WO2017039665A1

WO2017039665A1 - Method and apparatus for rating failure data structures

Info

Publication number: WO2017039665A1
Application number: PCT/US2015/048231
Authority: WO
Inventors: David Sean FARRELL
Original assignee: General Electric Company
Priority date: 2015-09-03
Filing date: 2015-09-03
Publication date: 2017-03-09

Abstract

Approaches are provided for a case management system where a case data structure (204, 300) is stored in a memory device (202). The case data structure represents characteristics of a case associated with an abnormality detected in an industrial machine or system. The case data structure includes evidence (392), an interpretation (304), and a recommendation (310) which may be evaluated to determine a rating (312). The rating is input and stored in a rating field of the case data structure.

Description

METHOD AND APPARATUS FOR RATING FAILURE DATA STRUCTURES

Background of the Invention Field of the Invention

[0001] The subject matter disclosed herein generally relates to remote equipment monitoring. More specifically, the subject matter relates to assessment of information contained in case data structures. More specifically, the subject matter relates to rating of information contained in case data structures.

Brief Description of the Related Art

[0002] In industrial operations, industrial machines and systems are monitored to ensure proper operation and/or detect anomalies which may arise. Remote Monitoring & Diagnostic (M&D) approaches often include personnel at one location communicating with personnel at an operating site located at a separate, geographically remote location. The M&D personnel view information related to industrial machines or systems located at the operating site.

[0003] During operation, problems oftentimes occur which may warrant an operator or maintenance engineer's involvement. Using known information related to the industrial machine or system, M&D personnel provide recommendations to personnel at the operating site.

[0004] The information viewed by M&D personnel may be unreliable or irrelevant to the underlying problem. At the time of recommendation, it is difficult for M&D personnel to appreciate the reliability or relevance of the information. The recommendations given by the M&D personnel to the operating site may, in some instances, not be useful, or in other instances, be outright incorrect.

[0005] The above-mentioned problems have resulted in some user dissatisfaction with previous approaches and sub-optimal application of remote monitoring and diagnostic approaches. Brief Description of the Invention

[0006] The approaches described herein provide for assessing case data structures, including assessing the usefulness and correctness of the information contained within the case data structures. A grading/rating system may be used to grade/rate both information and human factors.

[0007] In many of these embodiments, a case data structure is stored in a memory device.

The case data structure represents characteristics of a case associated with an abnormality detected in an industrial machine or system. The case data structure includes an evidence field with evidence. The evidence includes a characteristic associated with the industrial machine or system. For example, the evidence associated with the industrial machine or system may include: a measured temperature, a measured vibration, a measured pressured, a calculated efficiency, a structural defect, a lifespan of machine, a machine history, and/or a detected position shift.

[0008] The case data structure also includes an interpretation field with an interpretation.

The interpretation includes a user determined condition based at least in part on the evidence. For example, the interpretation may be: a case diagnosis, a case prognosis, a case impact (e.g., an indication of the potential scope of impact should the issue proceed to failure), and/or a case urgency (e.g., how soon the analyst feels the issue should be addressed).

[0009] The case data structure also includes a recommendation field with a

recommendation. The recommendation includes a user determined course of action to undertake with respect to the industrial machine or system based at least in part on the interpretation. For example, the recommendation may be: watch, wait, manual inspection, offline analysis, contact subject matter expert, contact original equipment manufacturer, change operation, invasive inspection, minor maintenance, schedule work, and/or shut down.

[0010] The case data structure also includes a rating field for storing a rating. The case data structure may also include a rating explanation field for storing a rating explanation. The case data structure may also include a rating provider field for storing a rating provider.

[0011] The evidence, interpretation, recommendation, rating, rating explanation, and/or rating provider may be received by a processor via an input of an interface. [0012] At least one of the evidence stored in the evidence field, the interpretation stored in the interpretation field, and the recommendation stored in the recommendation field are evaluated. The evaluation may be performed, for example, based upon predetermined criteria.

[0013] A rating is determined based upon the evaluation. The rating may be, for example, a rating of the evidence and a usefulness of the evidence, a rating of the interpretation and a correctness of the interpretation, and/or a rating of the recommendation and a correctness of the recommendation.

[0014] In some approaches, the rating is a binary rating. In other approaches, the rating is a graduated range of three or more values.

[0015] The rating is entered via an input and is stored in the rating field.

[0016] In some examples, the rating is selectively displayed to a user. The rating may be displayed, for example, at a presentation device. The rating may be displayed in response to the processor receiving a request via the input to present one or more ratings.

[0017] In some aspects, the rating is evaluated to determine a risk or threat assessment level.

Brief Description of the Drawings

[0018] For a more complete understanding of the disclosure, reference should be made to the following detailed description and accompanying drawings wherein:

[0019] FIG. 1 comprises a comprises an illustration of an informational flow chart for providing information relating to industrial machines or systems according to various

embodiments of the present invention;

[0020] FIG. 2 comprises a block diagram illustrating an exemplary apparatus for managing information relating to industrial machines or systems according to various

embodiments of the present invention; [0021] FIG. 3 comprises a block diagram illustrating an exemplary case data structure for managing information relating to industrial machines or systems according to various embodiments of the present invention; and

[0022] FIG. 4 comprises an operational flow chart illustrating an approach for case management according to various embodiments of the present invention.

[0023] Skilled artisans will appreciate that elements in the figures are illustrated for simplicity and clarity. It will further be appreciated that certain actions and/or steps may be described or depicted in a particular order of occurrence while those skilled in the art will understand that such specificity with respect to sequence is not actually required. It will also be understood that the terms and expressions used herein have the ordinary meaning as is accorded to such terms and expressions with respect to their corresponding respective areas of inquiry and study except where specific meanings have otherwise been set forth herein.

Detailed Description of the Invention

[0024] Referring now to FIG. 1, a system 100 for monitoring industrial machines includes an operating site 110, optionally, a data center 120, and a central monitoring center 130. The operating site 110 includes one or more industrial machines, equipment, or systems of industrial machines or equipment 112. Examples of industrial machines 112 monitored in system 100 include aircraft machinery (e.g., turbine engines), marine machinery, mining machinery, oil machinery, gas machinery, health care machinery, telecom machinery, to mention a few examples. Other examples are possible.

[0025] Industrial machine 112 is operably connected to a local computing device 114 such that the computing device 114 receives or obtains information from the industrial machine 112. The computing device 114 may be continuously connected to the industrial machine 112, or may be removably connected to the industrial machine 112. In one approach, the computing device 114 is located at the operating site 110. In other approaches, the computing device 114 is instead located remotely from the industrial machine 112. [0026] Information received at the computing device 114 from the industrial machine

112 includes operational characteristics of the industrial machine 112. Operational characteristics may include a measured temperature, a measured vibration, a measured pressured, a calculated efficiency, a structural defect, a lifespan of machine, a machine history, and/or a detected position shift. Other examples are possible.

[0027] The computing device 114 may be any type of hardware device such as a personal computer, a tablet, a cellular telephone, and/or a personal digital assistant. Other examples are possible. The computing device 114 may include a processor, an interface (e.g., a computer based program and/or hardware) having an input (which may also include a user input) and an output, a memory, and a display device (e.g., a screen or a graphical user interface which allows for a visualization to be made). In this way, a user of the computing device 114 is able to observe information at the computing device 114 (such as operational characteristics of the industrial machine 112), input information into the computing device 114, send information from the computing device 114 to a remote device (such as at the data center 120 or the central monitoring center 130), and receive information from a remote device. The computer device 114 may be configured to run specific software applications, such as a historian.

[0028] The computing device 114 is operably connected to a data storage module 116.

The data storage module 1 16 includes a memory for short- and/or long-term storage of information received from the computing device 114. Examples of information received and stored at the data storage module 116 include historical information relating to the industrial machine 112, or information received at the computing device from a remote device (such as at the data center 120 or the central monitoring center 130).

[0029] The optional data center 120 is in communication with the operating site 110

(preferably, with the computing device 114 at the operating site) such that the data center 120 can send and/or receive information pertaining to one or more industrial machines 112 located at the operating site 110. The data center 120 maybe located at the operating site 110, at the central monitoring center 130, or in a location geographically remote from the operating site 110 and the central monitoring center 130. In one approach, the data center 120 is disposed on a cloud based network. [0030] The data center 120 includes one or more data storage modules 122 having corresponding memories. The data center 120 may also include one or more computing devices 124 that include a processor, an interface having an input (which may include a user input) and an output, a memory, and a display device (e.g., a screen or a graphical user interface which allows for a visualization to be made). Various applications may be performed at the data center 120, including analytic modeling, anomaly detection, and/or calculations of key performance indicators.

[0031] The central monitoring center 130 includes a computing device 132 that is in communication with the data center 120 such that the central monitoring center 130 can send and/or receive information pertaining to one or more industrial machines 112 located at the operating site 110. Alternatively, the central monitoring center 130 is in communication with the operating site 110 (preferably, with the computing device 114 at the operating site) such that the central monitoring center 130 can send and/or receive information pertaining to one or more industrial machines 112 located at the operating site 110.

[0032] In one example of the operation of the system of 100 of FIG. 1, a case data structure is stored in a memory device that may be, for example, at the data center 120 or at the central monitoring center 130. The case data structure, discussed in greater detail elsewhere herein, represents characteristics of a case associated with an abnormality detected in an industrial machine or system 112. The case data structure includes an evidence field with evidence (e.g., a characteristic associated with the industrial machine or system), an

interpretation field with an interpretation (e.g., a user determined condition based at least in part on the evidence), a recommendation field with a recommendation (e.g., a user determined course of action to undertake with respect to the industrial machine or system based at least in part on the interpretation), and a rating field.

[0033] One or more of the evidence, the interpretation, and the recommendation is evaluated by personnel at the operating site 110, at the data center 120, and/or at the central monitoring center 130 to determine at least one rating. The rating is entered via an input and is stored in the rating field of the case data structure. [0034] With reference now to FIG. 2, an apparatus 200 (such as computing device 132 of

FIG. 1) includes a memory device 202. The memory device 202 stores a case data structure 204 (discussed in greater detail elsewhere herein). The memory device 202 may also store one or more work data plans 206 and/or one or more prior case histories 208. A work data plan 206 includes prior maintenance performed on an industrial machine or system (such as industrial machine 112), as well as scheduled maintenance to be performed on an industrial machine or system in the future. A prior case history 208 includes previous case data structures 204 associated with an industrial machine or system, or with one or more classifications of industrial machines or systems.

[0035] The apparatus 200 further includes an interface 210 including an input 212 (which preferably includes a user input) and an output 214. The apparatus 200 may also include a display device 216. The apparatus 200 includes processor 218 coupled to the memory device 202, and the interface 210, and optionally, the display device 216.

[0036] When an anomaly, abnormality, or incident is detected in an industrial machine or system (such as machine 112 of FIG. 1), a case data structure 204 (or combination of case data structures 204) associated with the case is created and stored in the memory device 202. As used herein, a "case" is associated with an anomaly, an abnormality, or an incident detected in an industrial machine or system, and a "case data structure" 204 includes a data structure that represents a compilation of characteristics of the case.

[0037] In one approach, the case data structure 204 is generated by personnel at the central monitoring center 130. In another approach, the case data structure 204 is generated at a local computing device (e.g., local computing device 114 at the operating site 110 shown in FIG. 1). In either approach, a user may link evidence, expert interpretation associated with the evidence, metadata describing the particular nature of the industrial machine at issue, and/or other relevant information such that a visual aid is created.

[0038] When problems occur with an industrial machine or system at an operating site,

M&D personnel (located, for example, at central monitoring center 130) use information contained in the case data structure 204 to make various interpretations regarding the problematic machine. Also based on this information, M&D personnel make recommendations to operating site personnel.

[0039] Prior to, during, or after resolution of the problem (i.e., the "case"), it often is apparent that information contained within the case data structure 204 provided varying value in resolving the case. For example, it may be apparent that evidence contained within the case data structure 204 was of varying usefulness, that interpretations were of varying correctness, and that recommendations or actions taken were of varying importance in efficiently resolving the case. Personnel at either location may evaluate the information contained within the case data structure 204 and apply a rating to the evaluated information. Such evaluations may be performed prior to, during, or after resolution of the case.

[0040] In one example of the apparatus 200 of FIG. 2, a processor 218 is configured to receive a rating at the input 212. The rating is an evaluation of one or more of the evidence, interpretation, and recommendation stored in the case data structure 204.

[0041] In one aspect, the rating is a rating of evidence stored in an evidence field of case data structure 204. For example, personnel evaluating evidence such as a measured temperature of problematic industrial machine 112 may determine the measured temperature was - or was not - useful in determining or resolving the problem.

[0042] In another aspect, the rating is a rating of an interpretation stored in an

interpretation field of case data structure 204. For example, personnel evaluating an

interpretation such as a case diagnosis of a problematic industrial machine may determine the diagnosis was - or was not - correct in diagnosing the problem.

[0043] In yet another aspect, the rating is a rating of a recommendation stored in a recommendation field of case data structure 204. For example, personnel evaluating a recommendation such as a recommendation to shut down a problematic industrial machine may determine the recommendation was - or was not - a correct recommendation in resolving the problem.

[0044] In some approaches, the determined rating is a binary rating selected from a group of two values. Binary ratings may be numerical (e.g., "0"/"l"; "l"/"2"), descriptive (e.g., "Yes'VNo"; "Good'VBad"; "Use'VDon't Use"; "Negative'V'Positive";

" Appro ve'V'Disapprove"), or graphical (e.g., a thumbs up/down icon; star/no star; a

smiling/frowning face).

[0045] In other approaches, the determined rating is selected from a group of three or more values. The values may be graduated numerical ranges (e.g., "1" through "5"), descriptive (e.g., "Yes"/"Maybe"/"No"; "Good"/"OK"/"Bad"; "Use"/"Consider"/"Don't Use";

"Negative"/"Neutral"/"Positive"; "Approve"/"Indifferent"/"Disapprove"), or graphical (e.g., one/two/three stars; a smiling/straight/frowning face).

[0046] The processor 218 is configured to store the rating in a rating field of the case data structure 204 in the memory 202.

[0047] Evaluations and corresponding ratings of information contained in case data structures 204 may be aggregated over time. Personnel across various locations (e.g., operating site 110 and/or central monitoring center 130) may subsequently view and consider previous ratings to better inform as to, for example, the importance or accuracy of various types of information, the usefulness of performing corrective actions, the accuracy of certain personnel's interpretations, or the effectiveness of certain personnel's recommendations. Knowledge of prior ratings may provide guidance in resolving subsequent cases.

[0048] In this regard, in response to receiving via the input 212 a request to present one or more ratings, the processor 218 is configured to provide via the output 214 the one or more ratings for presentation at a presentation device 216.

[0049] With reference now to FIG. 3, a case data structure 300 (such as case data structure 204 stored in memory device 202) may include an evidence field 302 with evidence. The evidence includes information associated with the anomaly and/or the industrial machine 112. For example, the evidence associated with the industrial machine or system may include: a measured temperature, a measured vibration, a measured pressured, a calculated efficiency, a structural defect, a lifespan of machine, a machine history, and/or a detected position shift. The evidence may be in the form of advisories, alarms, charts, or reports. [0050] The case data structure 300 also includes an interpretation field 304 with one or more interpretations. The interpretation includes a user determined condition based at least in part on the evidence. For example, the interpretation may be: a case diagnosis, a case prognosis, a case impact, and/or a case urgency. The interpretation field 304 may further include an impact field 306 for storing a case impact value, and an urgency field 308 for storing a case urgency value.

[0051] The case data structure 300 also includes a recommendation field 310 with one or more recommendations. The recommendation includes a user determined course of action to undertake with respect to the industrial machine or system based at least in part on the interpretation. For example, the recommendation may be: watch, wait, manual inspection, offline analysis, contact subject matter expert, contact original equipment manufacturer, change operation, invasive inspection, minor maintenance, schedule work, and/or shut down.

[0052] The case data structure 300 also includes a rating field 312 for storing one or more ratings. The rating field 312 may include an explanation field 314 for storing a rating explanation and/or a provider field 316 for storing a rating provider.

[0053] The case data structure 300 may also include a permission field 318, a case history field 320, and/or one or more widgets 322.

[0054] The information contained in a case data structure 300 may not only be used to assist an analyst in ascertaining a solution to the present case, but it also may be used in subsequent cases to better aid analysts from exploring resolutions which have been historically shown to be ineffective. Additionally, the system may be configured to automatically access past cases which may be related to the present case to assist the analyst in determining the best solution. Any information that is used in the present case may also be linked to provide additional information within the apparatus.

[0055] The case data structure 300 is structured so as to allow a user to provide updates to the case, to evidence relating to the case, to their expert interpretation as to the meaning and implication of the evidence (that is, what the issue might be, and what to do about it at a particular time), and to their recommendation regarding actions to be taken. Additional abnormalities which may occur prior to or after the creation of the case data structure 300 may also be linked to the created case data structure 300. Ancillary capabilities such as collaboration, workflow with assignment/request timers, analytic escalation notifications, and other constructs can be input and stored in the case data structure 300. That is, whatever data structure is used, the case data structure 300 is easily modified.

[0056] With reference now to FIG. 4, a method 400 includes storing 402 a case data structure (e.g., case data structure 300) in a memory device (e.g. memory device 202). As previously discussed, the case data structure represents characteristics of a case associated with an abnormality detected in an industrial machine or system and may include any one or more of the fields and/or information discussed with respect to case data structure 300, including an evidence field 302 with evidence, an interpretation field 304 with an interpretation, a

recommendation field 310 with a recommendation, and a rating field 312.

[0057] The method 400 further includes evaluating 404 one or more of the evidence, the interpretation, and the recommendation to determine at least one rating. The evaluation and determination of a rating may be performed, for example, by an operating engineer at an operating site (e.g., operating site 110) or by M&D personnel (e.g., at central monitoring center 130). In some approaches, multiple people (at the same or multiple locations) may each evaluate the information and respectively determine a rating.

[0058] In one aspect, the rating is a rating of evidence stored in an evidence field (e.g., evidence field 302). In another aspect, the rating is a rating of an interpretation stored in an interpretation field (e.g., interpretation field 304). In yet another aspect, the rating is a rating of a recommendation stored in a recommendation field (e.g., recommendation field 310).

[0059] The method 400 further includes inputting 406 the rating via an input. The initial entering of the rating may be performed for example, by an operating engineer at an operating site (e.g., operating site 110) or by M&D personnel (e.g., at central monitoring center 130).

[0060] The method 400 further includes storing 408 the rating in the rating field (e.g., rating field 312 of FIG. 3). As discussed, multiple stored ratings may be aggregated over time. [0061] In some aspects, the method 400 includes searching and recalling (often referred to as "mining") prior stored ratings. The mining of prior stored ratings may be related to, for example, a type of industrial machine or system (e.g., a specific jet turbine model), or a classification of evidence (e.g., mechanical vibration in all jet turbine models).

[0062] In some aspects, the method 400 includes selectively displaying the rating to a user. This may occur for example, when a processing device (e.g., processor 218) receives via an input a request to present one or more ratings. In response, the processing device provides via an output the one or more ratings for presentation. The one or more ratings may be presented at a presentation device, such as a television screen or computer monitor.

[0063] Aggregation and subsequent mining of prior ratings may provide valuable insight into the resolution of current cases involving industrial machines or systems. Such ratings allow for quick analysis of various factors of a case, such as the accuracy of source alarms and analytics that initially indicated a potential problem, the usefulness of supplementary information bound to a case data structure, the accuracy and usefulness of user interpretations in case diagnosis and prognosis, the usefulness of recommendations and actions taken during the lifecycle of the case, and the timing and efficiency of the case resolution.

[0064] Aggregation of ratings also provides insight into the knowledge and reliability of

M&D personnel. For example, when certain personnel consistently receive more negative ratings than others, this may inform both those receiving recommendations from the M&D personnel and those overseeing the M&D personnel that the M&D personnel is less knowledgeable or reliable than others. Such "benchmarking" provides better insights to the quality of the M&D structure.

[0065] The assignment and storage of ratings has the additional benefit of streamlining workflow during case resolution. For example, in response to observing that an operating engineer at an operating site (e.g., operating site 110) has given a recommendation a negative rating, M&D personnel (e.g., at central monitoring center 130) may revise the recommendation and provide the revised recommendation to the operating site. [0066] Preferred embodiments of this invention are described herein, including the best mode known to the inventors for carrying out the invention. It should be understood that the illustrated embodiments are exemplary only, and should not be taken as limiting the scope of the invention.

Claims

WHAT IS CLAIMED IS:

1. A method comprising:

storing a case data structure in a memory device, the case data structure representing characteristics of a case associated with an abnormality detected in an industrial machine or system, the case data structure comprising:

- an evidence field with evidence, the evidence being a characteristic associated with the industrial machine or system;

- an interpretation field with an interpretation, the interpretation being a user determined condition based at least in part on the evidence;

- a recommendation field with a recommendation, the recommendation being a user determined course of action to undertake with respect to the industrial machine or system based at least in part on the interpretation; and

- a rating field;

evaluating one or more of the evidence, the interpretation, and the recommendation to determine at least one rating;

inputting the rating at an input; and

storing the rating in the rating field.

2. The method of claim 1, wherein the evidence associated with the industrial machine or system comprises a measured temperature, a measured vibration, a measured pressured, a calculated efficiency, a structural defect, a lifespan of machine, a machine history, or a detected position shift.

3. The method of claim 1, wherein the interpretation comprises a case diagnosis, a case prognosis, a case impact, or a case urgency.

4. The method of claim 1, wherein the recommendation comprises watch, wait, manual inspection, offline analysis, contact subject matter expert, contact original equipment manufacturer, change operation, invasive inspection, minor maintenance, schedule work, or shut down.

5. The method of claim 1, wherein the rating is a rating of the evidence and a usefulness of the evidence.

6. The method of claim 1 , wherein the rating is a rating of the interpretation and a correctness of the interpretation.

7. The method of claim 1, wherein the rating is a rating of the recommendation and a correctness of the recommendation.

8. The method of claim 1, further comprising selectively displaying the rating to a user.

9. The method of claim 10, wherein the rating is displayed at a presentation device.

10. The method of claim 1, wherein the case data structure further comprises a rating explanation field.

11. The method of claim 1 , wherein the case data structure further comprises a rating provider field.

12. An apparatus, the apparatus comprising:

an interface comprising an input and an output;

a memory including a case data structure, the case data structure representing characteristics of a case associated with an abnormality detected in an industrial machine or system, the case data structure comprising:

- an interpretation field with an interpretation, the interpretation being a user determined condition based at least in part on the evidence; - a recommendation field with a recommendation, the recommendation being a user determined course of action to undertake with respect to the industrial machine or system based at least in part on the interpretation; and

- a rating field;

a processor coupled to the interface and the memory, the processor configured to receive a rating at the input, the rating being an evaluation of one or more of the evidence, interpretation, and recommendation, the processor configured to store the input rating into the rating field of the case data structure in the memory, and the processor configured to, in response to receiving via the input a request to present one or more ratings, provide via the output the one or more ratings for presentation at a presentation device.

13. The apparatus of claim 12, wherein the evidence associated with the industrial machine or system comprises a measured temperature, a measured vibration, a measured pressured, a calculated efficiency, a structural defect, a lifespan of machine, a machine history, or a detected position shift.

14. The apparatus of claim 12, wherein the interpretation comprises a case diagnosis, a case prognosis, a case impact, or a case urgency.

15. The apparatus of claim 12, wherein the recommendation comprises watch, wait, manual inspection, offline analysis, contact subject matter expert, contact original equipment

manufacturer, change operation, invasive inspection, minor maintenance, schedule work, or shut down.

16. The apparatus of claim 12, wherein the rating is a rating of the evidence and a usefulness of the evidence.

17. The apparatus of claim 12, wherein the rating is a rating of the interpretation and a correctness of the interpretation.

18. The apparatus of claim 12, wherein the rating is a rating of the recommendation and a correctness of the recommendation.

19. The apparatus of claim 12, further comprising a presentation device coupled to the processor.