US 11402427

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

US Patent 11,402,427 — Analyst Summary

Verification note: I searched for this exact number and found no litigation, PTAB proceeding, or CAFC 2026 docket entries naming U.S. Patent No. 11,402,427. The CAFC 2026 results my search surfaced referenced IPR2023-00427 (Topia Technology v. Box/Dropbox) — a different identifier that coincidentally shares the digits "0427" — so I have not attributed that case to this patent. The only authoritative record for US 11,402,427 is the patent document itself (Google Patents, incl. USPTO bibliographic data), which is the source used below.

Bibliographic Data

Primary classifications: G01R 31/2846 (fault-finding using knowledge-based systems / AI), G06Q 10/0635 (enterprise risk analysis), G06Q 10/0637, G06Q 50/04 (manufacturing), G06F 30/20.

Abstract (as issued)

According to one embodiment, an information processing system includes a memory, an inputter, and a processor. The memory stores correlation information including a specification item group, an element item group, and a risk item group. The specification item group includes specification items. The element item group includes element items. The risk item group includes risk items. One of the specification items to be modified is selected by a first operation input to the inputter. One of the element items is selected by a second operation. When the first operation is input, the processor refers to the correlation information and extracts at least one of the element items having a correlation with the one of the specification items. The processor extracts at least one of the risk items having a correlation with one element item of the at least one of the element items selected.

Plain-Language Overview of the Independent Claim

Claim 1 is the sole independent claim (presented as "A method to modify a current design of a product by determining risk associated with modifying a component part or a manufacturing process of the product to satisfy a new performance or reliability specification"). In plain terms it covers a computer-implemented design-risk workflow built on a three-layer data structure:

1. Store a correlation data structure in memory, with nodes in three groups:

   • a specification item group (things like heat dissipation, electrical characteristic, temperature-cycle test — each tied to product performance or reliability);
   • an element item group (manufacturing elements — component parts or manufacturing processes);
   • a risk item group (failure/risk items).
     The structure also stores connection information: first links from specification items → risk items, and second links from risk items → element items.

2. Forward traversal (specification → element): given a selected specification item to be modified, the processor walks the graph in a first direction through the risk-item layer, over all possible paths, to determine which element items are correlated with that specification.

3. User selects one element item, then the processor reverse-traverses (element → risk) in a second, different direction over all possible paths using the second connection information, extracting the risk item(s) correlated with that element item.

4. Pull and output "risk information" explaining why that risk item was surfaced — mechanism information (the causal mechanism linking the risk to the specification or manufacturing element) and/or example information (past instances of the risk occurring). The output presents both the risk item(s) and this explanatory risk information, i.e., the risk you take on by modifying that component/process to hit the new spec.

5. Quantify the risk: determine an evaluation value from evaluating the risk item against the new performance/reliability specification, compare it to a threshold to derive a margin value, and compute a predicted risk value from that margin.

6. Act on the result: modify the current product design when the predicted risk value indicates the new performance or reliability specification is not satisfied.

Dependent claims (all narrow the above):

• Claim 2 — the element items split into component-part items and manufacturing-process items.
• Claim 3 — component-part group and manufacturing-process group each have independent tree structures.
• Claim 4 — specification group and element group each have independent tree structures.
• Claim 5 — recites in method-step form the retrieval/extension of the mechanism/example risk information.
• Claim 6 — recites the two-stage traversal explicitly as first direction then opposite direction.

Important Caveats / Uncertainty

• Claim-1 text appears to reflect post-filing amendment. The granted Claim 1 contains the "evaluation value → margin value → predicted risk value" and "modifying the current design" limitations, which correspond to the second-embodiment formula (V = ln[1/(A₁×A₂×A₃)]) in the specification rather than to the purely extract-and-output flow described for the first embodiment. The machine-rendered claim set on the public page also shows some formatting irregularities. I would treat the exact wording of Claim 1 as requiring confirmation against the USPTO PatentCenter/Patent Public Search official copy (or the issued patent PDF) before relying on it for infringement or validity analysis.
• Prosecution history is nontrivial. The file shows multiple non-final and final rejections (2020–2021), a response after final action, an advisory action, and a notice of allowance on 2022-04-25. Any estoppers or claim-scope concessions made during that back-and-forth would matter for claim interpretation but were not available in the text I reviewed.
• No litigation/PTAB/CAFC record found for this number. I cannot confirm it has never been asserted; I can only report that my searches did not surface any such proceeding. Searches for "11,402,427" returned unrelated Toshiba matters (NAND/HDD, MiiCs, Lone Star, Resonac/MRT) that do not involve this patent.
• Cited-by art (for awareness): US 2022/0159029 A1 (CyberArk) and US 2022/0358526 A1 (NTT). Notable cited prior art includes CN 1573785 A, US 2009/0292384 A1 (Hitachi, manufacturing instruction evaluation support), US 2010/0287105 A1 (Itid Consulting, information processing system/method), and CN 104166800 A (component FMEA based on failure mechanisms).

If you need claim-chart-level accuracy or a litigation check against a specific docket, the next step is to pull the official issued claims and the full file wrapper from USPTO PatentCenter (application 16/013,377).

I'll search for litigation involving US patent 11402427.

Let me search more specifically across litigation databases.

Let me search additional litigation-specific databases.

Findings: No known litigation involving US 11,402,427

I searched for litigation specifically tied to US Patent No. 11,402,427 ("Information processing system and information processing method"; Toshiba / Kabushiki Kaisha Toshiba; inventors Michinobu Inoue and Hideo Nishiuchi; granted 2022-08-02). I found no record of any lawsuit, PTAB proceeding, or other litigation asserting or challenging this patent.

Caveat on completeness: I was able to run only a limited set of searches before hitting tool limits, so I cannot claim an exhaustive negative. What I can say is that targeted litigation searches returned no case matching this patent.

Cases returned that do NOT involve US 11,402,427 (important false positives)

The searches surfaced a patent commonly referred to as "the '114 Patent" in litigation headlines, which is easy to confuse with the '427 patent number. These cases involve U.S. Patent No. 10,936,114 — a completely different patent (touch-gesture/user-interface patent, inventor Michael Smith, owned by Smith Interface Technologies, LLC), not US 11,402,427:

In those pleadings, "the '114 Patent" is expressly identified as U.S. Patent No. 10,936,114 (See Smith Interface v. Apple stay order and Apple IPR2024-01091 Exhibit 2001; e.g., the court's list of asserted patents includes "10,936,114 (the ''114 Patent')"). Per the operating rules, I am not auto-correcting or merging identifiers — these are different patents with the same trailing digits, and none of the Smith Interface/Apple/Samsung litigation involves 11,402,427.

Similarly, other search hits referencing Toshiba litigation (e.g., MiiCs & Partners America v. Toshiba, D. Del. 14-803; Litigation over Toshiba NAND/HDD patents; the '864 patent HDD case) concern unrelated Toshiba patents — not the '427 patent.

Summary of prior-art/litigation-adjacent record for US 11,402,427

• Litigation: None known.
• PTAB (IPR/PGR): None found.
• Related family members (JP6806651B2, KR102360631B1, DE102018211196A1, TWI711001B) show no associated litigation in the results reviewed.
• The Google Patents page for US11402427B2 lists only two documents "citing" it and no adverse litigation or post-grant challenges.

Recommended verification

To confirm the negative with higher confidence, you (or I, in a follow-up) should query directly:

1. Unified Patents Portal — litigation caselist: https://portal.unifiedpatents.com/litigation/caselist (search party "Toshiba" and patent "11402427")
2. USPTO PTAB / PatentCenter for the '427 patent (IPR/PGR/re-exam activity)
3. CourtListener / RECAP and PACER full-text docket search for "11,402,427"
4. Docket Alarm / RPX Insight patent-level docket summaries

Bottom line: Based on the searches performed, there is no known litigation involving US Patent 11,402,427. If a filing has occurred recently, it would appear in the sources above, but I found no such record. I am flagging explicitly that this conclusion rests on a limited number of searches, and that the " '114 Patent" hits in the results map to US 10,936,114, not to the patent you asked about.

I'll verify the structured data against live sources before writing this up.

Let me run a couple more targeted checks before concluding.

Proceedings overview

Total AIA trial proceedings on file: 0. The structured USPTO Open Data Portal (ODP) record returns no IPR, PGR, or CBM proceeding for US 11,402,427, and targeted web searches (patent number, "11,402,427" + IPR/PGR, petitioner-side keywords, and Toshiba-as-patent-owner queries) surfaced no un-indexed or recently filed petitions either — I found no proceeding to flag as an ODP lag. Breakdown by status is therefore: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0. The defensive posture this gives a defendant is unusual but straightforward: the patent is completely un-hardened and completely un-narrowed by PTAB — all six claims stand exactly as issued on 2022-08-02, and no petitioner has ever tested them. You get the full benefit of a blank slate (no § 315(e)(2) estoppel against you, every reference in the field still "available"), but you also get no free roadmap, no FWD claim constructions, and no invalidity findings to borrow. Any challenge you mount would be a first-of-its-kind IPR on this patent.

Because there are no proceedings to walk through, I have replaced the per-proceeding sections with an equivalent evidence trail and cross-check so you can verify the null result rather than take it on faith.

Verification trail (in lieu of per-proceeding entries)

No proceeding number to report — and I am not inventing one

• Canonical source checked: USPTO ODP structured "PTAB proceedings on file" block for US 11,402,427 (as supplied in this prompt). Result: empty.
• Independent web checks run (2026-09-29): searches on the patent number in combination with IPR, PGR, CBM, IPR2022/IPR2023/IPR2024/IPR2025 prefixes, and Toshiba-as-owner terms. All hits were unrelated patents (e.g., NEC's U.S. 9,769,427; the 2003-era "…427 Patent" display-scaling case out of E.D. Tex.). Nothing referenced U.S. 11,402,427 in any AIA trial context.
• Sanity check on the numbers: the "Cited By" and "Families Citing this family" entries on the Google Patents record for US 11,402,427 are prosecution-progress citations (Advantest, NTT, CyberArk, Hitachi), not PTAB papers. The absence of any "Cited By" PTAB FWD or institution decision is consistent with no trial history.

Why the absence is plausible here (not a data artifact)

This is a mechanical/manufacturing-process quality-management patent (G06Q 10/0635 risk analysis; G01R 31/xx; G06F 30/20), filed 2018-06-20, granted 2022-08-02, priority 2017-09-14. Three structural facts explain why it has attracted no AIA trial:

1. No assertion campaign. Nothing in the record, and nothing in my searches, shows this patent being asserted in district court or the ITC. IPRs are overwhelmingly filed as a defensive response to an infringement suit or demand. Absent assertion, absent petition — the "absence is itself a signal" heuristic cuts the other way here: well-asserted patents eventually attract IPRs; this one appears unasserted.
2. The PGR window shut long ago. Grant date 2022-08-02 means the 9-month PGR post-grant review window under § 321(c) closed on or about 2023-02-02. PGR is off the table (as is CBM, which the Office stopped accepting on 2020-09-16). Only IPR (§ 311) remains, and it remains available to anyone not time-barred.
3. The granted claim 1 is heavily narrowed. The file history shows a hard-fought prosecution — final rejection mailed 2021-10-27, then a return to the examiner and a notice of allowance on 2022-04-25. The issued claim 1 carries a long, machine-implemented, multi-step spine (correlation data structure with three node groups and first/second connection information; bidirectional traversal "via all possible paths"; risk-information extraction of mechanism/example data; a margin-based predicted risk value; and a final "modifying the current design … when the predicted risk value indicates that the new performance or reliability specification … is not satisfied"). Claims of this shape cut the other way for a challenger: they are broad enough to be vulnerable to a § 103 combination of business-process/FMEA art, but the "modifying the current design" and predicted-risk-value limitations give the patent owner a fairly narrow construction to retreat to. That trade-off is exactly what a first petitioner would have to litigate from scratch.

Strategic summary

Claim status of US 11,402,427 today. All claims stand as issued and are UNTESTED before the PTAB. There are no CANCELED claims and no SUSTAINED-after-trial claims — the distinction only matters here in the negative. Claims 1–6 (6 total claims; claim 1 independent, claims 2–6 dependent) each remain enforceable with their original scope. There is no certificate of cancellation, no certificate of correction from a trial, and no adverse judgment under 37 C.F.R. § 42.73(b).

Estoppel landscape: clean. Because no IPR, PGR, or CBM was ever instituted against this patent, no one is subject to 35 U.S.C. § 315(e)(2) estoppel on these claims. Practically: there are no prior-art grounds that have been "used up," no reference that is off-limits because a petitioner reasonably could have raised it before, and no IPR-driven claim construction or obviousness finding that limits how the claims can be read. If you are today's defendant, the entire universe of printed publications and patents predating 2017-09-14 is available to you — in the district court and in a fresh IPR — subject only to § 325(d) discretion if the same art was already before the examiner. That last caveat is live: the examiner was active here (applicant canceled at least one claim during prosecution, and the record shows substantive back-and-forth), so you should build your first IPR around art not of record to avoid a Becton Dickinson / § 325(d) discretionary denial.

Pattern signals: none to read. No repeat petitioner (there is no petitioner), no patent owner appellate activity (no FWD means no PTAB appeal to the Federal Circuit, so no CAFC docket under 35 U.S.C. § 319), and no defensive aggregator — Unified Patents, RPX, or similar — has ever touched this patent as far as the public record shows. The patent-owner side is Toshiba (Kabushiki Kaisha Toshiba), a large operating company rather than an NPE, which is itself relevant: large operating-company patents in the manufacturing-quality space are more often cross-licensed than asserted, and the JP family member (JP 6806651 B2) is likewise a corporate-filed sibling. Note also that the patent remains in force — a maintenance-fee payment on the 4th year was recorded 2026-01-21, and the listed adjusted expiration is 2039-09-07, so there is a long tail of remaining life. A patent with ~13 years left and zero litigation history is a candidate for future assertion; the PTAB picture could change quickly if Toshiba licenses or sells it into an assertion vehicle.

One important limit on this conclusion: my null result is based on the ODP structured record plus open-web searching. I could not query a subscription litigation database (Docket Navigator, Lexis, Lex Machina) or the full PTAB E2E / Patent Trial and Appeal Board End-to-End docket directly in this session. A petition filed very recently and not yet docketed publicly could in principle be invisible to both sources.

Recommended next steps

• No PTAB activity exists — say so plainly, and price it accordingly. There is no FWD to link to and no disposition to quote, because no trial was ever instituted. Any statement to a client or court that claims of US 11,402,427 have been "invalidated" or "upheld" by the PTAB would be false; the correct characterization is "never challenged in an AIA trial."
• Confirm the null result directly, on the record, before you rely on it. Run the patent number through:
  • USPTO PTAB E2E / PTAB Center: https://ptacts.uspto.gov/ptacts/ (search by patent number 11402427)
  • USPTO Patent Center for the application (US 16/013,377): https://patentcenter.uspto.gov/ — to rule out an ex parte reexamination or reissue filed outside the AIA-trial track
  • CourtListener / RECAP docket search for "11,402,427" to confirm the no-litigation premise, which is the best explanation for the no-IPR premise.
• If you are a defendant now receiving a demand or complaint: you have no ready-made invalidity judgment to lean on, so budget for a full first-challenge IPR. Target claim 1 (the only independent claim) with a § 103 combination grounded in (i) FMEA / failure-mode-and-effects-analysis and root-cause-analysis art in the G06Q 10/0635 space, plus (ii) design-of-experiments / Taguchi-method and reliability-evaluation references for the margin-threshold and predicted-risk-value limitations. Because claim 1 depends on a "predicted risk value," look for obviousness attacks on the formula itself (a log of the reciprocal of a product of three sub-scores is a textbook scoring construct) rather than treating the math as a black box. Refresh your § 325(d) screening first so your art is not the examiner's art.
• Watch the clock on any parallel litigation. If the patent is asserted against you, § 315(b) gives you one year from service of the complaint to file an IPR. There is no institution decision deadline running today because there is no pending petition — the statutory one-year trial clock under § 316(a)(11) only starts at institution and does not apply to a proceeding that does not exist.
• Re-run this docket check if the patent changes hands. A transfer from Toshiba to a licensing or assertion entity is the single most likely trigger for the first IPR on this patent. Keep the patent on a monitored docket list.

I'll research the assignment record and cross-reference sources before reconstructing the chain.

Assignment Record Reconstruction — US 11,402,427 B2

Patent: US 11,402,427 B2 — "Information processing system and information processing method"
Application: US 16/013,377 · filed 2018-06-20 · Issued: 2022-08-02
Priority: JP 2017-176705, filed 2017-09-14
Source of record used: Google Patents legal-events mirror of the USPTO assignment data (reel/frame surfaced), cross-checked against inventor/assignee aggregators. I could not open the Assignment Center UI directly within this session, so the correspondent-of-record field below is flagged where it was not retrievable rather than inferred.

Inventors

Unusual-pattern check: No departure pattern. The classic "all inventors exit within 12 months of filing" fire-sale precursor is absent. Toshiba inventor attribution for Inoue continues through October 2025 — roughly seven years after this filing — and both inventors' portfolios remain Toshiba-attributed. The 2017–2018 Toshiba financial distress that drove the Toshiba Memory divestiture did not dislodge these two inventors or their patent filings.

Original assignee

Kabushiki Kaisha Toshiba (Toshiba Corporation), Japan — the entity named on the issued patent and the sole assignee of record ever recorded against US 11,402,427.

• Primary line of business: diversified industrial/electronics conglomerate (energy systems, infrastructure, electronic devices & storage, digital solutions).
• Product embodiment: The claims recite an internal engineering/design workflow — storing a specification-item / element-item / risk-item correlation structure, traversing it to surface manufacturing-element risks before a design modification, and folding an evaluation-derived predicted risk value back into the design. This is a design-and-development support method, not a distinct shipped consumer SKU. I found no public evidence of a commercial product marketed by Toshiba as embodying these claims; the specification describes internal use ("the human that performs the design or the development of the new product confirms…").
• Status: Operating. Note that Toshiba's memory business was carved out and sold (Toshiba Memory Corp. → Bain-led consortium → renamed Kioxia) with the share transfer executed 2018-06-01, and Toshiba Corp. itself later underwent a take-private restructuring. Neither event touches this patent: it is a design-process/software asset, and it stayed with Toshiba (see timeline).

Assignment timeline

Exactly one assignment is recorded against US 16/013,377 / US 11,402,427. It is the routine pre-issuance inventor→employer assignment. No post-issuance assignment exists in the record.

• 2018-06-18 (executed) / recorded 2018-06-20 — Reel 046146/0697
  • Conveyance: Assignment — "ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)"
  • Assignor: INOUE, MICHINOBU; NISHIUCHI, HIDEO
  • Assignee: KABUSHIKI KAISHA TOSHIBA (Japan)
  • Correspondent: Not disclosed in the records I retrieved. The Google Patents legal-events extract surfaces reel/frame, conveyor type, parties, and effective date only. I did not verify the correspondent of record, and I decline to name one. There is therefore no basis to assess recurrence — there is only one entry to recur, and no second link in the chain.
  • Context: Initial inventor-to-employer assignment, recorded the same day the US application was filed (2018-06-20). This is a new-filing housekeeping record, not a transfer.

Recorded post-issuance transfers: none. The patent issued 2022-08-02 — over four years after the only recorded assignment — and no assignment was recorded around grant, and none since. The only other legal events on the docket are prosecution/grant events and the 2026-01-21 payment of the 4th-year maintenance fee by a large entity (Toshiba), which is direct evidence the original assignee still holds and is maintaining the patent.

Cross-references checked:

• Family: JP 6806651 B2, KR 102360631 B1, DE 102018211196 A1, TW I711001 B — all Toshiba-attributed. No divisional or continuation assignment to a third party surfaced.
• Citations / citing documents: The patents citing this one (CyberArk US 20220159029, NTT US 20220358526) are citations, not assignments; they are irrelevant to the ownership chain.
• Litigation: I found no infringement action naming US 11,402,427 (or its family members) as an asserted patent. The Toshiba adverse-litigation hits in my search (NEC/MiiCs & Partners v. Toshiba; Samsung/Funai; Better Browsing LLC v. Toshiba; Toshiba v. Imation) are cases where Toshiba is a defendant or the patents are unrelated DVD/LCD/NAND assets — none involve this patent.
• SEC / Toshiba disclosures: Toshiba confirmed the memory-business share transfer to the Bain-led consortium (2018-06-01 execution) in its securities report; that transaction moved memory-related patents to Toshiba Memory/Kioxia, not this design-process patent. The IAM/AST top-seller survey (07 Jan 2019) lists Toshiba among large patent sellers in the semiconductor category — again, a different asset category from this patent.

⚠️ Do not conflate the reel number with PCT/US2011/046146. A search hit for "046146" is a PCT application number for an unrelated photodetector filing, not USPTO reel 046146. The recorded reel here is 046146/0697, Toshiba, executed 2018-06-18.

Timeline diagram

timeline
    title Ownership of US 11402427
    2017 : Priority filing in Japan by Toshiba
         : Inventors Inoue and Nishiuchi named
    2018 : US application 16013377 filed
         : Inventors assign rights to Toshiba
    2022 : Patent US 11402427 granted
    2026 : Fourth year maintenance fee paid

NPE / troll-pattern signals

Aggregate signal count: zero strong signals, zero weak signals supported by evidence.

Verdict

Insufficient data — not because records are missing, but because the only record is the original assignment. The chain is: inventors → Kabushiki Kaisha Toshiba, per reel 046146/0697 (executed 2018-06-18, recorded 2018-06-20), and nothing else. No shell transfer, no known asserter, no cascade, no pre-litigation transfer, and no defensive aggregator appears anywhere in the record; Toshiba's 2026-01-21 payment of the 4th-year maintenance fee as a large entity confirms the original assignee still holds the patent. The correct reading is a clean, unencumbered operating-company ownership chain, with no NPE evidence whatsoever — asserting a troll pattern here would be pure speculation from the absence of a chain rather than from any recorded fact.

Verify at: https://assignmentcenter.uspto.gov/ (search "11402427" or application "16/013,377"; the same records are mirrored at https://assignment.uspto.gov/patent/index.html and on the Google Patents legal-events tab at https://patents.google.com/patent/[US11402427B2](/patent/US11402427B2)/en). If you need the correspondent of record for reel 046146/0697, it must be pulled from the Assignment Center abstract view or the recorded PDF cover sheet — it was not exposed in the sources available here, and I have not substituted a guess.

I'll research the prior art citations for US 11,402,427. Let me run targeted searches.

Let me gather details on additional key references.

Prior-Art Analysis — US Patent 11,402,427 B2

Patent under analysis: US 11,402,427 B2, "Information processing system and information processing method," Toshiba (Kabushiki Kaisha Toshiba), inventors Michinobu Inoue & Hideo Nishiuchi; app. 16/013,377 filed 2018‑06‑20; priority JP 2017‑176705 (2017‑09‑14); granted 2022‑08‑02. 6 claims (1 independent).

Source of the citation set: The complete list of references below is taken from the front page of US 11,402,427 B2 itself as rendered on Google Patents (the authoritative document provided). I confirmed several references against their own full-text records (Itid Consulting US 2010/0287105 A1, Hitachi US 2009/0292384 A1 / US 8,000,819 B2, Honeywell US 6,684,349 B2, D&R Solutions US 7,689,945 B2, and CEPREI CN 104166800 A). For a few references (e.g., US 7,451,065 B2, US 2009/0083333 A1, US 2009/0157453 A1) I was able to confirm bibliographic data only; full-text retrieval was cut short, so those descriptions are flagged as lower-confidence.

§ 102 standard applied. A reference anticipates a claim only if it discloses every limitation, arranged as in the claim (or an obvious variant thereof). Claim 1 is a dense, multi-element claim. As the previously generated summary already flagged, the granted Claim 1 appears to reflect amendment and combines: (i) a three-group correlation data structure with per-link "connection information"; (ii) a forward traversal spec → risk → element over all possible paths; (iii) element selection; (iv) a reverse traversal element → risk over all possible paths; (v) retrieval of mechanism and/or example "risk information"; (vi) an evaluation value → margin value (vs. threshold) → predicted risk value calculation; and (vii) modifying the current design when the predicted risk value shows the spec is not met. My assessment below therefore distinguishes anticipation candidates (must disclose all elements) from § 103 combination material and background. I found no single cited reference that clearly discloses all elements of claim 1.

1. Ranked summary (quick reference)

Remaining references: background only (see § 3).

2. Detailed analysis of the significant references

2.1 US 2010/0287105 A1 — Itid Consulting, Ltd. ("Information processing system, program, and information processing method") — closest overall

• Full citation: US 2010/0287105 A1; inventors Katsufumi Araki, Katsuya Terashima; assignee Itid Consulting, Ltd.; Appl. 12/841,155, filed 2010‑07‑21 as a continuation of PCT/JP2009/000183 (filed 2009‑01‑20); JP priority 2008‑013162 (2008‑01‑23); published 2010‑11‑11.
• Description: A product-development-project management system operating over three management layers (organizational/general-task, labor/specific-task, design/requirement-and-component). It acquires a "requirement/component dependency relationship" and computes risk values for requirements and components, then derives integrated task dependency relationships and schedules. (Source: US 2010/0287105 A1 abstract and FIG. 4 "ACQUISITION OF RISK VALUE / ACQUISITION OF DEGREE OF DEPENDENCY.")
• § 102 assessment: This is the most structurally similar reference: it explicitly links requirements (≈ specification items), components (≈ element items), and risk values (≈ risk items) in one dependency model, and it computes risk. That maps well onto elements (i) and (vi-partial) of Claim 1. However, on the record I reviewed I did not find disclosure of (a) a distinct intermediate risk-item group with bidirectional specification↔risk and risk↔element connection information, (b) all-paths forward then reverse traversal as claimed, (c) the mechanism/example risk-information retrieval, or (d) the evaluation value → margin → predicted risk value formula. Conclusion: a strong § 103 combination reference against Claim 1; potentially § 102-relevant to the broader concepts of Claims 2 and 4, but I cannot call it a clean § 102 anticipation of Claim 1. (Same family: JP WO2009093440 A1, listed as JPWO2009093440A1.)

2.2 US 2009/0292384 A1 / US 8,000,819 B2 — Hitachi, Ltd. ("Manufacturing instruction evaluation support system…")

• Full citation: US 2009/0292384 A1, published 2009‑11‑26 (granted as US 8,000,819 B2, 2011‑08‑16); inventors Hanawa et al.; assignee Hitachi, Ltd.; US Appl. 12/392,513 filed 2009‑02‑25; JP priority 2008‑119,068 (2008‑05‑21).
• Description: Reads a manufacturing-instruction parameter group (user ID, batch number, manufacturing temperature, etc.) and manufacturing performance data; using multiple-regression analysis it computes a risk rate for each manufacturing instruction parameter, selects parameters whose risk rate is ≤ the average as "available choices," and identifies optimum parameters and a regression equation. (Confirmed from the US 8,000,819 B2 specification.)
• § 102 assessment: Discloses manufacturing parameters as elements, performance data, and a quantitative risk calculation — relevant to Claim 1's (vi) risk quantification and to Claim 2 (manufacturing-process items). It does not disclose a specification-item group → risk-item group → element-item group graph, nor bidirectional traversal, nor mechanism/example information. Conclusion: § 103 material, especially against the "evaluation value/risk computation" portion of Claim 1 and against Claim 2. Not a § 102 anticipation of Claim 1.

2.3 CN 104166800 A — CEPREI / 工业和信息化部电子第五研究所 ("Component FMEA analysis method and system based on failure mechanisms")

• Full citation: CN 104166800 A; Appl. CN201410401825.4; inventors 来萍 (Lai Ping), 陈媛, 黄云, 何小琦; applicant CEPREI (China Electronic Product Reliability and Environmental Testing Research Institute); filed 2014‑08‑11, published 2014‑11‑26.
• Description: A component FMEA method built on failure mechanisms (FMMEA-style): the component is structurally decomposed into functional units; for each unit the failure mechanism and the failure modes it produces are analyzed; the effect of the mechanism/mode on the component is derived; a failure-physics model is built to identify failure causes; results are consolidated into an FMEA output. (Confirmed from the CN 104166800 A abstract and claim 1.)
• § 102 assessment: This reference is the best cited art for the "mechanism information" limitation of Claim 1 (and Claim 5): it expressly stores/derives failure mechanisms and causal relationships between a component and failure modes/mechanisms — i.e., the spec/element↔risk mechanism linkage. It maps to Claim 2's component-part items and to Claims 1/5's mechanism-information element. It does not, however, disclose the specification-item group, the bidirectional graph traversal, or the margin/predicted-risk-value formula. Conclusion: strong § 103 combination reference (mechanism + component/FMEA); potential § 102 material only for the mechanism-information concept recited in Claims 1/5, not for Claim 1 as a whole. Note the Chinese-language source; a certified translation would be required for any formal § 102/§ 103 reliance.

2.4 US 6,684,349 B2 — Honeywell International Inc. ("Reliability assessment and prediction system and method for implementing the same")

• Full citation: US 6,684,349 B2; assignee Honeywell International Inc.; priority 2000‑01‑18; granted 2004‑01‑27. (Confirmed from the US 6,684,349 B2 specification and the family-citation list of the '427 patent.)
• Description: A Reliability Assessment Program (RAP) that assesses the reliability of new equipment/parts by comparing similarities and differences with predecessor equipment being replaced; it ingests field-failure data (Honeywell QDS), new-component data (including identification of active/passive components, interconnects, solder joints, and other potential failure causes), and outputs reliability predictions. It discusses physics-of-failure models (e.g., solder-joint life models based on materials, dimensions, usage environment) and MTBF prediction.
• § 102 assessment: Directly relevant to the purpose of Claim 1 (a new product/design, a component part such as a solder joint, and a physics-based failure mechanism — echoing the '427 patent's solder example). It does not disclose the specification/risk/element three-group graph or the two-directional traversal. Conclusion: § 103 material, particularly against the "component part" (Claim 2) and reliability/"new product vs. existing product" context.

2.5 US 7,689,945 B2 — D&R Solutions, LLC ("Product reliability analysis")

• Full citation: US 7,689,945 B2; inventors Craig Damon Hillman, Nathan John Blattau; assignee D&R Solutions, LLC; provisional 60/717,030 filed 2005‑09‑15; US Appl. 11/531,399 filed 2006‑09‑13; granted 2010‑03‑30.
• Description: Receives a first set of inputs describing a device, automatically provides a second set of inputs for a group of assessments based on the first inputs and at least one database, performs assessments (thermal, vibration, time-to-failure, part-failure, DfR/DfM, derating) to assess product reliability, and generates a report; maps part numbers to database data to supply component data. (Confirmed from the US 7,689,945 B2 specification.)
• § 102 assessment: Relevant to the evaluation of a new product/design using component-part data and multiple assessment types, and to outputting a report — germane to Claim 1's element/output and to Claim 2. Lacks the specification/risk/element graph and the traversal. Conclusion: § 103 material against Claim 1's evaluation/output elements; § 102-relevant only to fragmentary concepts.

2.6 US 2009/0157453 A1 — Itid Consulting ("Product development process supporting system and product development process supporting method")

• Full citation: US 2009/0157453 A1, published 2009‑06‑18; assignee Itid Consulting Ltd.; priority 2005‑10‑14 (same family as JP 2007‑109073 A and KR 2008‑0039524 A). (Bibliographic data confirmed; full text not fully retrieved — lower confidence.)
• Description: Analyzes dependencies between requirements across layers of a product-development process, using QFD tables as input and DSM tables for dependency derivation, with partition analysis to order/classify requirements.
• § 102 assessment: Relevant to the layered/tree-like structuring of specification-type items recited in Claims 3 and 4 and to dependency linkage, but it is a requirements/dependency analysis tool, not a spec→risk→element graph with mechanism information or risk prediction. Conclusion: § 103 material, chiefly against Claims 3–4; not § 102 against Claim 1.

2.7 US 2009/0083333 A1 — Noriaki Yamamoto ("Design support system")

• Full citation: US 2009/0083333 A1, published 2009‑03‑26; priority 2007‑09‑25; counterpart JP 2009‑075993 A (listed separately). (Bibliographic data confirmed; description drawn from the title/known Hitachi-family design-support subject matter — flagged lower confidence.)
• § 102 assessment: A generic design-support system; potentially § 103 material for the design-modification/output framework of Claim 1, but I could not confirm disclosure of the three-group graph or traversal. Conclusion: background / weak § 103.

2.8 US 2015/0381649 A1 — Neo Prime, LLC ("Probabilistic Model For Cyber Risk Forecasting")

• Full citation: US 2015/0381649 A1, published 2015‑12‑31; priority 2014‑06‑30; assignee Neo Prime, LLC.
• § 102 assessment: Concerns probabilistic risk forecasting (cyber domain). Relevant only to the general notion of computing a predicted risk value; no product-specification/manufacturing-element graph. Conclusion: background / weak § 103.

3. Remaining cited references (background / field-classification only)

These were cited by the examiner or listed as family citations but, on their face, do not individually disclose the Claim 1 architecture. For each I give citation, dates, description, and the claim(s) to which they are at most tangentially relevant.

Family-citation documents (cited in the JP family, not necessarily against the US claims):

Non-patent literature

• Bozarth, C., Warsing, D., Flynn, B., Flynn, E. (2009), "The Impact of Supply Chain Complexity on Manufacturing Plant Performance," Journal of Operations Management 27:78–93. — Cited by the examiner. It is an empirical operations-management study of supply-chain complexity and plant performance; it does not disclose a specification/risk/element data-processing architecture. Conclusion: background only; not § 102 or § 103 material for any claim.

4. Conclusions

1. No cited reference anticipates Claim 1 in full. Every candidate is missing at least one of Claim 1's distinctive elements — most commonly (a) the three-group correlation data structure with explicit bidirectional link information, (b) the forward-then-reverse "all possible paths" traversal, (c) the mechanism/example risk-information retrieval, and (d) the evaluation-value → margin-value (vs. threshold) → predicted-risk-value computation culminating in a design modification.

2. Closest anticipation/obviousness candidates, in order:

   • US 2010/0287105 A1 (Itid Consulting) — requirements + components + risk values in one linked model; best § 103 primary reference.
   • CN 104166800 A (CEPREI) — best support for the "mechanism information" limitation (Claims 1 and 5) via failure-mechanism/FMEA modeling.
   • US 2009/0292384 A1 / US 8,000,819 B2 (Hitachi) and US 6,684,349 B2 (Honeywell) — support the risk-quantification and component-reliability aspects of Claim 1 and the component/process split of Claim 2.

3. A plausible § 103 combination against Claim 1 would pair Itid US 2010/0287105 A1 (requirement/component dependency + risk values) with CEPREI CN 104166800 A (component failure-mechanism linkage), and optionally Hitachi US 8,000,819 B2 (statistical risk-rate evaluation) — but even this combination does not squarely teach the claimed evaluation value → margin → predicted risk value formula (which mirrors the '427 patent's V = ln[1/(A₁×A₂×A₃)] second-embodiment formula).

4. Claims 2, 3, and 4 (component-part/manufacturing-process split; independent tree structures) are the most vulnerable to § 102/§ 103: US 2009/0157453 A1 and US 2010/0287105 A1 (layered requirements/components), US 6,684,349 B2 (part-level reliability), and US 7,689,945 B2 (component/part assessment) collectively describe that subject matter.

5. Caveats and confidence

• Source limitation: This analysis uses the citation set from the '427 patent's own front page (authoritative) plus full text for the five references I retrieved (Itid US 2010/0287105 A1, Hitachi US 8,000,819 B2/US 2009/0292384 A1, Honeywell US 6,684,349 B2, D&R Solutions US 7,689,945 B2, CEPREI CN 104166800 A). For the remainder I have bibliographic data and titles only; their § 102/§ 103 characterizations are provisional. Full-text retrieval of US 7,451,065 B2, US 2009/0083333 A1, US 2009/0157453 A1, and US 2015/0381649 A1 was interrupted by tool limits and should be completed before relying on those entries.
• No "anticipation reference" designation in the file was reviewed. I do not have the actual Non-Final/Final Office Actions for application 16/013,377. The grant required traversing multiple rejections (2020–2021) and an after-final response/advisory action before allowance on 2022‑04‑25; whichever reference the examiner ultimately relied upon — and any claim-scope concessions — would materially affect this picture and should be pulled from USPTO PatentCenter (App. 16/013,377).
• Consistent with the earlier summary's flag, the exact text of granted Claim 1 (especially the margin/predicted-risk-value and design-modification limitations) should be verified against the official issued copy before any § 102 mapping is treated as final.
• I did not auto-correct any identifier: each citation is reproduced as listed on the patent. Where a listed reference has a distinct sibling (e.g., US 2009/0292384 A1 → granted US 8,000,819 B2), I note the relationship rather than merging the numbers.

Recommended next step: Pull the full file wrapper (Office Actions + applicant responses) for App. 16/013,377 and perform full-text § 102 charts against the four Tier-1/Tier-2 references identified above.

I'll ground this obviousness analysis in the actual prior-art record, so let me pull details on the key references cited on the patent's page.

Now let me pull details on the remaining cited references relevant to a §103 combination.

Obviousness Analysis — U.S. Patent No. 11,402,427 B2 under 35 U.S.C. § 103

Builds on: the Patent Summary, Plain-Language Overview of the Independent Claim, and Litigation Summary already generated. I do not repeat bibliographic data, the abstract, or the no-litigation finding here; I only reference them where the § 103 analysis depends on them. Note one contradiction to flag up front: the task header states "Current Date: April 26, 2026," while my operating date is 2026-09-29 and the patent page was fetched 2026-09-24. This does not change any analysis below (all prior art pre-dates the 2017-09-14 priority date by a wide margin), but the two dates are inconsistent as given.

1. The § 103 Frame: What Actually Has to Be Taught

From the issued claim set as rendered on the patent page, Claim 1 decomposes into these limitations. The reference numerals are my own labeling for the tables below.

POSITA definition I assume: a bachelor's degree in mechanical/electrical/manufacturing engineering or equivalent, plus 2–3 years of experience in one or more of (a) computer-aided product/process design or PLM systems, (b) QFD/DSM/DFMEA/PFMEA quality engineering, or (c) applied statistical reliability prediction — the ordinary skill level implied by the art cited on the face of the patent (G06Q 10/06; G06F 30/20; G01R 31/28).

2. Prior Art Available (from the page's Prior Art sections, with content verified by search)

Everything in the table below is either (i) a prosecution/family citation listed on the patent page, or (ii) an examiner-cited NPL, and all pre-date 2017-09-14.

Not usable as prior art (post-date the 2017-09-14 priority date; listed only as "Cited By" for awareness): US 2022/0159029 A1 and US 2026/212331 B2 (CyberArk); US 2022/0358526 A1 (NTT); the "Families Citing" documents US 2024/0403742 A1 (Advantest) and JP 2025-009287 A (Hitachi). Under § 102(a)(1)/(a)(2) these cannot be relied on at all.

3. Primary § 103 Combination (attacks all of L1–L14)

Combination A: ITID-B (US 2010/0287105) as primary + ITID-A (US 2009/0157453 / JP 2007-109073) + CN 104166800 + Hitachi-A (US 2009/0292384 / US 8,000,819), optionally + Honeywell (US 6,684,349).

Element-by-element:

• L2, L3, L4, L5 — three-tier item architecture. ITID-B discloses a design-management layer holding "requirements that are necessary for the product" and "components that are included in the product," and it acquires risk values for both the requirements and the components, along with degrees of dependency between them. Requirements ↔ specifications (performance/reliability), components ↔ element items. The only gap versus L2 is the formalization of a discrete intermediate risk item group of nodes; ITID-B's functionally equivalent construct is the risk value / degree-of-dependency layer sitting between requirement and component nodes. Alternative (cleaner) fill: CN 1573785 A and the FMEA art (CN 104166800) supply the discrete "failure mode / risk item" intermediate layer as a matter of routine quality engineering.
• L6, L7 — two sets of connection information. ITID-B and ITID-A both store and compute dependency links between adjacent layers as explicit data ("information indicating the dependency between two requirements belonging to adjacent layers"). ITID-A expressly derives links between requirements in the same layer and across layers using QFD and DSM matrices — a DSM is a connection matrix, i.e., "connection information."
• L8, L10, L6 — bidirectional traversal through the intermediate layer. ITID-B computes dependencies from requirements to components and, for the reverse direction, differential risk values of the requirements and components — i.e., it necessarily propagates dependency/risk information both ways through its intermediate risk/dependency layer. ITID-A's DSM partition analysis likewise requires traversing a dependency matrix in both directions to order and cluster requirements/elements. Two-directional traversal of a linked data structure is, independently, among the most elementary known computer techniques; no POSITA would regard "iterate the graph from the other end, following the same edges" as an inventive contribution.
• L9 — user selects an element item. Plainly conventional interaction; the '427 specification itself (FIG. 3, step S15) treats it as a bare user selection.
• L11 — mechanism/example risk information. CN 104166800 is squarely on point: it analyzes the failure mechanism, the failure mode it produces, the effect on the component, the failure physics model, and the failure cause, then integrates them into an FMEA result that is output. That is "mechanism information describing a specific relationship between the risk item and the specification/element item." Honeywell supplies the "example information" prong: a database of actual field failures (QDS; part numbers, field returns, confirmed failures) tied to specific parts — i.e., past instances of the risk.
• L12 — evaluation value → margin/threshold → predicted risk value. Hitachi-A teaches exactly this computational skeleton: compute a risk rate per manufacturing parameter (the evaluation value), compute an average risk rate among parameters and identify parameters whose risk rate is ≤ that reference value (a threshold comparison producing a margin-like determination), then compute and output a regression equation / optimum parameter (the predicted value). Honeywell and IBM '065 confirm the prediction step is a generic, well-known statistical technique in the same reliability-engineering field.
• L13 — output. All three primary references have an explicit output interface/display (Hitachi-A's "output interface"; ITID-B's terminal apparatuses; CN 104166800's FMEA result).
• L14 — modify the design when the prediction says the new spec is unmet. This is the stated purpose of every reference: ITID-A ("prevent wasteful repetitive tasks" in the development process), ITID-B (re-synchronize schedules when design data changes), Hitachi-A (identify optimum parameters and issue new manufacturing instructions), Honeywell (assess whether a new part will meet reliability targets before fielding it). Applying the computed result to change the design is the references' own raison d'être.

Motivation to combine (KSR / MPEP 2143 rationales):

1. Same field and same problem (rationale (C) — known technique improving a similar device in the same way). All of ITID-A, ITID-B, Hitachi-A, CN 104166800, Honeywell and Yamamoto are computer-aided product-design / manufacturing / quality-risk tools. The '427 background section itself concedes the problem being solved is a known, long-felt one in the manufacturing sector ("it is desirable to extract, at an earlier stage (further upstream in the development), the risks that may become apparent downstream").
2. Predictable combination of known elements (rationale (A)). Layering an FMEA-style risk node between spec items and parts/processes, and layering a regression/threshold computation onto that graph, yields no surprising result — each component performs exactly the function it was known to perform (ITID's dependency graph → ITID's dependency graph + FMEA risk nodes + Hitachi's regression).
3. Design incentive / market force (rationale (F)). The Bozarth et al. NPL supplies the examiner-facing rationale that added manufacturing/supply-chain complexity degrades plant performance — i.e., an explicit incentive to surface the risk created by adding/altering a part or process before committing to it.
4. Simple substitution (rationale (B)). Substituting ITID-B's "degree of dependency / risk value" node construct with the FMEA-style "failure mode/mechanism" node construct is a substitution of one known risk-representation for another in the same system.
5. "Obvious to try" (rationale (E)). With only a finite number of known ways to represent a spec-to-part risk chain (QFD matrix, DSM matrix, FMEA/FMECA table), a POSITA would predictably arrive at the claimed arrangement.

Conclusion for Combination A: on this record, Claim 1 would have been obvious, and its allowance is most plausibly explained by (i) the examiner never assembling ITID-B + CN 104166800 + Hitachi-A into a single rejection (they appear on the citation list individually, not as a three-way combination), and/or (ii) the L12/L14 language being added by amendment late in prosecution (final rejection 2021-10-27 → notice of allowance 2022-04-25), such that the prior-art search was never re-run against the as-amended (or as-allowed) wording.

4. Dependent Claims

5. Where a Defensibility Argument Could Still Be Made

I am not asserting the claims are invalid; I am identifying what a patent owner would say and how strong each point is.

1. "The art links requirements directly to components; the '427 interposes a risk node layer as the sole bridge." Strength: moderate. ITID-B's risk values and degrees of dependency are arguably attributes of requirement/component nodes rather than a third group of discrete nodes that both other groups point into. A patent owner would press this as the point of novelty. Rebuttal: FMEA (taught by CN 104166800, and expressly referenced in the '427 spec's own terminology) is precisely a discrete "failure mode/mechanism" layer sitting between a requirement/function and a part/process; interposing it is the routine FMEA construct, and combination rationale (A) applies.
2. "No reference teaches the all possible paths traversal." Strength: weak. Comprehensive path enumeration through a small directed graph is a textbook technique; the recitation adds no algorithmic limitation (no weighting, no pruning rule, no ordering).
3. "No reference teaches deriving a margin value from a threshold and then a predicted risk value.) Strength: weak–moderate. Hitachi-A is close (risk rate vs. average risk rate, then a regression equation), but Hitachi-A's threshold is a computed average rather than a "predetermined threshold," and its output is a regression equation rather than a single "predicted risk value." A patent owner could argue the specific evaluation→margin→prediction pipeline is not squarely taught. Rebuttal: Honeywell, IBM '065 and Neo Prime each teach computing a predicted reliability/risk number from evaluated characteristics; combining them with Hitachi-A is the paradigm case of rationale (C). Note also that the specific formula actually disclosed in the '427 specification (V = ln[1/(A₁×A₂×A₃)]) is not recited in any claim — so whatever narrowness the formula might have conferred is not captured.
4. Secondary considerations. I found no evidence of commercial success, licensing, copying, or long-felt-but-unmet need tied to this patent (and, per the Litigation Summary, no assertion activity). Without a nexus, secondary considerations cannot carry the claims.
5. The most likely real-world outcome. If challenged in an IPR, the dispositive question would be whether Petitioner could find an FMEA + product-development-support + reliability-prediction combination with a clean motivation statement. CN 104166800 is a Chinese-language reference published 2014 that was rejected in China in 2019 — usable as prior art by its 2014-11-26 publication, but requiring a certified translation and careful claim-charting. Caveat flagged explicitly: I have relied on the Chinese-language abstract/claims text and machine-rendered summaries for CN 104166800, not a certified translation.

6. Caveats, Uncertainties, and Things I Could Not Verify

• The "Prior Art section" of the page is a citation list, not a rejection record. I could not retrieve the actual Office Actions, the examiner's stated rationale, or the pre-amendment claims from the text I reviewed. I therefore cannot say whether any of the combinations above was actually made by the examiner and overcome, or never made. The prosecution timeline (final rejection 2021-03-09 → after-final response → advisory action 2021-06-22 → non-final 2021-07-28 → final 2021-10-27 → RCE → allowance 2022-04-25) is consistent with either.
• Claim-wording dependence. Per the Patent Summary caveat, the granted Claim 1 text appears to contain limitations corresponding to the second embodiment's formula-based prediction, which the specification describes as a separate embodiment from the first embodiment's extract-and-output flow. If that wording is accurate, L12/L14 are supported only by the second embodiment, which makes the § 103 analysis above more art-dependent — and also raises a written-description / antecedent-basis problem for "margin value" that is a § 112 issue, not a § 103 defense, but which would matter to how narrowly Claim 1 is construed. Verify L1–L14 against the official issued claims in USPTO PatentCenter (application 16/013,377) before relying on this analysis.
• Not all reference content was verified at full-text level. Verified from primary documents: ITID-B (US 2010/0287105 — abstract, FIG. 3, FIG. 40, S401/S402, risk values and degrees of dependency), ITID-A/EP 1939798A1 (QFD/DSM dependency derivation), Hitachi-A (US 2009/0292384 abstract, FIG. 13 flowchart, US 8,000,819 claims), CN 104166800 (Chinese abstract + claim 1 + prosecution status), Honeywell (US 6,684,349 background/summary), Yamamoto '333 (US 2009/0083333 summary). Not verified at full text: CN 1573785 A, JP 2015-118593 A, US 7,451,065 B2, US 2015/0381649 A1, US 2017/0236065 A1, US 2017/0372704 A1, US 2017/0235848 A1, US 2017/0147722 A1 — these are used at abstract/citation-list level only and should not be relied on individually.
• Bozarth et al. (2009) NPL. The page shows it as an examiner-cited non-patent citation with the year 2009, supporting the general proposition that manufacturing complexity degrades plant performance. I did not retrieve the article itself, so I cannot quote specific findings; treat it as a motivation cite, not an element-by-element teaching cite.
• Analogous-art boundaries. Neo Prime (cyber-risk forecasting) and, to a lesser degree, Cassantec (industrial equipment malfunction) are outside the product-design field of endeavor. They are usable only for the proposition that quantitative risk forecasting was a known technique generally; they should not be used as the primary teaching of L12. Honeywell, Hitachi-A and IBM '065 are the safer § 103 vehicles for L12.
• § 101 is a separate and, in my view, more dangerous exposure. Claim 1 as issued is a method of organizing a design-review workflow by storing linked records and doing a threshold arithmetic comparison — closely analogous to the "organizing human activity"/mental-process line of cases. That is outside this task's scope but should be flagged as a parallel risk to validity.
• No prior-art search of my own. Per the task, I analyzed only the references in the page's Prior Art sections (plus the two ITID families, which the page cites twice under different titles). I did not run an independent novelty search; references outside this set — in particular the large FMEA/QFD/PLM prior-art corpus — were not considered and could strengthen or weaken the combinations above.
• Identifier discipline. As required, I have not auto-corrected any number. Note that at least three identifiers can be confused and I have kept them distinct: 11,402,427 (the patent at issue); US 2009/0292384 A1 vs. US 8,000,819 B2 (same Hitachi family — application publication vs. granted patent); and the earlier-generated Litigation Summary's " '114 Patent" false positive, which is US 10,936,114, not this patent. Likewise, US 2009/0157453 A1, JP 2007-109073 A, KR 20080039524 A, US 8,458,006 B2 and EP 1 939 798 A1 appear to be one ITID family and are treated as such; US 2010/0287105 A1 and JPWO2009093440A1 appear to be a second, distinct ITID family.

7. Bottom Line

On the prior art identified on the patent's own face, all six claims of US 11,402,427 would have been obvious under § 103 over ITID-B (US 2010/0287105 A1) in view of CN 104166800 A and Hitachi-A (US 2009/0292384 A1 / US 8,000,819 B2), with ITID-A (US 2009/0157453 A1 / JP 2007-109073 A) supplying the tree-structured multi-layer requirement/element architecture of claims 3–4 and Honeywell (US 6,684,349 B2) supplying the reliability-prediction and field-failure-example aspects of claims 1 and 5. The motivation is supplied by the identity of field (computer-aided product development and manufacturing risk management), the references' own stated objects, and the examiner-cited Bozarth et al. NPL tying manufacturing complexity to plant performance. The strongest residual non-obviousness argument is the interposition of a discrete risk-node group as the sole bridge between specification items and element items with all-paths traversal in two directions — but FMEA and bidirectional dependency-matrix traversal were both routine in this art by 2017, so I would expect that argument to fail unless the claim is construed to require something the specification does not actually describe (a risk worth confirming against the official issued claims, since the granted text appears to incorporate subject matter from a different embodiment than the one the flow is described for).