WO2000060464A1

WO2000060464A1 - A method and system for analyzing operational data for diagnostics of locomotive malfunctions

Info

Publication number: WO2000060464A1
Application number: PCT/US2000/008661
Authority: WO
Inventors: David Richard Gibson; Nicholas Edward Roddy; Anil Varma
Original assignee: General Electric Company
Priority date: 1999-04-02
Filing date: 2000-03-31
Publication date: 2000-10-12
Also published as: AU758061B2; CA2368818C; AU4060200A; CA2368818A1; MXPA01009957A

Abstract

A method for analyzing fault log data (200) and snapshot operational parameter data (25) from railroad locomotives and other large land-based, self-powered transport equipment, undergoing diagnostics is provided. A receiving step (72) allows for receiving fault log data comprising a plurality of faults from the equipment. Respective executing steps allow for executing a set of noise-reduction filters (27) upon the received fault log data to generate noise-reduced fault log data, and for executing a set of candidate snapshot anomalies (28) upon the noise-reduced data to generate data predictive of malfunctions of the equipment.

Description

A METHOD AND SYSTEM FOR ANALYZING OPERATIONAL DATA FOR DIAGNOSTICS OF LOCOMOΗVE MALFUNCTIONS ^IΛΛΛJMU UVU

BACKGROUND OF THE INVENTION

The present invention relates generally to diagnostics of railroad locomotives and other large land-based, self-powered transport equipment, and, more specifically, to a system and method for hybrid processing of snapshot operational parameter data and fault log data to

5 facilitate analysis of machine equipment undergoing diagnostics.

A machine, such as a locomotive or other complex systems used in industrial processes, medical imaging, telecommunications, aerospace applications, power generation, etc., includes elaborate controls and sensors that generate faults when anomalous operating conditions of

10 the machine are encountered. Typically, a field engineer will look at a fault log and determine whether a repair is necessary.

Approaches like neural networks, decision trees, etc., have been employed to learn over input data to provide prediction, classification, and function approximation capabilities in the context of diagnostics.

15 Often, such approaches have required structured and relatively static and complete input data sets for learning, and have produced models that resist real-world interpretation.

Another approach, Case Based Reasoning (CBR), is based on the observation that experiential knowledge (memory of past

20 experiences - or cases) is applicable to problem solving as learning rules or behaviors. CBR relies on relatively little pre-processing of raw knowledge, focusing instead on indexing, retrieval, reuse, and archival of cases. In the diagnostic context, a case refers to a problem/solution description pair that represents a diagnosis of a problem and an appropriate repair.

25 CBR assumes cases described by a fixed, known number of descriptive attributes. Conventional CBR systems assume a corpus of fully valid or "gold standard" cases that new incoming cases can be matched against. U.S. Patent No. 5,463,768 discloses an approach which uses error log data and assumes predefined cases with each case associating an input error log to a verified, unique diagnosis of a problem. In particular, a plurality of historical error logs are grouped into case sets of common malfunctions. From the group of case sets, common patterns, i.e., consecutive rows or strings of data, are labeled as a block. Blocks are used to characterize fault contribution for new error logs that are received in a diagnostic unit. Unfortunately, for a continuous fault code stream where any or all possible fault codes may occur from zero to any finite number of times and where the fault codes may occur in any order, predefining the structure of a case is nearly impossible.

U.S. Patent Application Serial No. 09/285,611, (Attorney Docket No. RD-26576), assigned to the same assignee of the present invention, discloses a system and method for processing historical repair data and fault log data, which is not restricted to sequential occurrences of fault log entries and which provides weighted repair and distinct fault cluster combinations, to facilitate analysis of new fault log data from a malfunctioning machine. Further, U.S. Patent Application Serial No. 09/285,612, (Attorney Docket No. 20-LC-1927), assigned to the same assignee of the present invention, discloses a system and method for analyzing new fault log data from a malfunctioning machine in which the system and method are not restricted to sequential occurrences of fault log entries, and wherein the system and method predict one or more repair actions using predetermined weighted repair and distinct fault cluster combinations.

It is believed that the inventions disclosed in the foregoing patent applications provide substantial advantages and advancements in the art of diagnostics. It would be desirable, however, to provide a system and method that uses snapshot observations of operational parameters from the machine in combination with the fault log data in order to further enhance the predictive accuracy of the diagnostic algorithms used therein. It would be further desirable, through the use of noise reduction filters, to substantially eliminate undesirable noise, e.g., unreliable or useless information that may be present in the fault log data and/or the operational parameter data. This noise reduction would advantageously allow for increasing the probability of early detection of actual incipient failures in the machine, as well as decreasing the probability of falsely declaring nonexistent failures.

It would be further desirable to have a process and system that allows a field or diagnostic engineer or any other personnel involved in maintaining and/or servicing the machine to systematically analyze the fault log data so as to identify respective faults and/or respective combinations of faults that otherwise could be missed. It will be appreciated that since the fault log data contains useful information in order to detect incipient failures, it is desirable to accurately identify any such faults and/or combinations so that such personnel is able to proactively make repair recommendations and thus avoid loss of good will with clients as well as costly delays that could result in the event of a mission failure of the machine. An example of a mission failure would be a failed locomotive unable to deliver cargo to its destination and possibly causing traffic gridlock in a given railtrack. It would be further desirable to be able to provide to such personnel an inexpensive and user-friendly fault analysis kit, such as a flowchart, check list, software program, etc., that would quickly allow such personnel to compare any new fault log data downloaded from the machine with prior fault log data of the same machine so as to be able to issue accurate and reliable repair recommendations to the entity responsible for operating the locomotive.

BRIEF SUMMARY OF THE INVENTION

Generally speaking, the present invention fulfills the foregoing needs by providing a method for analyzing fault log data from railroad locomotives and other large land-based, self-powered transport equipment undergoing diagnostics. A receiving step allows for receiving fault log data comprising a plurality of faults from the equipment.

Respective executing steps allow for executing a set of noise-reduction filters upon the received fault log data to generate noise-reduced fault log data, and for executing a set of candidate snapshot anomalies upon the noise-reduced data to generate data predictive of malfunctions of the equipment. In another embodiment, the methods allows for receiving fault log data including a plurality of faults from the equipment and for receiving operational parameter data including a plurality of operational parameters from the equipment. A executing step allows for executing a set of candidate snapshot anomalies upon the fault log data and upon the operational parameter data. A combining step allows for combining any candidate snapshot anomalies respectively triggered by the fault log data with any candidate snapshot anomalies respectively triggered by the parameter data to generate data predictive of malfunctions of the equipment.

In yet another embodiment, the present invention fulfills the foregoing needs by providing a process for analyzing fault log data from railroad locomotives and other large land-based, self-powered transport equipment to identify respective faults and/or fault combinations predictive of equipment malfunctions. The process allows for downloading new fault log data from the equipment. The process further allows for retrieving prior fault log data of the equipment. The prior fault log data may be obtained during an earlier download relative to a present download of new fault log data. A comparing step allows for comparing the new fault log data against the prior fault log data, and an adjusting step allows for adjusting any repair recommendations for the earlier download of fault log data based upon the comparison of the new fault log data and the prior fault log data. In another aspect of this invention, it will be appreciated that the foregoing process may be used for developing a fault analysis kit, either in electronic form suitable for computerized processing or otherwise, e.g., a check list, flowchart, instruction chart, software program, etc., that enables respective users to systematically and accurately analyze the fault log data from the equipment so as to be able to identify the respective faults and/or fault combinations which are predictive of malfunctions of the equipment. The present invention further fulfills the foregoing needs by providing a system for analyzing fault log data and operational parameter data from railroad locomotives and other large land-based, self-powered transport equipment undergoing diagnostics. The system includes a module for receiving fault log data including a plurality of faults from the equipment. A memory unit is configured to store a set of noise-reduction filters and a set of candidate snapshot anomalies. A processor is respectively coupled to the module for receiving fault log data and to the memory unit. The processor includes processor means for executing the set of noise-reduction filters upon the received fault log data to generate noise- reduced fault log data, and further includes processor means for executing the set of candidate snapshot anomalies upon the noise-reduced data to generate data predictive of malfunctions of the equipment.

BRIEF DESCRIPTION OF THE DRAWINGS

The features and advantages of the present invention will become apparent from the following detailed description of the invention when read with the accompanying drawings in which:

FIG. 1 is one embodiment of a block diagram of a system of the present invention that uses a processor for processing operational parameter data and fault log data from railroad locomotives and other large land-based, self-powered transport equipment and diagnosing malfunctioning equipment;

FIG. 2 is an illustration of exemplary repair log data; FIG. 3 is an illustration of exemplary fault log data; FIG. 4 is an illustration of exemplary hybrid data including in part fault log data and operational parameter data and further including exemplary noise reduction filters and candidate snapshot anomalies used to analyze the hybrid data;

FIG. 5 is a flow chart illustrating one exemplary embodiment for using the noise reduction filters and the candidate snapshot anomalies; FIG. 6 illustrates further details regarding the processor of

FIG. 1;

FIG. 7 is a flowchart describing steps for selecting a respective repair for a predicted malfunction upon analysis of the fault data and/or the operational parameter data; FIG. 8 is flow chart describing steps for generating a plurality of respective cases, including predetermined repairs, fault cluster combinations and operational parameter observations for each case; FIG. 9 is a flowchart describing the steps for adding a new case to the case database and updating the weighted repair, distinct fault cluster combinations and respective weights for the candidate snapshot anomalies; FIG. 10 is a flow chart of an exemplary embodiment of the process of the present invention for analyzing fault log data so as to identify respective faults and/or fault combinations predictive of equipment malfunctions;

FIG. 11 is a flow chart illustrating further details in connection with the process of FIG. 10; and

FIG. 12 is flow chart describing steps for generating a plurality of respective cases, including predetermined repairs, fault cluster combinations and operational parameter observations for each case.

DETAILED DESCRIPTION OF THE INVENTION FIG. 1 diagrammatically illustrates one embodiment of a diagnostic system 10 of the present invention. System 10 provides a process for automatically harvesting or mining repair data comprising a plurality of related and unrelated repairs and fault log data comprising a plurality of faults, from one or more machines, such as railroad locomotives and other large land-based, self-powered transport equipment, and generating weighted repair and distinct fault cluster combinations which are diagnostically significant predictors to facilitate analysis of new fault log data from a malfunctioning locomotive. In one aspect of the invention, system 10 allows for hybridly analyzing the fault log data jointly with operational parameters from the machine.

Although the present invention is described with reference to a locomotive, system 10 can be used in conjunction with any machine in which operation of the machine is monitored, such as a chemical, an electronic, a mechanical, or a microprocessor machine. Exemplary system 10 includes a processor 12 such as a computer (e.g., UNIX workstation) having a hard drive, input devices such as a keyboard, a mouse, magnetic storage media (e.g., tape cartridges or disks), optical storage media (e.g., CD-ROMs), and output devices such as a display and a printer. Processor 12 is operably connected to and processes data contained in a repair data storage unit 20 and a fault log data storage unit 22. Processor 12 is further respectively connected to and processes noise-reduction filters stored in a storage unit 27, and candidate snapshot anomalies stored in a storage unit 28.

Repair data storage unit 20 includes repair data or records regarding a plurality of related and unrelated repairs for one or more locomotives. FIG. 2 shows an exemplary portion 30 of the repair data contained in repair data storage unit 20. The repair data may include a customer identification number 32, a locomotive identification or unit number 33, the date 34 of the repair, the repair code 35, a repair code description 36, a description of the actual repair 37 performed, etc.

Fault log data storage unit 22 includes fault log data or records regarding a plurality of faults occurring prior to the repairs for the one or more locomotives. FIG. 3 shows an exemplary portion 40 of the fault log data contained in fault log data storage unit 22. The fault log data may include a customer identification number 42, a locomotive identification number or unit 44, the date 45 when the fault occurred, a fault code 46, a fault code description 48, etc.

As suggested above, additional data used in the analysis of the present invention include operational parameter data indicative of a plurality of operational parameters or operational conditions of the machine. The operational parameter data may be obtained from various sensor readings or observations, e.g., temperature sensor readings, pressure sensor readings, electrical sensor readings, engine power readings, etc. Examples of operational conditions of the machine may include whether the locomotive is operating in a motoring or in a dynamic braking mode of operation, whether any given subsystem in the locomotive is undergoing a self-test, whether the locomotive is stationary, whether the engine is operating under maximum load conditions, etc. It will be appreciated by those skilled in the art that the repair data storage unit, the fault log data storage unit, and the operational parameter data storage unit may respectively contain repair data, fault log data and operational parameter data for a plurality of different locomotives. It will be further appreciated that the operational parameter data may be made up of snapshot observations, i.e., substantially instantaneous readings or discrete samples of the respective values of the operational parameters from the locomotive. Preferably, the snapshot observations are temporally aligned relative to the time when respective faults are generated or logged in the locomotive. For example, the temporal alignment allows for determining the respective values of the operational parameters from the locomotive prior, during or after the logging of respective faults in the locomotive. The operational parameter data need not be limited to snapshot observations since substantially continuous observations over a predetermined period of time before or after a fault is logged can be similarly obtained. This feature may be particularly desirable if the system is configured for detection of trends that may be indicative of incipient failures in the locomotive. FIG. 4 shows an exemplary data file 50 that combines fault log data and operational parameter data 52, such as locomotive speed, engine water temperature, engine oil temperature, call status, etc. FIG. 4 further illustrates exemplary candidate snapshot anomalies and noise reduction filters that may be conveniently used to enhance the predictive accuracy of the algorithms of the present invention.

One example of a noise reduction filter is detection of a predetermined code, e.g., the letter S adjacent to bracket 54 in the column designated "call". In this case, this noise reduction filter allows for ignoring any faults that may have been logged while a self-test is being conducted. Another example of a noise reduction filter may be based on ignoring suspect or unreliable information, such as may occur upon detection of two or more consecutive faults within a predetermined period of time. For example, as illustrated in bracketed row 56, this noise reduction rule allows to ignore the snapshot observations of operational parameters 52 whenever any fault is logged within a predetermined period of time, e.g., t, ms, as any previously logged fault. Similarly, as illustrated in bracketed row 58, another example of a noise reduction filter would allow for ignoring snapshot observations if any fault is logged within a predetermined period of time, e.g., t₂ ms, as any previously two logged faults.

As suggested above, one key feature of the present invention allows for using candidate snapshot anomalies to process, the fault log data and/or the operational parameter data. As used herein, candidate snapshot anomalies refer to one or more conditions that may be triggered based upon deviations in the snapshot data and/or the fault log data. One example of a candidate snapshot anomaly that conveniently uses predetermined operational parameter data is illustrated in the data field entry adjacent to bracket 60. In this case, the candidate snapshot anomaly would be triggered if the engine water temperature exceeds the engine oil temperature by a predetermined temperature, e.g., T,° C, and if the water temperature is above another predetermined temperature e.g., T₂° C. Upon such conditions being met by the respective snapshot parameters, then this exemplary candidate snapshot anomaly would be triggered and would allow for declaring a cooling subsystem malfunction with a higher level of confidence than would otherwise be feasible if one were to rely on fault log data alone. It will be appreciated that using the foregoing candidate snapshot anomaly in combination with detection of one or more faults regarding the cooling subsystem will increase the probability that in fact there is a malfunction of the cooling subsystem as compared to detection of cooling subsystem faults by themselves. Another example of a candidate snapshot anomaly is illustrated by the data field entry adjacent to bracket 62. In this case, the candidate snapshot anomaly would be triggered when the oil engine temperature exceeds the water engine temperature by a predetermined temperature, e.g., T,° C, and if the oil temperature is above another predetermined temperature e.g., T₂° C. Upon being triggered this other exemplary candidate snapshot anomaly would allow for declaring a malfunction in the lubrication subsystem of the engine with a higher level of confidence than would otherwise be possible. Once again it will be appreciated that the foregoing candidate snapshot anomaly in combination with detection of one or more faults regarding the lubrication subsystem will increase the probability that in fact there is a malfunction of the lubrication subsystem as compared to detection of lubrication subsystem faults by themselves. For the sake of clarity of understanding, the foregoing examples of candidate snapshot anomalies and noise reduction filters were chosen to be relatively straightforward. However, as will be recognized by those skilled in the art, the construction of noise reduction filters and/or candidate snapshot anomalies construction may involve searching for combinations of clusters or groups of faults as well as searching for respective combinations of observations of multiple snapshot operational parameters, using the analysis techniques disclosed in the foregoing patent applications. FIG. 5 illustrates a flow chart of one exemplary embodiment of the present invention. Subsequent to start of operations in step 70, step 72 allows for receiving the fault log data. As suggested above, step 74 allows for executing a set of noise reduction filters upon the received fault log data to generate noise-reduced data. Prior to return step 78, step 76 allows for executing a set of candidate snapshot anomalies upon the noise- reduced data to generate data predictive of malfunctions of the machine. Each predicted malfunction may be correlated with the repair data using statistical correlation techniques well-understood by those skilled in the art. As illustrated in FIG. 4, the repair data may include respective repair codes 64 and may further indicate one or more corrective actions to be taken once a specific malfunction is detected. The indication, for example, may be for the operator to disengage a respective handbrake unintentionally activated, or suggest the replacement of a given replaceable unit, or in more complex situations may suggest to the operator to bring the locomotive to a selected repair site where needed specialized tools may be available to perform the repair. Preferably, prior to generating a respective repair code for a predictive malfunction, a respective repair weight should be retrieved from a directed weight data storage unit 26 (FIG. 1) to verify that the predicted malfunction and selected repair meet the respective weight assigned to the predicted malfunction or repair. It will be appreciated that the initial values for the directed weight data may be obtained based on the knowledge of experts and/or empirical data, that is, the values of the directed weight data may be initially assigned. However, as additional cases are used to populate a case data storage unit 24 (FIG. 1), the system may be configured to automatically adjust or adapt the respective values of the directed weight data based on the cumulative knowledge acquired from such additional cases. Similarly, both the noise-reduction filters and the candidate snapshot anomalies may be adapted or modified based on the cumulative knowledge extracted from the additional cases.

FIG. 6 illustrates an exemplary embodiment wherein a candidate snapshot anomaly processor module 208, which may be part of processor 12, receives fault log data 100 and operational parameter data 52 that may be optionally filtered or noise-reduced by respective noise reduction filters 202 and 204 using the noise reduction filters discussed in the context of FIG. 4. Preferably, processor module 206 receives noise- reduced data both for the fault log data and the operational parameter data. However, processor module 206 is not restricted to operating on noise- reduced data since the candidate snapshot anomalies could also be directly executed on unfiltered or raw fault log data and/or operational parameter data.

FIG. 7 illustrates a flow chart illustrating additional processing steps performed by processor module 206 regarding executing step 76 (FIG 5). For example, step 208 allows for combining candidate snapshot anomalies triggered by the fault log data with candidate snapshot anomalies triggered by the operational parameter data to generate data predictive of malfunctions of the machine. Prior to return step 212, step 210 allows for selecting at least one repair for each predicted malfunction using a plurality of weighted repairs and, as suggested above, respective combinations of distinct clusters of faults and/or observations of operational parameters.

FIG. 8 is a flowchart of an exemplary process 150 of the present invention for selecting or extracting repair data from repair data storage unit 20, fault log data from fault log data storage unit 22, and operational parameter data from operational parameter data storage unit 29 and generating a plurality of diagnostic cases, which are stored in a case storage unit 24. As used herein, the term "case" comprises a repair and one or more distinct faults or fault codes in combination with respective observations of one or more operational parameters.

With reference still to FIG. 8, process 150 comprises, at 152, selecting or extracting a repair from repair data storage unit 20 (FIG. 1). Given the identification of a repair, the present invention searches fault log data storage unit 22 (FIG. 1) to select or extract, at 154, distinct faults occurring over a predetermined period of time prior to the repair. Similarly, operational parameter data storage unit 29 (FIG. 1) may be searched to select or extract, at 155, respective observations of the operational parameter data occurring over a predetermined period of time prior to the repair. Once again, the observations may include snapshot observations, or may include substantially continuous observations that would allow for detecting trends that may develop over time in the operational parameter data and that may be indicative of malfunctions in the machine. The predetermined period of time may extend from a predetermined date prior to the repair to the date of the repair. Desirably, the period of time extends from prior to the repair, e.g., 14 days, to the date of the repair. It will be appreciated that other suitable time periods may be chosen. The same period of time may be chosen for generating all of the cases.

At 156, the number of times each distinct fault occurred during the predetermined period of time is determined. At 157, the respective values of the observations of the operational parameters is determined. A plurality of repairs, one or more distinct fault cluster and respective observations of the operational parameters are generated and stored as a case, at 160. For each case, a plurality of repair, respective fault cluster combinations, and respective combinations of clusters of observations of the operational parameters is generated at 162.

As shown in FIG. 9, a process 250 of the present invention provides for updating directed weight data storage unit 26 to include one or more new cases. For example, once a new case is generated, a new repair, fault log data, and operational parameter data from a malfunctioning locomotive is received at 252. At 254, a plurality of distinct fault cluster combinations and clusters of observations of the operational parameters is generated.

The number of times each fault cluster occurred for related repairs is updated at 256 and the number of times each fault cluster occurred for all repairs are updated at 258. Similarly, respective values of the clusters of observations of the operational parameters that triggered respective candidate snapshot anomalies for related repairs may be averaged and updated at 260 and respective values of the operational parameters that triggered respective candidate snapshot anomalies for all repairs may be averaged and updated at 262. Thereafter, the weighted repair, the distinct fault cluster combinations and the respective weight values for the candidate snapshot anomalies are redetermined at 264. For example, although the candidate snapshot anomaly may have initially postulated that if the engine water temperature exceeds the engine oil temperature by T,⁰ C, and if the water temperature is above T₂° C, then the candidate snapshot anomaly would declare a cooling subsystem malfunction. However consistent with the adaptive features of the present invention, at step 260, the learning algorithm would conveniently allow for redetermining the respective temperature values required to trigger the candidate snapshot anomaly, in view of the accumulated knowledge gained from each new case. In addition, the candidate snapshot anomalies themselves could be modified to add observations of new parameters or delete observations from parameters that were initially believed to be statistically meaningful but in view of the cumulative knowledge acquired with each new case are proven to be of little value for triggering a respective candidate snapshot anomaly, i.e., equivalent to a "Don't Care" variable in Boolean logic. As suggested above, further analysis of the repair data could indicate that ambient temperature may be another parameter that could aid the candidate snapshot anomaly to trigger more accurately the prediction of malfunctions of the cooling subsystem.

As noted above, the system provides prediction of malfunctions and repair selection from hybrid analysis of fault log data and operational parameter data from a malfunctioning machine. Desirably, after verification of the repair(s) for correcting a malfunction the new case can be inputted and updated into the system.

From the present invention, it will be appreciated by those skilled in the art that the repair, respective fault cluster combinations and observations of operational parameters may be generated and stored in memory when generating the weights therefor, or alternatively, be stored in either the case data storage unit, directed weight storage unit, or a separate data storage unit.

Thus, the present invention provides in one aspect a method and system for automatically harvesting potentially valid diagnostic cases by interleaving repair, fault log data which is not restricted to sequential occurrences of faults or error log entries and operational parameter data that could be made up of snapshot observations and/or substantially continuous observations. In another aspect, standard diagnostic fault clusters and suitable candidate snapshot anomalies using operational parameters and/or fault data can be generated in advance so they can be identified across all cases and their relative occurrence tracked.

The present invention further allows use of noise reduction filters that may enhance the predictive accuracy of the diagnostic algorithms used therein. In yet another aspect, the invention allows for readjusting the assigned weights to the repairs, the candidate snapshot anomalies and the noise reduction filters based on extracting knowledge that is accumulated as each new case is closed.

In addition, when initially setting up case data storage unit 24, a field engineer may review each of the plurality of cases to determine whether the collected data, either fault log data and/or operational parameter data, provide a good indication of the repair. If not, one or more cases can be excluded or removed from case data storage unit 24. This review by a field engineer would increase the initial accuracy of the system in assigning weights to the repair, candidate snapshot malfunctions and fault cluster combinations.

FIG. 10 shows a flow chart of an exemplary embodiment of a process 350 for analyzing fault log data so as to avoid missing detection or identification of fault log data which is statistically and probabilistically relevant to early and accurate prediction of machine malfunctions. Upon start of operations at step 352, step 354 allows for downloading new fault log data from the machine. Step 356 allows for verifying predetermined identification parameters of the newly downloaded fault log data so as to avoid unintentionally attributing faults to the wrong locomotive. Exemplary identification parameters may include road number, time of download, time fault was logged, etc. For example, this step may allow for verifying that the road number in a previously downloaded fault log actually matches the road number of the locomotive fault log presently intended to be downloaded and may further allow for verifying that the date and time in the fault log matches the present date and time. Step 358 allows for retrieving prior fault log data of the machine. The prior fault log may be obtained during an earlier download, such as the last download executed prior the download of step 354. As described in greater detail in the context of FIG. 11 below, step 360 allows for comparing the new fault log data against the prior fault log data. Prior to return step 364, step 362 allows for adjusting any repair recommendations for the earlier download of fault log data based upon the comparison of the new fault log data and the prior fault log data.

FIG. 11 is a flowchart that illustrates further details regarding process 350 (FIG. 10). Subsequent to start step 370, step 372 allows for determining whether any new faults have occurred since the last download. If new faults have not been logged since the last download, then step 374 allows for reviewing and updating the last repair recommendation. If new faults were logged at step 372, then step 376 allows for determining whether any of the new faults are repeats of the previously logged faults, e.g., faults that previously required a recommendation. If there are repeat faults, then, as suggested above, step 374 would allow for reviewing and updating the last repair recommendation. If there are no repeat faults, then step 380 allows for determining if the newly downloaded faults are related to any previously logged faults. By way of example and not of limitation, related faults generally affect the same machine subsystem, such as power grid faults and dynamic braking faults, both generally related to the dynamic braking subsystem of the locomotive. If the newly downloaded faults are related to previously logged faults, then once again, step 374 would allow for reviewing and updating the last repair recommendation. Step 382 allows for determining whether there are any active faults. If there are active faults, then step 384 allows for assigning a respective repair action. For example, the repair assignment may require to determine if the locomotive engineer should reset the faults, or if the locomotive should be checked first by one or more repair specialists. By way of example, any open or non-reset faults will show 0.00 in the reset column. An externally-derived set of instructions, such as may be contained in a fault analysis electronic database or hardcopy may be conveniently checked so as to determine whether any given fault is the type of fault that could result in locomotive damage if reset prior to conducting detailed investigation as to the cause of that fault. If no faults are active, then step 386 allows for conducting expert analysis on the fault. By way of example and not of limitation, the expert analysis may be performed by teams of experts who preferably have a reasonably thorough understanding of respective subsystems of the locomotive and their interaction with other subsystems of the locomotive. For example, one team may address fault codes for the traction subsystem of the locomotive. Another team may address faults for the engine cooling subsystem, etc. As suggested above, each of such teams may also interact with the diagnostics experts in order to insure that the newly identified faults and/or respective combinations thereof are fully compatible with any of the diagnostics techniques used for running diagnostics on any given locomotive. FIG. 12 is a flowchart of an exemplary process 450 for selecting or extracting repair data from repair data storage unit 20, fault log data from fault log data storage unit 22, and operational parameter data from operational parameter data storage unit 29 and generating a plurality of diagnostic cases, which are stored in a case storage unit 24. As used herein, the term "case" comprises a repair and one or more distinct faults or fault codes in combination with respective observations of one or more operational parameters.

With reference still to FIG. 12, process 450 comprises, at 452, selecting or extracting a repair from repair data storage unit 20 (FIG. 1). Given the identification of a repair, the present invention searches fault log data storage unit 22 (FIG. 1) to select or extract, at 454, distinct faults occurring over a predetermined period of time prior to the repair. Similarly, operational parameter data storage unit 29 (FIG. 1) may be searched to select or extract, at 455, respective observations of the operational parameter data occurring over a predetermined period of time prior to the repair. Once again, the observations may include snapshot observations, or may include substantially continuous observations that would allow for detecting trends that may develop over time in the operational parameter data and that may be indicative of malfunctions in the machine. The predetermined period of time may extend from a predetermined date prior to the repair to the date of the repair. Desirably, the period of time extends from prior to the repair, e.g., 14 days, to the date of the repair. It will be appreciated that other suitable time periods may be chosen. The same period of time may be chosen for generating all of the cases.

At 456, the number of times each distinct fault occurred during the predetermined period of time is determined. At 457, the respective values of the observations of the operational parameters is determined. A plurality of repairs, one or more distinct fault cluster and respective observations of the operational parameters are generated and stored as a case, at 460. For each case, a plurality of repair, respective fault cluster combinations, and respective combinations of clusters of observations of the operational parameters is generated at 462. As suggested above, the present invention provides in one of its aspects, a process and system for developing a fault log data analysis kit that enables users to analyze fault log data from a machine so as to identify faults and/or fault combinations predictive of machine malfunctions. It will be appreciated by those skilled in the art that the kit may be deployed in any suitable form, electronic or otherwise, e.g., a flowchart, a checklist, a computer program product configured in any suitable computer-usable medium and having a computer-readable code therein for executing the respective process steps discussed above in the context of FIGS. 10 and 11, etc. By way of example, the output of the fault analysis kit and/or process of the present invention could be used for opening respective cases in the case data storage unit 24, such as during situations when system 10 may be unavailable, for example, due to maintenance and/or upgrading. Thus, in another aspect of the present invention, the tool analysis kit of the present invention may provide backup to system 10 as well as enhance any CBR analysis performed on the fault log data by system 10.

While the preferred embodiments of the present invention have been shown and described herein, it will be obvious that such embodiments are provided by way of example only. Numerous variations, changes and substitutions will occur to those of skill in the art without departing from the invention herein. Accordingly, it is intended that the invention be limited only by the spirit and scope of the appended claims.

Claims

WHAT IS CLAIMED IS

1. A method for analyzing fault log data from railroad locomotives and other large land-based, self-powered transport equipment undergoing diagnostics, the method comprising: receiving (72) fault log data comprising a plurality of faults from the equipment; executing (74) a set of noise-reduction filters upon the received fault log data to generate noise-reduced fault log data; and executing (76) a set of candidate snapshot anomalies (28) upon the noise-reduced data to generate data indicative of malfunctions of the equipment.

2. The method of claim 1 further comprising receiving operational parameter data (25) comprising a plurality of snapshot observations of operational parameters from the equipment, executing a set of noise reduction filters (27) upon the operational parameter data so as to generate noise-reduced operational parameter data and executing the set of candidate snapshot anomalies (28) upon the noise-reduced operational parameter data.

3. The method of claim 2 further comprising a step

(208) of considering any candidate snapshot anomalies respectively triggered by the noise-reduced fault log data in light of any candidate snapshot anomalies respectively triggered by the noise-reduced parameter data so as to enhance the accuracy of the data indicative of malfunctions of the equipment.

4. The method of claim 1 further comprising selecting (210) at least one repair for each indicated malfunction using a plurality of respective weighted repairs, and respective combinations of distinct clusters of faults.

5. The method of claim 4 further comprising selecting (210) at least one repair for each indicated malfunction using a plurality of respective weighted repairs, and respective combinations of distinct clusters of faults and or operational parameters.

6. A method for analyzing fault log data and operational parameter data from railroad locomotives and other large land- based, self-powered transport equipment undergoing diagnostics, the method comprising: receiving fault log data (200) comprising a plurality of faults from the equipment; receiving operational parameter data (25) comprising a plurality of snapshot observations of operational parameters from the equipment; executing (76) a set of candidate snapshot anomalies upon the fault log data and upon the operational parameter data; and considering (208) any candidate snapshot anomalies (28) respectively triggered by the fault log data in light of any candidate anomalies respectively triggered by the parameter data to generate data predictive of malfunctions of the equipment.

7. The method of claim 6 wherein prior to executing the set of candidate snapshot anomalies a step of executing a set of noise- reduction filters (27) is performed upon the fault log data and the operational parameter data.

8. The method of claim 6 wherein the considering of data comprises comparing the fault log data and parameter data.

9. The method of claim 6 wherein the considering of data comprises combining the fault log data and parameter log data.

10. A system for analyzing fault log data and operational parameter data from railroad locomotives and other large land-based, self- powered transport equipment undergoing diagnostics, the system comprising: a module for receiving fault log data (200) comprising a plurality of faults from the equipment; a memory unit (27) configured to store a set of noise- reduction filters and a set of candidate snapshot anomalies; and a processor (12) respectively coupled to the module for receiving fault log data and to the memory unit, the processor comprising: a processor for executing (74 or 202) the set of noise- reduction filters upon the received fault log data to generate noise-reduced fault log data; and a processor for executing (76) the set of candidate snapshot anomalies upon the noise-reduced data to generate data predictive of malfunctions of the equipment.

11. The system of claim 10 further comprising a memory unit receiving operational parameter data (25) comprising a plurality of snapshot observations of respective operational parameters from the equipment; a processor for executing (204) the set of noise reduction filters upon the operational parameter data so as to generate noise-reduced operational parameter data; and a processor for executing (206) the set of candidate snapshot anomalies upon the noise-reduced operational parameter data.

12. The system of claim 10 further comprising a processor for considering (208) any candidate snapshot anomalies respectively triggered by the noise-reduced fault log data in light of any candidate snapshot anomalies respectively triggered by the noise-reduced parameter data so as to enhance the accuracy of the data indicative of malfunctions of the equipment.

13. A process for analyzing fault log data from railroad locomotives and other large land-based, self-powered transport equipment so as to identify respective faults and or fault combinations indicative of machine malfunctions, the process comprising: a) collecting a set of (354) new fault log data from the equipment; b) retrieving (358) prior fault log data of the equipment held in memory, the prior fault log data having been collected at an earlier time than the data collected in step a); c) considering (360) the new fault log data in light of the prior fault log data; and d) adjusting (362) any repair recommendations based on the prior fault log data in light of the consideration of the new fault log data.

14. The process of claim 13 wherein the considering of data comprises determining (372) respective occurrences of any new faults since the last collection of data.

15. The process of claim 13 wherein the considering of data comprises determining (376) respective occurrences of any repeat faults.

16. The process of claim 13 wherein the repair recommendation is based upon considering respective repair entries in an externally-derived file of repairs, each fault being associated with at least one or more repairs.

17. The process of claim 13 wherein the fault log data further comprises snapshot observations of predetermined operational parameters from the equipment.

18. A system for analyzing fault log data from railroad locomotives and other large land-based, self-powered transport equipment to identify respective faults and/or fault combinations indicative of equipment malfunctions, the system comprising: memory for collecting and storing (354) new fault log data from the equipment; a processor for retrieving (358) prior fault log data of the equipment, the prior fault log data obtained during an earlier collection of data relative to the collection of new fault log data; a process for considering (360) the new fault log data in light of the prior fault log data; and a processor for adjusting (362) any repair recommendations for the earlier fault log data based in light of the consideration of the new fault log data.

19. The system of claim 18 further comprising memory for an externally-derived file of repairs, each active fault being associated with at least one or more repairs.