DETAILED ACTION
Claims 1-5, 9-12 and 14-16 are pending. Claims 6-8 and 13 are cancelled.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Acknowledgement is made of applicant’s claim for foreign priority under 35 U.S.C. 119 (a)-(d) to Japanese Patent Application No. 2022-029736, filed on 2/28/2022.
Response to Arguments
Applicant’s arguments, filed 9/3/26, have been fully considered but are not persuasive.
Applicant argues that the claims do not recite an abstract idea because they are directed to ‘a specific machine with specific functional components that operate in a defined manner… define a specific and concrete architecture for translating internal AI computational states into human-readable diagnostic knowledge’ (page 8).
It is respectfully submitted that the conventional computer arrangement, see Fig. 9 of the specification, is merely a generic computer and the ‘concrete architecture’ is merely an Applicant defined partition of the generic computer’s functions and thus not considered a ‘specific machine’. Applicant’s argument is therefore not persuasive.
Applicant argues that the claims are not directed to ‘the abstract idea of "displaying information" or "processing data." They are directed to a specific technical implementation: the use of internal processing information (intermediate layer vectors) from one trained machine learning model as the input to a second trained machine learning model (the description presentation model), which outputs operator-aligned diagnostic knowledge items. This is a specific technical solution to the specific technical problem of making AI-generated abnormality diagnoses understandable to human operators’ (pages 8-9).
It is respectfully submitted that ‘displaying information’ is moot because this is not considered part of the abstract idea — it is an insignificant additional element. Further, processing data using a known method (intermediate layer vectors) and a machine learning model is merely considered applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C, using a known algorithm. It is noted that the ‘intermediate layer vector’ feature is only recited in the dependent claims. Applicant’s argument is therefore not persuasive.
Applicant argues that the claims integrate the abstract idea into a practical application because they recite ‘a concrete improvement’ (page 9).
It is respectfully submitted that improving a mental process (abstract idea) is merely an improvement to an abstract idea and eligibility "cannot be furnished by the unpatentable law of nature (or natural phenomenon or abstract idea) itself." Genetic Techs. v. Merial LLC, 818 F.3d 1369, 1376, 118 USPQ2d 1541, 1546 (Fed. Cir. 2016) as cited in MPEP 2106.04, i.e. an improvement to the abstract idea itself (diagnosing an abnormality by analyzing data and determining a corresponding description) is still merely an abstract idea. It is also noted that any improvement is considered obvious — see the current rejection under 35 U.S.C. § 103. Applicant’s argument is therefore not persuasive.
Applicant argues that the claims recite significantly more than an abstract idea because they include: description presentation model storage unit holding a machine learning model, description presentation unit using processing information from the abnormality diagnosis unit (specifically, intermediate layer vectors) as input to the description presentation model, definition unit holding diagnosis knowledge items as relationship information between confirmation items and diagnosis results in the context of manual operator diagnosis (data structure) and a diagnosis flow display (page 10).
It is respectfully submitted that each of these elements is analyzed below in the current rejection under 35 U.S.C. § 101 below and have not been found to constitute significantly more than the abstract idea. As indicated below, analyzing data using machine learning on a generic computer is not significantly more; a ‘data structure’ is merely an abstract construct; and displaying data is not considered significantly more than the abstract idea. Inputting data to a computer, processing the data and displaying results constitutes a routine application of generic computer technology — see for example Tanaka or Honda — and is therefore not ‘a specific technical architecture’. Applicant’s argument is therefore not persuasive.
Applicant argues that the combination of Tanaka and Honda fails to teach a description presentation processing unit that uses processing information from the abnormality diagnosis unit to output diagnosis knowledge items via a trained description presentation model (pages 11-12).
It is respectfully submitted that Tanaka teaches these features because Tanak describes the abnormality diagnosis device 102 is described below. The abnormality diagnosis device 102 includes a control unit 51 and various kinds of databases (DBs). The control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25 [0031]; abnormality level determination unit 22 acquires the past disturbance torque applied to the motor mounted on the respective speed reducers 14 from the sensor DB 31, and calculates an abnormality level indicating a degree of abnormality of the disturbance torque acquired [0039, Fig. 1]; abnormality prediction unit 23 determines or predicts an abnormality caused in the respective speed reducers 14 in accordance with the abnormality level calculated for each speed reducer 14 [0043]; In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9 [0069, Fig. 4]; In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71 [0079-0083, Figs. 8-10]; the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model. A method of the machine learning may be known “supervised learning” (supervised/manually) [0104-0110]; control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25. The databases include a sensor DB 31, the maintenance history DB 32 (the maintenance history storage unit), an abnormality prediction DB 33, a correlation storage DB 34 (a correlation storage unit), and a work history DB 35 [0031-0035, 0043-0050].; the flowchart shown in FIG. 3 illustrates a process of the correlation analysis processing executed by the correlation analysis unit 25 [0061-0064, Fig. 3]; that notification content setting unit 24 sets the contents to be provided to the user through the indication on the display 62 of the user interface 103. The notification content setting unit 24 generates a tree image 73 indicating, with a tree structure, the abnormality diagnosis results of the speed reducers 14 determined to have an abnormality by the abnormality prediction unit 23[0052]; and displaying the diagnosis results on the display 62 [0021, Fig. 1]; and maintenance history image 72 displayed includes the image of sign K2 and the image indicating “Speed reducer 14b: requiring attention”. The same images as the sign K2 and the letters are indicated in the block q4 of the tree image 73[0081]. This is detailed below in the current rejection under 35 U.S.C. § 103. Applicant’s argument is therefore not persuasive.
Applicant’s arguments regarding the ‘intermediate layer vector’ features (pages 11-12) are moot because neither Tanaka nor Honda is cited as teaching these features — they are taught by the newly cited reference, Oba.
Applicant’s arguments regarding the alleged deficiencies in Honda (page 12) are moot because Honda is not cited as teaching these features. Applicant’s argument is therefore not persuasive.
Applicant argues that ‘The Office Action has not identified - and the cited references do not teach – the use of internal processing information from an AI-based abnormality diagnosis unit as the input to a separately trained description presentation model that outputs diagnosis knowledge items aligned with manual operator diagnostic practice’ (page 13).
It is respectfully submitted that this argument is moot, at least because the claims do not recite ‘artificial intelligence’ or ‘AI’ and neither do they explicitly cite a ‘separately trained description presentation model’. However, Tanaka teaches an AI-based abnormality diagnosis system that outputs diagnosis knowledge items aligned with manual operator diagnostic practice because Tanaka describes that the correlation analysis unit 25 executes the machine learning (AI) for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… A method of the machine learning may be known “supervised learning” (supervised/manually) [0104-0110]. Applicant’s argument is therefore not persuasive.
Applicant argues that the diagnosis knowledge item is not taught by Tanaka or Honda (page 13).
It is respectfully submitted that the relevant teachings of Tanaka were cited above and are detailed in the rejection below under 35 U.S.C. § 103. Tanaka teaches the diagnosis knowledge item, at least, because Tanaka describes the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model. A method of the machine learning may be known “supervised learning” (supervised/manually) [0104-0110]. Note that supervised (by an operator) learning encodes an operators knowledge. It is also noted that ‘a confirmation item’ is very broadly claimed. Applicant’s argument is therefore not persuasive.
Applicant’s arguments regarding the alleged deficiencies in Honda (pages 13-14) are moot because Honda is not cited as teaching these features. Applicant’s argument is therefore not persuasive.
Applicant argues that there is no proper motivation to combine Tanaka and Honda (pages 14-15).
It is respectfully submitted that the motivation to combine the teachings of Tanaka and Honda was specifically given on page 14 of the non-final office action and Applicant has not rebutted the reasoning presented there. Applicant’s argument is therefore not persuasive.
Applicant’s argument involving the ‘intermediate layer vector from an AI diagnosis model’ (page 15) is moot. The ‘AI’ feature was addressed above and neither Tanaka nor Honda is cited as teaching the ‘intermediate layer vector’ feature, that is taught by the newly cited reference, Oba. Applicant’s argument is therefore not persuasive.
For at least these reasons, the rejection of the claims is maintained.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
Claim limitations in this application that use the word “means” (or “step”) are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action. Conversely, claim limitations in this application that do not use the word “means” (or “step”) are not being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, except as otherwise indicated in an Office action.
Instances in the claims such as ‘abnormality diagnosis unit’, ‘description presentation processing unit’, ‘diagnosis model storage unit’, ‘definition unit’, ‘training unit’ etc. are interpreted under 35 U.S.C. 112(f) as incorporating a processor in accordance with [0034] of the specification/PGPub.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claim(s) 1-5, 9-12 and 14-16 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a non-statutory subject matter. The claims do not fall within at least one of the four categories of patent eligible subject matter because the claimed invention is directed to the abstract idea (mental process) of diagnosing an abnormality by analyzing data and determining a corresponding description.
Claim 1 recites an abnormality diagnosis device, i.e. a machine, which is a statutory category of invention. The claim recites:
an abnormality diagnosis unit which diagnoses an abnormality of a diagnosis target by using data related to the diagnosis target as an input;
a description presentation processing unit, wherein the description presentation processing unit comprises: a diagnosis knowledge item that is relationship information between a confirmation item and the result of diagnosis in a case where an operator manually performs diagnosis of the diagnosis target that may be performed in the human mind, or by a human using a pen and paper. Thus the claim recites an abstract idea (mental processes), see MPEP 2106.04(a).
This judicial exception is not integrated into a practical application because the additional elements, i.e. , an abnormality diagnosis device with an abnormality diagnosis unit and a description presentation processing unit, a definition unit which holds a diagnosis knowledge item (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C — see 112(f) interpretation above), a description presentation model storage unit which holds a machine learning model based on a past abnormality case as a description presentation model (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C); and a description presentation unit which presents the diagnosis knowledge item related to the abnormality of the diagnosis target in association with the result of diagnosis of the abnormality using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the description presentation model held in the description presentation model storage unit and presenting description data on a display unit which displays the result of diagnosis and a description corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit on a screen, wherein the description is the diagnosis knowledge item presented in association with the result of diagnosis (insignificant extra-solution activity — see MPEP 2106.04(a)(2) III A regarding displaying information and MPEP 2106.05(d)) do not impose any meaningful limits on practicing the abstract idea. The claim is therefore directed to an abstract idea.
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to integration of the abstract idea into a practical application an abnormality diagnosis device with an abnormality diagnosis unit and a description presentation processing unit, a definition unit which holds a diagnosis knowledge item (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C — see 112(f) interpretation above), a description presentation model storage unit which holds a machine learning model based on a past abnormality case as a description presentation model (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C); and a description presentation unit which presents the diagnosis knowledge item related to the abnormality of the diagnosis target in association with the result of diagnosis of the abnormality using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the description presentation model held in the description presentation model storage unit and presenting description data on a display unit which displays the result of diagnosis and a description corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit on a screen, wherein the description is the diagnosis knowledge item presented in association with the result of diagnosis (insignificant extra-solution activity — see MPEP 2106.04(a)(2) III A regarding displaying information and MPEP 2106.05(d)) are not considered significantly more. Considering the additionally elements individually and in combination and the claim as a whole, the additional elements do not provide significantly more than the abstract idea. Thus the claim is not patent eligible.
Claim 2 recites a diagnosis model storage unit (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C) which stores a machine learning model in which a relationship between a feature of data related to the diagnosis target and a diagnosis result is trained as a diagnosis model, wherein the abnormality diagnosis unit diagnoses the abnormality of the diagnosis target using a feature extracted from related to the diagnosis target at a time of abnormality diagnosis and the diagnosis model stored in the diagnosis model storage unit (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C), and outputs a diagnosis result of the diagnosis to the description presentation processing unit (insignificant extra-solution elements – merely using generic computer technology, see MPEP 2106.05 I A, MPEP 2106.05(g) MPEP 2106.05(d) e.g. receiving or transmitting data over a network), wherein the processing information used by the description presentation unit is an intermediate layer vector output by the diagnosis model during processing of data related to the diagnosis target (abstract data processed by a computer). Thus this claim recites an abstract idea.
Note that machine learning and neural networks are well-understood, routine and conventional, see for example Agarwal et al. U.S. Patent Publication No. 20200137585 [0028], Redmond et al. U.S. Patent Publication No. 20210342730 [0131], or Gross et al. U.S. Patent Publication No. 20210256389 [0019-0028].
Claim 3 recites the diagnosis model stored in the diagnosis model storage unit is a neural network (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C). Thus this claim recites an abstract idea.
Claim 4 recites the diagnosis result is a factor of the abnormality (abstract data). Thus this claim recites an abstract idea.
Claim 5 recites the diagnosis result is a countermeasure to be taken against the abnormality (abstract task description). Thus this claim recites an abstract idea.
Claim 9 recites the description presentation model stored in the description presentation model storage unit (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C) is a machine learning model obtained by training a correspondence between processing information obtained by processing a past abnormal case by the abnormality diagnosis unit (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C) and the diagnosis knowledge item held in the definition unit (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C). Thus this claim recites an abstract idea.
Claim 10 recites the description presentation processing unit further comprises a training unit (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C)), and the training unit trains a machine learning model for estimating the diagnosis knowledge item from the processing information by using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the diagnosis knowledge item held in the definition unit (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C)), and stores the trained machine learning model (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C)). Thus this claim recites an abstract idea.
Claim 11 recites the display unit displays the diagnosis knowledge item corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit on a screen in association with an abnormality diagnosis flow that represents a flow of processing (insignificant extra-solution activity — see MPEP 2106.04(a)(2) III A regarding displaying information and MPEP 2106.05(d)) in a case where an operator manually performs diagnosis of the diagnosis target (mental process), wherein the abnormality diagnosis flow includes combinations and an order of confirmation items used by the operator in manual diagnosis (abstract data). Thus this claim recites an abstract idea.
Claim 12 recites an abnormality diagnosis method for diagnosing an abnormality of a diagnosis target using an abnormality diagnosis device, i.e. a process, which is a statutory category of invention. However, the process is similar to that recited in claim 1 and is rejected under the same rationale.
Claim 14 recites a description presentation model storage unit (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C), the abnormality diagnosis method comprising presenting, by the description presentation processing unit, a diagnosis knowledge item related to the abnormality of the diagnosis target as a description corresponding to the abnormality of the diagnosis target in association with a diagnosis result of the abnormality by using an intermediate layer vector output by the abnormality diagnosis unit during processing of data related to the diagnosis target as the processing information and the description presentation model held in the description presentation model storage unit (insignificant extra-solution elements – merely using generic computer technology, see MPEP 2106.05 I A, MPEP 2106.05(g) MPEP 2106.05(d) e.g. receiving or transmitting data over a network) which holds a machine learning model based on a past abnormality case as a description presentation model (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C). Thus this claim recites an abstract idea.
Claim 15 recites displaying the diagnosis knowledge item corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit on the screen of the display unit in association with an abnormality diagnosis flow that represents a flow of processing (insignificant extra-solution activity — see MPEP 2106.04(a)(2) III A regarding displaying information and MPEP 2106.05(d)) in a case where an operator manually performs diagnosis of the diagnosis target (mental process), wherein the abnormality diagnosis flow includes combinations and an order of confirmation items used by the operator in manual diagnosis (abstract data). Thus this claim recites an abstract idea.
Claim 16 recites definition unit, the abnormality diagnosis method comprising training, by the description presentation processing unit, a machine learning model for estimating the diagnosis knowledge item from the processing information by using the processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit (applying the exception with generic computer technology using a known algorithm, see MPEP 2106.04(a)(2) III C) and the diagnosis knowledge item held in the definition unit, and storing the trained machine learning model (applying the exception with generic computer technology, see MPEP 2106.04(a)(2) III C). Thus this claim recites an abstract idea.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim(s) 1, 9-10, 12 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tanaka U.S. Patent Publication No. 20210178615 (hereinafter Tanaka) in view of Honda et al. U.S. Patent Publication No. 20230132587 (hereinafter Honda).
Regarding claim 1, Tanaka teaches an abnormality diagnosis device [0021, Figs. 1-2 — the abnormality diagnosis device 102] comprising:
an abnormality diagnosis unit which diagnoses an abnormality of a diagnosis target by using data related to the diagnosis target as an input [0031 — the abnormality diagnosis device 102 is described below. The abnormality diagnosis device 102 includes a control unit 51 and various kinds of databases (DBs). The control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25; 0039, Fig. 1 — abnormality level determination unit 22 acquires the past disturbance torque applied to the motor mounted on the respective speed reducers 14 from the sensor DB 31, and calculates an abnormality level indicating a degree of abnormality of the disturbance torque acquired; 0043 — abnormality prediction unit 23 determines or predicts an abnormality caused in the respective speed reducers 14 in accordance with the abnormality level calculated for each speed reducer 14.];
a description presentation processing unit which presents a description corresponding to the abnormality of the diagnosis target based on a result of diagnosis by the abnormality diagnosis unit [0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71 ], wherein the description presentation processing unit comprises:
a definition unit which holds a diagnosis knowledge item that is relationship information between a confirmation item and the result of diagnosis in a case where an operator manually performs diagnosis of the diagnosis target [0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model. A method of the machine learning may be known “supervised learning” (supervised/manually)];
a description presentation model storage unit which holds a machine learning model based on a past abnormality case as a description presentation model; and a description presentation unit which presents the diagnosis knowledge item related to the abnormality of the diagnosis target in association with the result of diagnosis of the abnormality using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the description presentation model held in the description presentation model storage unit [0031-0035, 0043-0050 — control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25. The databases include a sensor DB 31, the maintenance history DB 32 (the maintenance history storage unit), an abnormality prediction DB 33, a correlation storage DB 34 (a correlation storage unit), and a work history DB 35.; 0061-0064, Fig. 3 — The flowchart shown in FIG. 3 illustrates a process of the correlation analysis processing executed by the correlation analysis unit 25; 0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34; 0052 — notification content setting unit 24 sets the contents to be provided to the user through the indication on the display 62 of the user interface 103. The notification content setting unit 24 generates a tree image 73 indicating, with a tree structure, the abnormality diagnosis results of the speed reducers 14 determined to have an abnormality by the abnormality prediction unit 23]; and
a display unit which displays the result of diagnosis and a description corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit, [0021, Fig. 1 — display the diagnosis results on the display 62; 0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71; 0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71; 0081 — maintenance history image 72 displayed includes the image of sign K2 and the image indicating “Speed reducer 14b: requiring attention”. The same images as the sign K2 and the letters are indicated in the block q4 of the tree image 73], wherein the description is the diagnosis knowledge item presented in association with the result of diagnosis [0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71; 0081 — maintenance history image 72 displayed includes the image of sign K2 and the image indicating “Speed reducer 14b: requiring attention”. The same images as the sign K2 and the letters are indicated in the block q4 of the tree image 73].
But Tanaka fails to clearly specify displaying on a screen.
However, Honda teaches displaying on a screen [0037 — a display unit 24 including a display screen 241; 0042-0046 — display step S103 includes displaying, on the display screen 241, the at least one response procedure RP for each abnormal event AE in the abnormal device 2A… the plurality of response procedures RP are displayed on the display screen 241].
Tanaka and Honda are analogous art. They relate to abnormality detection and diagnosis systems.
Therefore at the time the invention was made, it would have been obvious to a person of ordinary skill in the art to simply substitute the known displaying on a screen, as taught by Honda, for the known displaying on a generic display, as taught by Tanaka, for the predictable result of a device that displays abnormality information on a screen.
Regarding claim 9, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches the description presentation model stored in the description presentation model storage unit is a machine learning model obtained by training a correspondence between processing information obtained by processing a past abnormal case by the abnormality diagnosis unit and the diagnosis knowledge item held in the definition unit [0034 — The maintenance history DB 32 stores the maintenance data on maintenance made for the robot 101 acquired when an abnormality is caused or the occurrence of an abnormality is predicted in the respective speed reducers 14; 0044 — abnormality prediction unit 23 detects an abnormality caused in each of the speed reducers 14a and 14b according to the maintenance data (such as the maintenance time and the contents of the maintenance) of the speed reducers 14a and 14b stored in the maintenance history DB 32 and the correlation stored in the correlation storage DB 34; 0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model].
Regarding claim 10, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches the description presentation processing unit further comprises a training unit, and the training unit trains a machine learning model for estimating the diagnosis knowledge item from the processing information by using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the diagnosis knowledge item held in the definition unit, and stores the trained machine learning model [0034 — The maintenance history DB 32 stores the maintenance data on maintenance made for the robot 101 acquired when an abnormality is caused or the occurrence of an abnormality is predicted in the respective speed reducers 14; 0044 — abnormality prediction unit 23 detects an abnormality caused in each of the speed reducers 14a and 14b according to the maintenance data (such as the maintenance time and the contents of the maintenance) of the speed reducers 14a and 14b stored in the maintenance history DB 32 and the correlation stored in the correlation storage DB 34; 0104-0110, Figs. 1-2 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model].
Regarding claim 12, Tanaka teaches an abnormality diagnosis method for diagnosing an abnormality of a diagnosis target using an abnormality diagnosis device [0005 — the present invention provides an abnormality diagnosis device and an abnormality diagnosis method; 0021, Figs. 1-2 — the abnormality diagnosis device 102] comprising an abnormality diagnosis unit [0021, Figs. 1-2 — FIG. 1 is a block diagram illustrating a configuration of an abnormality diagnosis device according to one embodiment of the present invention, and peripheral apparatuses. As illustrated in FIG. 1, the abnormality diagnosis device 102 according to the present embodiment is connected to a robot 101 (an apparatus) and a user interface 103 (indicated by reference sign “UI” in FIG. 1). The abnormality diagnosis device 102 diagnoses an abnormality of the robot 101, and outputs data regarding the diagnosis results to a display 62 (a display unit) provided in the user interface 103 so as to display the diagnosis results on the display 62. ], a description presentation processing unit [0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71], and a display unit [0021, Fig. 1 — display the diagnosis results on the display 62; 0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71], the abnormality diagnosis method comprising:
diagnosing an abnormality of the diagnosis target by the abnormality diagnosis unit using data related to the diagnosis target as an input [0031 — the abnormality diagnosis device 102 is described below. The abnormality diagnosis device 102 includes a control unit 51 and various kinds of databases (DBs). The control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25; 0039, Fig. 1 — abnormality level determination unit 22 acquires the past disturbance torque applied to the motor mounted on the respective speed reducers 14 from the sensor DB 31, and calculates an abnormality level indicating a degree of abnormality of the disturbance torque acquired; 0043 — abnormality prediction unit 23 determines or predicts an abnormality caused in the respective speed reducers 14 in accordance with the abnormality level calculated for each speed reducer 14.];
presenting, by the description presentation processing unit based on a result of diagnosis by the abnormality diagnosis unit, a diagnosis knowledge item related to the abnormality of the diagnosis target in association with a result of diagnosis of the abnormality, wherein the description presentation processing unit comprises a definition unit holding the diagnosis knowledge item as relationship information between a confirmation item and the result of diagnosis in a case where an operator manually performs diagnosis of the diagnosis target [0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model. A method of the machine learning may be known “supervised learning” (supervised/manually); 0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71], a description presentation model storage unit holding a machine learning model based on a past abnormality case as a description presentation model, and a description presentation unit presenting the diagnosis knowledge item using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the description presentation model [0031-0035, 0043-0050 — control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25. The databases include a sensor DB 31, the maintenance history DB 32 (the maintenance history storage unit), an abnormality prediction DB 33, a correlation storage DB 34 (a correlation storage unit), and a work history DB 35.; 0061-0064, Fig. 3 — The flowchart shown in FIG. 3 illustrates a process of the correlation analysis processing executed by the correlation analysis unit 25; 0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34; 0052 — notification content setting unit 24 sets the contents to be provided to the user through the indication on the display 62 of the user interface 103. The notification content setting unit 24 generates a tree image 73 indicating, with a tree structure, the abnormality diagnosis results of the speed reducers 14 determined to have an abnormality by the abnormality prediction unit 23]; and
displaying the result of the diagnosis and the diagnosis knowledge item corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit on the display unit [0021, Fig. 1 — display the diagnosis results on the display 62; 0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71; 0081 — maintenance history image 72 displayed includes the image of sign K2 and the image indicating “Speed reducer 14b: requiring attention”. The same images as the sign K2 and the letters are indicated in the block q4 of the tree image 73].
But Tanaka fails to clearly specify displaying on a screen.
However, Honda teaches displaying on a screen [0037 — a display unit 24 including a display screen 241; 0042-0046 — display step S103 includes displaying, on the display screen 241, the at least one response procedure RP for each abnormal event AE in the abnormal device 2A… the plurality of response procedures RP are displayed on the display screen 241].
Tanaka and Honda are analogous art. They relate to abnormality detection and diagnosis systems.
Therefore at the time the invention was made, it would have been obvious to a person of ordinary skill in the art to simply substitute the known displaying on a screen, as taught by Honda, for the known displaying on a generic display, as taught by Tanaka, for the predictable result of a method that displays abnormality information on a screen.
Regarding claim 16, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches the abnormality diagnosis device further comprises a definition unit, the abnormality diagnosis method comprising training, by the description presentation processing unit, a machine learning model for estimating the diagnosis knowledge item from the processing information by using the processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the diagnosis knowledge item held in the definition unit, and storing the trained machine learning model [0034 — The maintenance history DB 32 stores the maintenance data on maintenance made for the robot 101 acquired when an abnormality is caused or the occurrence of an abnormality is predicted in the respective speed reducers 14; 0044 — abnormality prediction unit 23 detects an abnormality caused in each of the speed reducers 14a and 14b according to the maintenance data (such as the maintenance time and the contents of the maintenance) of the speed reducers 14a and 14b stored in the maintenance history DB 32 and the correlation stored in the correlation storage DB 34; 0104-0110, Figs. 1-2 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model].
Claim(s) 2-5 and 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Tanaka and Honda in view of Oba U.S. Patent Publication No. 20190095300 (hereinafter Oba).
Regarding claim 2, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches a diagnosis model storage unit which stores a machine learning model in which a relationship between a feature of data related to the diagnosis target and a diagnosis result is trained as a diagnosis model [0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34],
wherein the abnormality diagnosis unit diagnoses the abnormality of the diagnosis target using a feature extracted from related to the diagnosis target at a time of abnormality diagnosis and the diagnosis model stored in the diagnosis model storage unit, and outputs a diagnosis result of the diagnosis to the description presentation processing unit [0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34; 0052 — notification content setting unit 24 sets the contents to be provided to the user through the indication on the display 62 of the user interface 103. The notification content setting unit 24 generates a tree image 73 indicating, with a tree structure, the abnormality diagnosis results of the speed reducers 14 determined to have an abnormality by the abnormality prediction unit 23].
But the combination of Tanaka and Honda fails to clearly specify that processing information used is an intermediate layer vector output by the diagnosis model during processing of data related to the diagnosis target.
However, Oba teaches that that processing information used is an intermediate layer vector output by the diagnosis model during processing of data related to the diagnosis target [0001 –The present disclosure relates to an anomaly diagnosis method and an anomaly diagnosis apparatus; 0104-0117, Figs. 7-8 — the autoencoder is a type of neural network and extracts the amount of information on a feature representing an input by employing dimensionality reduction. The autoencoder is used for machine learning in which, in a neural network with at least three layer that includes one or more intermediate layers of dimensions lower than the dimension of an input layer and the dimension of an output layer, weighting is determined such that the value of an input feature vector and the value of an output vector are equal to each other… score J(x) is the mean squared error with respect to an output vector when diagnostic data is fed into the autoencoder… the determination unit 160 determines that the fed diagnostic data 212 is anomalous (S26) and notifies the identification unit 170 of the determination result.].
Tanaka, Honda and Oba are analogous art. They relate to abnormality detection and/or diagnosis systems.
Therefore at the time the invention was made, it would have been obvious to a person of ordinary skill in the art to modify the above device, as taught by the combination of Tanaka and Honda, by incorporating the above limitations, as taught by Oba.
One of ordinary skill in the art would have been motivated to do this modification in order to accurately and easily determine an anomaly using a technique whose performance is known to be good, as taught by Oba [0130-0133].
Regarding claim 3, the combination of Tanaka, Honda and Oba teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches the diagnosis model stored in the diagnosis model storage unit is a machine learning model [0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34].
Further, Oba teaches that a model is a neural network [0104-0117 — the autoencoder is a type of neural network and extracts the amount of information on a feature representing an input by employing dimensionality reduction. The autoencoder is used for machine learning in which, in a neural network with at least three layer that includes one or more intermediate layers of dimensions lower than the dimension of an input layer and the dimension of an output layer, weighting is determined such that the value of an input feature vector and the value of an output vector are equal to each other… score J(x) is the mean squared error with respect to an output vector when diagnostic data is fed into the autoencoder… The determination unit 160 loads from the anomaly detection model DB 150, for example, the learned autoencoder (a type of neural network)… the determination unit 160 determines that the fed diagnostic data 212 is anomalous (S26) and notifies the identification unit 170 of the determination result].
Therefore at the time the invention was made, it would have been obvious to a person of ordinary skill in the art to simply substitute the known neural network model, as taught by Oba, for the known machine learning model, as taught by Tanaka, for the predictable result of an abnormality diagnosis device that utilizes a neural network.
Regarding claim 4, the combination of Tanaka, Honda and Oba teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches that the diagnosis result is a factor of the abnormality [0053, Figs. 8-10 — in the block q2, an image indicating “Abnormality level 2.1>Reference value 1.0” is created in which the numerical value of the abnormality level “2.1” calculated this time is greater than the reference value “1.0”. In addition, an image of sign K1 is created for easy recognition of the correspondence with the waveform of the abnormality level, as described below. The term “sign” as used herein encompasses a concept including characters, a predetermined mark, and an icon].
Regarding claim 5, the combination of Tanaka, Honda and Oba teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches that the diagnosis result is a countermeasure to be taken against the abnormality [0055, Figs. 8-10 — an image indicating the contents of the maintenance to be executed for this abnormality. The abnormality in this case is presumed to be caused in the speed reducer 14a or the speed reducer 14b, and an image indicating the execution of the maintenance, “Need to measure the density of iron powder in the grease for the speed reducers 14a and 14b”, is creased as shown in the block q5. The tree image 73 includes maintenance commands for the speed reducer 14a and for the speed reducer 14b predicted to have an abnormality, and is associated with the relation between the occurrence of the abnormality and the maintenance commands].
Regarding claim 14, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches the abnormality diagnosis device further comprises a description presentation model storage unit, the abnormality diagnosis method comprising presenting, by the description presentation processing unit, a diagnosis knowledge item related to the abnormality of the diagnosis target as a description corresponding to the abnormality of the diagnosis target in association with a diagnosis result of the abnormality by using processing information obtained by processing data related to the diagnosis target by the abnormality diagnosis unit and the description presentation model held in the description presentation model storage unit which holds a machine learning model based on a past abnormality case as a description presentation model [0031-0035, 0043-0050 — control unit 51 includes a communication unit 21, an abnormality level determination unit 22, an abnormality prediction unit 23, a notification content setting unit 24, and a correlation analysis unit 25. The databases include a sensor DB 31, the maintenance history DB 32 (the maintenance history storage unit), an abnormality prediction DB 33, a correlation storage DB 34 (a correlation storage unit), and a work history DB 35.; 0061-0064, Fig. 3 — The flowchart shown in FIG. 3 illustrates a process of the correlation analysis processing executed by the correlation analysis unit 25; 0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34; 0052 — notification content setting unit 24 sets the contents to be provided to the user through the indication on the display 62 of the user interface 103. The notification content setting unit 24 generates a tree image 73 indicating, with a tree structure, the abnormality diagnosis results of the speed reducers 14 determined to have an abnormality by the abnormality prediction unit 23].
But the combination of Tanaka and Honda fails to clearly specify using an intermediate layer vector output by the abnormality diagnosis unit during processing of data related to the diagnosis target as the processing information.
However, Oba teaches using an intermediate layer vector output by the abnormality diagnosis unit during processing of data related to the diagnosis target as the processing information [0001 –The present disclosure relates to an anomaly diagnosis method and an anomaly diagnosis apparatus; 0104-0117, Figs. 7-8 — the autoencoder is a type of neural network and extracts the amount of information on a feature representing an input by employing dimensionality reduction. The autoencoder is used for machine learning in which, in a neural network with at least three layer that includes one or more intermediate layers of dimensions lower than the dimension of an input layer and the dimension of an output layer, weighting is determined such that the value of an input feature vector and the value of an output vector are equal to each other… score J(x) is the mean squared error with respect to an output vector when diagnostic data is fed into the autoencoder… the determination unit 160 determines that the fed diagnostic data 212 is anomalous (S26) and notifies the identification unit 170 of the determination result.].
Tanaka, Honda and Oba are analogous art. They relate to abnormality detection and/or diagnosis systems.
Therefore at the time the invention was made, it would have been obvious to a person of ordinary skill in the art to modify the above method, as taught by the combination of Tanaka and Honda, by incorporating the above limitations, as taught by Oba.
One of ordinary skill in the art would have been motivated to do this modification in order to accurately and easily determine an anomaly using a technique whose performance is known to be good, as taught by Oba [0130-0133].
Claim(s) 11 and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of Tanaka and Honda in view of Phung et al. U.S. Patent Publication No. 20020007237 (hereinafter Phung).
Regarding claim 11, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above.
Further, Tanaka teaches the display unit displays the diagnosis knowledge item corresponding to the abnormality of the diagnosis target presented by the description presentation processing unit in association with an abnormality diagnosis flow that represents a flow of processing in a case where an operator manually performs diagnosis of the diagnosis target [0104-0110 — the correlation analysis unit 25 executes the machine learning for learning patterns of the maintenance data having a high probability of the occurrence of abnormality, in accordance with the maintenance data on the respective speed reducers 14 at least within a part of periods stored in the maintenance history DB 32. The correlation analysis unit 25 also detects an abnormality of the respective speed reducers 14 in accordance with the results of the machine learning… The leaning model created is stored in the correlation storage DB 34… The machine learning extracts the rules from the past abnormality data and maintenance data included in the maintenance history so as to create the learning model. A method of the machine learning may be known “supervised learning” (supervised/manually); 0069, Fig. 4 — In step S34, the notification content setting unit 24 generates the display data on the disturbance torque and the abnormality level to be displayed on the display 62 of the user interface 103. The abnormality level display image 71 as the display data is thus generated as illustrated in FIG. 9; 0079-0083, Figs. 8-10 — In step S44, the notification content setting unit 24 generates the display data of the tree image 73 as illustrated in FIG. 8. The tree image 73 is configured to allow the operator to easily recognize the details resulting in the contents of operation for instructing the operator in response to the detection of the abnormality in the speed reducer 14a… the abnormality level display image 71. The abnormality diagnosis image 70 is thus displayed with the configuration easy to recognize the correspondence relation between the tree image 73 and the abnormality level display image 71; 0081 — maintenance history image 72 displayed includes the image of sign K2 and the image indicating “Speed reducer 14b: requiring attention”. The same images as the sign K2 and the letters are indicated in the block q4 of the tree image 73].
Further, Honda teaches displaying on a screen [0037 — a display unit 24 including a display screen 241; 0042-0046 — display step S103 includes displaying, on the display screen 241, the at least one response procedure RP for each abnormal event AE in the abnormal device 2A… the plurality of response procedures RP are displayed on the display screen 241].
But the combination of Tanaka and Honda fails to clearly specify that the abnormality diagnosis flow includes combinations and an order of confirmation items used by the operator in manual diagnosis.
However, Phung teaches the abnormality diagnosis flow includes combinations and an order of confirmation items used by the operator in manual diagnosis [0039, Fig. 4 – combinations and an order of confirmation items are shown.; 0057, Fig. 8 — Continuing with the problem resolution 70 process, checks pertaining to the trouble code selected path would also be processed with the statistics/prioritization routine 80. The statistics/prioritization routine 80 communicates calculated results from the database system 200 to the client system 20 based on learned historical repair cases. The most probable repair checks are suggested, display checks 88, however the user ultimately decides on the troubleshooting path via the checks selection routine 90 (manual diagnosis)].
Tanaka, Honda and Phung are analogous art. They relate to abnormality detection and/or diagnosis systems.
Therefore at the time the invention was made, it would have been obvious to a person of ordinary skill in the art to modify the above device, as taught by the combination of Tanaka and Honda, by incorporating the above limitations, as taught by Phung.
One of ordinary skill in the art would have been motivated to do this modification in order to allow the technicians/operators to navigate dynamic trouble diagnostic trees based on identified problems and symptoms to arrive at the most likely causes and fixes quickly and efficiently, as taught by Phung [0019].
Regarding claim 15, the combination of Tanaka and Honda teaches all the limitations of the base claims as outlined above and this claim is otherwise rejected based on the same rationale as claim 11.
Note that any citations to specific, pages, columns, lines, or figures in the prior art references and any interpretation of the reference should not be considered to be limiting in any way. A reference is relevant for all it contains and may be relied upon for all that it would have reasonably suggested to one having ordinary skill in the art. See MPEP 2123.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BERNARD G. LINDSAY whose telephone number is (571)270-0665. The examiner can normally be reached Monday through Friday from 8:30 AM to 5:30 PM EST.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Mohammad Ali can be reached on (571)272-4105. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center for authorized users only. Should you have questions about access to Patent Center, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant may call the examiner or use the USPTO Automated Interview Request (AIR) Form at https://www.uspto.gov/patents/uspto-automated- interview-request-air-form.
/BERNARD G LINDSAY/
Primary Examiner, Art Unit 2119