drives both the master display and a per-station transaction counter, with a 15-second time delay. This is the claim-1 "terminal at the service point signaling that its point is free" in essentially anticipatory detail.

BofA '094 (US3893094) detects the absence of a customer at a teller window and activates a flashing arrow / "Teller Available" sign to direct customers to that free service region. Importantly for claim 4, its background states: "Heretofore, in previous systems wherein it has been desired to provide an indication of 'availability' of a teller at a particular window… systems have depended upon the manual operation by the teller of a switch located at the teller station." That is an express prior-art admission of teller-manual availability switches.

Lau '553 — a "registering and calling system for waiting numbers," i.e., the take-a-number paradigm: sequential numbers allotted to waiting customers and called in order.

3. Claim 1 — the combination case

Claim 1 is a "common queue + selective affinity + skip-and-hold" claim. Mapping it against NCR alone shows the only unmet limitations:

Combination A (primary): Lau '553 + Vanway '578 + BofA '094 + NCR '257

The differences reduce to two data-handling features: (i) accept and store a counter preference with the turn-number, and (ii) at dispatch, skip numbers whose preference ≠ the free counter.

Motivation to combine, grounded in the references themselves:

1. Specialized counters make station selection a functional necessity, not a design whim. NCR defines its own problem space narrowly — stations "performing essentially the same service functions." Every bank/post-office PHOSITA knew that real branches have non-fungible windows (foreign exchange, safe deposit, business/teller vs. personal, loans). NCR's own premise therefore identifies the unmet need in the art: extending a common-queue dispatcher to heterogeneous counters requires the customer to be routed only to a counter that can serve him.
2. Directing customers to a specific station and displaying which station is free were both old (Vanway's numeral-per-station master display and flashing lamps; BofA's flashing "Teller Available" arrows; NCR's admission that "lighted directing devices and flashing beacons" predate it). Combining a sequential-number allocator (Lau) with a per-station free indicator (Vanway) is the classic, predictable marriage of two known sub-systems serving the same purpose.
3. The dispatch logic is table-driven code on a general-purpose processor. NCR literally implements "GET NEXT AVAILABLE TELLER" by reading and deleting an entry from a memory table. Adding a one-field preference to each queue record and a single comparison (if pref = 0 or pref = free_station) is the kind of routine programming modification that KSR and Perfect Web Techs. v. InfoUSA treat as within ordinary skill, not an inventive leap. There is no new hardware in claim 1; it is a change in stored data and a branch condition.
4. The "skip-and-hold" rule is the logically compelled consequence of honoring a preference. Once a system stores "customer wants Counter 2," and Counter 2 is busy, the system has exactly two options: call him elsewhere (defeating the preference) or hold him. Claim 1 recites the second, in the alternative with calling the next uncommitted number. Reciting the necessary implementation of a stored preference is not an inventive contribution — it is the "predictable use of prior art elements according to their established functions."
5. Fairness/FIFO is preserved, which is the express purpose of the common-sequence systems of Lau/NCR — so there is no teaching-away; the claim's scheme improves throughput and match-rate simultaneously, both recognized desiderata (NCR: minimizing waiting time; BofA: "more efficiently handle and distribute the flow of incoming business").

Reasonable expectation of success: high. The modification is a data-flag and comparison on hardware NCR already discloses, with no unproven physics, no new transducer, and a fully deterministic result.

Conclusion on claim 1: Obvious under § 103 over Lau + Vanway (or NCR) in view of BofA, and a fortiori over NCR + Lau + Vanway, since NCR supplies the "terminals signal a free, identified station" element, the memory-table dispatch, and the admission that flashing/lighted station-directing devices were old.

4. Dependent claims 2–10

5. Objective indicia (§ 103 secondary considerations)

• No nexus evidence. Any commercial success of the "Bela" system (the cited non-patent literature) would need to be tied to the claimed selective-affinity feature, not to the general convenience of take-a-number systems, which was long known.
• Widespread later adoption (Addax US 5,006,983; Q-Nomy; Lo-Q/Disney flow-management families, per the citing-art list) is double-edged and, absent a nexus and a "failure of others," supports that the technique was the natural next step rather than that it was non-obvious.
• The non-patent literature is the most underused asset here. The "Bela Elektroniska Kösystem" brochure and the Borgshammar newspaper article are cited as prior art of record. If either is dated before 1981-07-20 (or before 1981-07-20 minus one year for § 102(b)), each is a printed publication that could independently show a prior public system, potentially as an anticipatory or near-anticipatory reference against claims 1 and 4. I could not verify their dates from the sources available to me — this is the single most valuable lead for anyone pursuing the invalidity case, per the earlier section's note that the Darts-IP family flag remains unverified.

6. Bottom line

The strongest single-reference attack on the claims that matter commercially (claims 1 and 4) is NCR '257, with Vanway '578 as the alternative source for station terminals/master display and BofA '094 supplying both the free-region-directing concept and the admission that manual teller availability switches and flashing directing devices were already known. The dependent mechanical/transmission claims (7–10) are the residual validity risk for any challenger working only from the face-of-patent citations.