System for monitoring operation and maintenance of industrial equipment

Abstract

The invention relates to a system (1) for monitoring the operation and maintenance of a device (2) in a regulated facility (3), wherein the device (2) operates until a maintenance (6) to be performed and has at least one subsequent planned maintenance (7), generating a manufacturing and maintenance log (9), a usage log (11), a log (120) of the state (13). The correlation (14) between the causes and consequences of the aging of the device (2) is determined by characterizing the tasks (90) and conditions (110) that affect the state (13). For other devices, data corresponding to said correlation (14) are recovered and extracted in order to train a virtual model (16). On the day before the maintenance (6) to be performed, based on these logs (9 11), the tasks (90) and the planned conditions (110) of the scenario (8) of the repair (6) to be performed, said model (16) generates a planned state (130) compared to the minimum operating state (17) of the device (2).

Description

[0001] This invention pertains to the field of industrial equipment supervision and aims to manage the lifecycle of industrial equipment.

[0002] - Optimize and adjust the manufacturing of industrial equipment for given usage scenarios;

[0003] - Optimize and adjust the maintenance of industrial equipment to meet expected usage conditions;

[0004] - Determine the maintenance of industrial equipment, thereby optimizing maintenance throughout the long service life of the industrial equipment;

[0005] - Identify usage that meets the criteria of zero failures (no faults) during operation until the next maintenance;

[0006] - Identify and ensure zero-failure operation throughout the long service life of industrial equipment;

[0007] - Provides instructions and optimizations for extending the lifespan of industrial equipment based on anticipated usage scenarios. Technical Field

[0008] This invention is preferably, but not limitingly, applied to the regulation of medium- to high-risk industrial equipment and facilities, because the equipment in question needs to meet reliability requirements, especially when both reliability and safety requirements need to be met simultaneously. The equipment specifically includes:

[0009] - Equipment whose unavailability disrupts the service or production of the entire facility to which it belongs;

[0010] - Equipment that causes or exacerbates accidental or unforeseen circumstances in the related facilities due to equipment unavailability.

[0011] Therefore, this invention can be applied not only to the transportation sector (road, air, sea, and rail), but also to the processing industry sector ("community service" utilities, such as the production of electricity, water, and fuel) or the defense sector.

[0012] Therefore, in the context of this invention, the general term "equipment" includes components of a facility, functional components that are essential to the proper operation and / or safety of the facility.

[0013] Therefore, the equipment under discussion consists of components of mobile facilities, such as machinery or vehicles (e.g., submersibles, ships, aircraft or spacecraft), or components of fixed facilities, such as buildings or infrastructure (e.g., power plants, oil platforms or refineries).

[0014] Because this invention provides superior reliability guarantees in manufacturing, maintenance, and operation, it meets the high expectations of industrial participants involved in the value chain of such equipment, particularly designers, manufacturers, maintenance technicians, and operators.

[0015] The reliability of the equipment under discussion is crucial because equipment failure can lead to the shutdown of related facilities. The reliability of the equipment becomes even more critical when it is a safety requirement of the facility to which it belongs.

[0016] In this case, the present invention makes the following choice: the reliability of the device depends on the manufacturing quality of the device, the past maintenance and use of the device, and the expected future use of the device.

[0017] Therefore, the issue of equipment reliability highlights all the benefits of controlling how the design, manufacture, maintenance, and use of the equipment affect its reliability.

[0018] Therefore, it is necessary to answer the following questions:

[0019] - Considering the intended use of the equipment, which tasks (gestures, parts, and settings) used to manufacture the equipment can optimize the equipment's behavior during operation?

[0020] - Which equipment maintenance tasks should be performed to achieve the best ratio between maintenance costs and the probability of zero failures until the next maintenance (considering the expected future use of the equipment and the equipment's own manufacturing, maintenance, and usage logs)?

[0021] - How will the observed equipment perform under extreme usage conditions until the next maintenance, i.e., how will it meet the maximum usage condition of zero failures until the next maintenance?

[0022] Another useful aspect is consistently determining the optimal task for each maintenance operation from the remaining lifespan of the equipment and its remaining service life (considering future usage and the equipment's own logs). In practice, this allows for:

[0023] - Customize maintenance schedules for each device in this series;

[0024] - Updates and anticipated availability of spare parts (nature of parts and replacement date);

[0025] - Update and expected industrial maintenance schedule;

[0026] - The total cost of updates and expected future maintenance over the lifespan of the equipment.

[0027] Another useful approach is to always be certain:

[0028] - The equipment's compliance will result in zero-failure limit use observed during every future operating cycle of the equipment's lifespan;

[0029] - Remaining useful life of the device (taking into account the future usage and the device's own logs).

[0030] In practice, this allows for actual usage planning of the equipment based on its remaining lifespan, thus enabling long-term use.

[0031] Furthermore, because it is possible to update and anticipate the equipment's maximum usage over the remainder of its lifespan, as well as the total cost of future optimal maintenance operations, the residual value of the equipment can always be determined from the perspective of potential resale.

[0032] In short, the ability to control the impact of equipment manufacturing, maintenance, and use on the equipment's reliability is crucial because such control can at least:

[0033] - Enable pathways to optimize manufacturing and maintenance;

[0034] - Facilitate interaction with industry maintenance stakeholders;

[0035] - Enable pathways to control the reliability of the equipment (by eliminating uncertainties about reliability during operation);

[0036] - Enables ways to control and optimize the management of the assets represented by the device. Background Technology

[0037] To date, the existing methods that best influence equipment reliability are:

[0038] -Predictive maintenance;

[0039] -Physical model;

[0040] - Authentication procedures on the testing platform;

[0041] - Feedback (usually abbreviated as "FB").

[0042] However, these methods cannot provide the simulations needed to understand the impact of manufacturing, maintenance, and use of control equipment on its reliability. In this sense, these methods only partially meet the manufacturer's expectations.

[0043] As for predictive maintenance, this technology observes and analyzes equipment operation to detect degradation. Once degradation is detected, the technology constructs an initial trend of degradation development to show an initial prediction of the future failure date. Based on subsequent observations, this trend and failure date are repeatedly fine-tuned.

[0044] In doing so, predictive maintenance involves several methodological biases:

[0045] Predictive maintenance attempts to infer the future behavior of a device from observed past behavior, thereby trying to infer the outcome of future aging without considering the causes of past and future aging;

[0046] Predictive maintenance assumes that future usage will be the same as past usage; this is almost always wrong.

[0047] Predictive maintenance uses only runtime data and does not take into account functional assessments.

[0048] Therefore, predictive maintenance has the following characteristics:

[0049] - Initial predictions are vague and slow, requiring approximately fifteen days of observation time after maintenance operations;

[0050] - The speed at which fair predictions are displayed is slow;

[0051] - Short-sighted: Predictive maintenance only covers a medium timeframe, at most about six months.

[0052] In terms of other consequences resulting from methodological biases in predictive maintenance, it lacks foresight. In reality, predictive maintenance only reacts to degradation that occurs during operation, rather than predicting its occurrence. Furthermore, predictive maintenance requires predictions to be made while the equipment is operational, which is an unacceptable operational defect. More generally, predictive maintenance does not require any simulations, particularly simulations of the equipment's behavior based on its manufacturing or maintenance practices and planned use.

[0053] Therefore, predictive maintenance is limited to providing predictions of failure dates with insufficient accuracy and foresight. Predictive maintenance cannot offer manufacturing, maintenance, or usage recommendations to optimize equipment behavior, especially to postpone failure dates or extend service life.

[0054] With the emergence of computational methods and the increasing value of data, various predictive maintenance solutions have arisen. However, these solutions rely solely on operational data, are subject to the same methodological biases, and therefore suffer from the same limitations, making them virtually indistinguishable from one another.

[0055] As for physical models, from a technical, and especially economic, perspective, it is impractical to fully and customarily model a given device using physical models. These models are extremely expensive and cover an overly restricted area of ​​physical phenomena. Furthermore, these physical models are often vague.

[0056] In reality, physical models can only characterize the behavior of basic components of a device exposed to physical phenomena. Furthermore, this modeling is an approximation of reality: it includes inaccurate elements. The development of physical models is linked to fundamental research and involves significant time and cost resources. Moreover, to characterize the aging of each basic component in the device, it is advisable to generate multiple physical models, approximately several hundred, before achieving a complete model of the device.

[0057] Therefore, modeling the behavior of this series of standard devices using physical models requires significant cost and time. Furthermore, this holistic modeling based on these physical models is difficult to customize for each individual device, especially for the complete logs of each device.

[0058] As for the bench tests conducted during the design of a given series of benchmark devices, particularly during the qualification of those devices, these tests are theoretically capable of approximating the behavior of that series of benchmark devices throughout their entire service life under one or more constraints simulating use and aging. These tests provide trends and orders of magnitude; therefore, the indication is only fuzzy.

[0059] Regardless of the specificity of a given device in this series that is tested individually (that is, without considering the manufacturing, maintenance, and usage logs of a given device that these tests cannot actually predict), these tests cannot actually characterize the future behavior of the device in response to future use of the device, and in particular cannot determine the parameters of maintenance to be performed that could be used to optimize the behavior of the device during operation.

[0060] In practice, bench testing involves characterizing the behavior (e.g., steady-state operating point or transient response) of a benchmark device in the series under one or more given usage and aging conditions: then, a sample device representing the series is subjected to accelerated aging, which is assumed to represent actual aging. Because aging is accelerated during testing, and the usage considered during testing is independent of the actual usage of each device in the series during its respective service life, the characterization of behavior remains ambiguous.

[0061] Therefore, the behavior of the benchmark devices in this series, especially their lifespan, is only an approximation.

[0062] As for feedback, often abbreviated as "FB," its completion is based on an approximate estimate of trends and orders of magnitude from bench testing. FB also fails to account for the specificity of a given device in the series, specifically the complete logs of that given device, or therefore characterizes the future behavior of the device in question in response to given future use, and in particular, cannot determine the maintenance parameters that will optimize the operating behavior of that given device. Data from FB is also insufficient due to its unsystematic and irregular operation.

[0063] In reality, Facebook compares average behavior (such as lifetime) with average usage. This approach does not yield precise data.

[0064] Furthermore, Facebook is not being used systematically. When it is used, it is irregular, and the refresh frequency is too low, with a cycle of approximately one to two years.

[0065] The purpose of this invention is to overcome the shortcomings of the prior art by proposing a digital system for monitoring the operation and maintenance of at least one industrial device within a facility, capable of generating and using behavioral models specific to that series of devices.

[0066] The generated behavioral model is meaningful:

[0067] -It links the causes and consequences of equipment aging;

[0068] -Its specificity is due to the physical phenomena present in each device within this series;

[0069] - It can be customized for each device in this series;

[0070] - It has the accuracy of an empirical model trained with observed data (“data-driven” model).

[0071] The generated behavioral model is powerful: it can simulate the behavior of a series of devices for a given maintenance decision and plan, especially taking into account the device's manufacturing, maintenance, and usage logs. The model is even more powerful because it can perform this type of simulation over the long lifespan of the device.

[0072] For a given device in the series under discussion, taking into account the device's intended planned use and its manufacturing, maintenance, and usage logs, the present invention is therefore able to provide an overall value proposition, particularly:

[0073] - For a given maintenance procedure for the equipment, the system can determine the optimal maintenance decision. This decision achieves an optimal balance between maintenance constraints and the likelihood of zero failures for the equipment until the next maintenance, as well as the expected planned usage. This optimal decision is also customized based on the equipment's manufacturing, maintenance, and usage logs;

[0074] -Based on the operating cycle of the equipment during its service life, the system is able to determine the limit of use, that is, the maximum possible use with zero failures until the next maintenance;

[0075] - Taking into account the actual use of the equipment since its last maintenance and the expected planned use until the next maintenance, the system is able to update the equipment's maximum usage during the operation of the facility;

[0076] Similarly, the system is able to always determine the optimal future maintenance schedule for the equipment for the remainder of its service life, as well as the limits of use that the equipment should adhere to for each operating cycle for the remainder of its service life.

[0077] - This system always determines the end date of the device's service life;

[0078] Finally, for a given planned use scenario, especially one that depends on the use segment that intersects with the geographical use segment, the system is able to determine manufacturing tasks (gestures, parts, and settings) to optimize the lifespan of the baseline equipment in the series.

[0079] To this end, the present invention reasoned about the scale of a series of devices. For a given series, the present invention first determines the correlation between the causes and consequences of device aging in the series under discussion. This correlation associates the manufacturing and maintenance logs, as well as the usage logs, of a given device in the series with the state of the device resulting from these logs. The determined correlation is specific to the physical phenomena present in each device within the series. Secondly, the present invention models this correlation in the form of a virtual model trained using manufacturing, maintenance, usage, and state data of the devices in the series under discussion. Thirdly, the present invention uses this model as a simulator tailored to each device in the series: this model is effectively capable of simulating the state of the device at a given moment to make maintenance decisions and simulating usage after that moment, taking into account the manufacturing and maintenance logs, as well as the usage logs specific to that device and prior to the maintenance under discussion. This model enables the realization of the value proposition of the present invention.

[0080] Furthermore, by modeling the behavior of the standard equipment in this series, the present invention correlates the observed behavior of the equipment during the manufacturing, maintenance, and use of each device at the series scale: thus, the present invention is a method using feedback. The advantage of this feedback lies in modeling the functional correlations of related aspects. The present invention also enables the use of system feedback in a continuous or quasi-continuous manner by training a model based on updated manufacturing, maintenance, use, and status data of the equipment in the series under discussion, according to the most suitable period determined by the designer, manufacturer, maintenance technician, and operator.

[0081] Therefore, the present invention relates to a digital system for monitoring the operation and maintenance of at least one industrial device within a facility, the monitoring being performed by at least one computing terminal.

[0082] Advantageously, the system includes at least the following steps:

[0083] - Install at least one industrial device produced by the manufacturing process and representing a series within the facility, and then operate the device for at least the period prior to the maintenance steps to be performed;

[0084] - After at least one scenario where the device is scheduled to be used within the planned operating cycle, the scheduled maintenance is limited to at least one maintenance following the maintenance to be performed.

[0085] This system is characterized in that the manufacture, installation, operation, and maintenance of the equipment produce at least the following:

[0086] - Manufacturing and maintenance logs, which include:

[0087] The task of manufacturing the at least one device up to the installation;

[0088] Optionally, at least one previous maintenance task of the device,

[0089] - The usage log of the device during the period between installation and the maintenance to be performed, the usage log including the usage conditions of the device during the period;

[0090] - The status log of the device, the status log including the material specifications of the device;

[0091] in

[0092] - By conducting a technical analysis of the equipment, at least one correlation is determined between at least one of the tasks and / or at least one of the usage conditions and at least one of the material indicators of the state, and the correlation establishes at least one link between the causes and consequences of aging of the equipment.

[0093] In this correlation:

[0094] - The task is characterized by those tasks identified as critical, and / or the conditions of use are characterized by those conditions to which the device is sensitive and exposed during operation or when stopped, the tasks and conditions in question affecting the state of the device;

[0095] - The material condition of the equipment is characterized by being identified as an indicator representing that condition of the equipment;

[0096] Then, within this correlation, the following is determined:

[0097] - A physical quantity or a function of the measured physical quantity that characterizes the device's sensitivity to and exposure to the conditions of use during operation or when it is stopped;

[0098] - A measured physical quantity or a function of the measured physical quantity that characterizes the material state of the device at a given moment;

[0099] Then

[0100] - For other devices in the series, recover and extract data related to these tasks and data associated with physical quantities or functions of physical quantities relating to these usage conditions and these material parameters, as identified in the correlation, in order to obtain a dataset;

[0101] - Train at least one virtual model based on the dataset.

[0102] The system is also characterized by

[0103] During the maintenance to be performed on the device, the following values ​​will be submitted to the model:

[0104] - At least one of the tasks of creating and maintaining logs, and at least one of the conditions for using logs.

[0105] - and at least one of the maintenance tasks to be performed and at least one of the planned usage conditions of the scenario;

[0106] The model generates a planned state for the device after the maintenance to be performed, and the planned state is compared with the minimum state identified as required for the operation of the device. Attached Figure Description

[0107] Other features and advantages of the invention will become apparent from the following detailed description of non-limiting embodiments of the invention with reference to the accompanying drawings, in which:

[0108] [ Figure 1 The diagram schematically depicts a view of a system architecture implemented under the supervision of a device, specifically showing that the device's manufacturing and maintenance logs, as well as usage logs, along with planned usage conditions for maintenance tasks and scenarios, are submitted to a trained virtual model the day before maintenance is scheduled. The virtual model then generates the planned status of the device on the next scheduled maintenance date.

[0109] [ Figure 2 The diagram schematically depicts a detailed view of the architecture, with particular emphasis on the manufacturing and maintenance log, which includes manufacturing decisions up to the installation of the equipment, as well as previous maintenance decisions for the equipment. It also includes maintenance decisions to be performed and decisions to be scheduled for maintenance, with the task of each decision and their timing highlighted in particular.

[0110] [ Figure 3The diagram schematically depicts a detailed view of the architecture, with particular emphasis on the device's usage logs and planned usage scenarios between installation and maintenance to be performed. The usage logs and scenarios highlight the usage conditions of the device associated with their respective cycles.

[0111] [ Figure 4 The diagram schematically depicts a detailed view of the architecture, particularly showing the log of the device's continuous status and the device's continuous planned status, with material indicators of these statuses associated with the corresponding cycles of the device status highlighted.

[0112] [ Figure 5 The diagram schematically depicts a view of the details of the regulatory system architecture, particularly illustrating the process of recovering and extracting data from a single device in the series, and the process of determining the correlations to be trained.

[0113] [ Figure 6 The diagram schematically depicts another detailed view of the architecture, specifically illustrating the process of recovering and extracting data for several devices in the series to obtain a dataset and train a virtual model;

[0114] [ Figure 7 The diagram schematically depicts a view of the regulatory system architecture, specifically showing the planned state of the device generated by the virtual model at a certain moment, along with its two material indicators and a comparison with the minimum state required for the operation of the device.

[0115] [ Figure 8 This diagram illustrates a calculation example of the limit of the planned conditions for a usage scenario following the maintenance to be performed, where the planned state of the device is equal to the lowest state at the next planned maintenance.

[0116] [ Figure 9 This schematically depicts another view of the example of another calculation for another limit of another planned condition in a usage scenario following the maintenance to be performed, for which the planned state of the device is equivalent to the lowest state at the next planned maintenance;

[0117] [ Figure 10 The view schematically depicts the proportions of the values ​​with certain planned use conditions in the form of polygons, specifically indicating the target values ​​and maximum limits of the planned conditions associated with the scenario;

[0118] [ Figure 11 The view of the example is schematically depicted using a set of curves on a graph, particularly highlighting the maximum limit of the third planning condition through the variation of the two planning conditions in the planning conditions;

[0119] [ Figure 12 [Illustratively depicting something similar to] Figure 1 The view of the long-term architecture specifically illustrates the first step of cyclically implementing the monitoring system to predict the planned state from maintenance operations to be performed to scheduled maintenance operations;

[0120] [ Figure 13 [Illustratively depicting something similar to] Figure 12 The long-term view specifically illustrates the second step of cyclically implementing the monitoring system to predict the planned status from the planned maintenance operation to subsequent planned maintenance operations;

[0121] [ Figure 14 A simplified view of an example of long-term forecasting is schematically depicted, with a particular illustration of a curve representing the change in equipment status over time during two consecutive planning cycles;

[0122] [ Figure 15 A simplified view of an example of long-term forecasting is schematically depicted, with particular emphasis on curves representing changes in equipment status over time over several consecutive planning cycles;

[0123] [ Figure 16 A simplified view of the architecture of the monitoring system is schematically depicted, with particular emphasis on the virtual model of the moment when the failure date of the equipment is determined due to inadequate maintenance decisions. Detailed Implementation

[0124] The present invention relates to a system 1 for monitoring the operation and maintenance of at least one piece of equipment 2 within an industrial facility 3.

[0125] This regulatory system 1 (hereinafter referred to as "System 1") is expected to be digital. In other words, it is at least one software program designed to be executed by at least one computing terminal.

[0126] Typically, such a computing terminal can be of any type, particularly a computer server or computer. Furthermore, the computing terminal enables the recording, reading, modification, and generation of data in digital form via suitable storage devices. The computing terminal is also envisioned to be accessible locally or remotely via a suitable communication network.

[0127] Furthermore, this system 1 includes a series of steps, which are described in a non-exhaustive manner below. According to the invention, system 1 can therefore be connected to a method.

[0128] Some steps are performed in a real manner, particularly by a person interacting with the computing terminal, especially via a virtual interface, or particularly when an operator interacts with the device 2.

[0129] Other steps are performed virtually, particularly when data processing is performed by a computing terminal. System 1 therefore relates to digital elements, similar to virtual technological means advantageously formed and implemented within the scope of this invention.

[0130] The system 1 provides oversight of a single device 2 within a facility 3, several devices 2 within the same facility 3, or even several devices 2 within several facilities 3.

[0131] As previously stated, the equipment 2 is characterized in that it is essential for the operation of the facility 3. Therefore, the unavailability of equipment 2 may interrupt the production of facility 3, thereby inducing or exacerbating accidents or unforeseen circumstances in facility 3.

[0132] The device 2 can be of any type, such as a thermal engine, electric motor, alternator, pump, or solenoid valve.

[0133] The device 2 thus forms part of the device 3 and is necessary for its proper operation.

[0134] Such facility 3 can be mobile, such as machinery or vehicles (e.g., submarines, ships, aircraft or spacecraft), or fixed, such as structures or infrastructure (e.g., power plants, oil platforms or refineries).

[0135] Therefore, as an initial condition, system 1 includes at least one piece of equipment 2, produced by manufacturing process 4 and representing a series, installed within industrial facility 3. In other words, equipment 2 has been integrated into facility 3 in the past.

[0136] Furthermore, the manufacturing of device 2 4 includes assembling several components until device 2 is obtained, and then installing device 2. Prior to this manufacturing process 4 is the design 40 of device 2.

[0137] It should be noted that device 2 represents a series, which includes all devices 2 manufactured identically according to a single design plan. Part of device 2 in this series is no longer operational, another part is currently operational, and yet another part that has been manufactured may still be awaiting installation and operation.

[0138] After device 2 is installed in facility 3, the device will operate for at least a period of time 5 until maintenance step 6 is required. Therefore, period 5 corresponds to the operating cycle of the device between installation and the maintenance to be performed 6, or to the alternation of operating cycles and maintenance steps since installation.

[0139] It should be noted that maintenance step 6 is a specific operation that requires intervention from at least one operator at the geographical location of facility 3 and to be performed directly on the equipment 2.

[0140] In this context, and as previously stated, monitoring system 1 envisions managing the future lifecycle of device 2, namely:

[0141] - Define the nature of future maintenance tasks for device 2 in order to optimize them and adapt them to the expected use of device 2, doing so for a long period of time and for the entire service life of device 2;

[0142] - For each planned operating cycle of device 2, identify the limit of zero-failure use during operation until the next maintenance, do so for a long period of time and for the entire service life of device 2;

[0143] -Refreshes the extreme usage of device 2 in real time during operation, meeting the zero-failure standard;

[0144] - By defining the nature of the last maintenance task of equipment 2, which is optimized and adapted to the expected use of equipment 2, the service life of equipment 2 is extended.

[0145] In terms of the scale of this series of equipment, the monitoring system 1 also envisions optimizing the manufacturing of standard equipment representing this series in order to adapt it to a given usage situation, that is, determining the manufacturing parameters 4 for the life cycle of the optimized equipment 2.

[0146] Therefore, at least one planned maintenance 7 is required after the maintenance 6 to be performed.

[0147] The future cycle occurring between the maintenance 6 to be performed and the planned maintenance 7 corresponds to at least one planned operation cycle 70, wherein the device 2 operates in at least one scenario 8, which is characterized by a planned usage condition 110 (referred to as "planned usage condition 110"). The future cycle may also correspond to the alternation of planned cycles for operation and maintenance steps.

[0148] The future cycle can be extended to the end of the equipment's lifespan.

[0149] Figure 1 The design 40 and manufacturing process 4 of the equipment 2 are shown in particular, as well as the cycle 5 (between the installation of the equipment 2 in the facility 3 and the maintenance 6 to be performed, and including any prior maintenance 10), and the planned cycle 70 up to the planned maintenance 7.

[0150] In this context, the steps of manufacturing process 4, the installation of equipment 2, and then the steps of operating and maintaining equipment 2 during cycle 5, all generate several types of data that the monitoring system 1 needs to consider.

[0151] The manufacture, installation, operation, and maintenance of the device 2 generate at least one manufacturing and maintenance log 9 (or "log 9"), which includes:

[0152] -For manufacturing the device 2 up to the installation task 90;

[0153] - Optionally, at least one previous maintenance task 90 of the device 2.

[0154] In other words, the manufacturing and installation steps include at least manufacturing and installation tasks 90. Furthermore, period 5 optionally includes at least one prior maintenance 10 of the device 2, which includes maintenance task 90. ​​All manufacturing tasks 90 up to the installation of the at least one piece of device 2, and any task 90 of the at least one prior maintenance 10 of the device 2, generate a manufacturing and maintenance log 9.

[0155] It should be noted that the manufacturing task 90 and the task 90 of each maintenance operation in the maintenance operation 10 are defined in a non-limiting manner by gestures performed by the operator, by components installed during manufacturing or removed during the maintenance step to be replaced, and by adjustments made to the device 2.

[0156] like Figure 2 As shown, system 1 specifically includes:

[0157] - Manufacturing and maintenance log 9, which is the log of manufacturing and maintenance decisions made regarding manufacturing process 4 and installation (level 0) and previous maintenance 10 (from level 1 to level j-1);

[0158] - Maintenance decision regarding maintenance level 6 (j) to be performed;

[0159] - Maintenance decision regarding planned maintenance 7 (level j+1).

[0160] Each manufacturing or maintenance decision of level k is characterized by its ratio to various possible tasks 90 (optional execution of task 90, the characterization of task 90 during execution, and timer 91 characterizing the duration after task 90 is executed on device 2).

[0161] In addition, the steps will generate a usage log 11 (or "log 11") for the device 2 during the cycle 5 between the installation and the maintenance 6 to be performed. The usage log 11 includes conditions 110 (also referred to as "condition 110") for the use of the device 2 during the cycle 5.

[0162] In fact, the cycle 5 between the installation of device 2 and the maintenance 6 to be performed includes at least one operating cycle of device 2 in the usage environment.

[0163] This use is associated with the manner in which the device has been used, namely, the set of usage conditions 110 of the device 2 during the period.

[0164] For example, in the case where device 2 corresponds to facility 3 such as a truck engine, these usage conditions 110 could be the transport load, mileage traveled, speed, ambient air temperature, and the average gradient of the route used.

[0165] This usage thus generates a usage log 11 for the device during the cycle 5 between installation and the maintenance 6 to be performed. For example... Figure 3 As shown, the device 2, such as a pump (for which one of the conditions 110 is temperature), is illustrated. The system 1 specifically includes:

[0166] - Usage log 11 of device 2 during cycle 5 (from installation at level 0 to maintenance to be performed at level j 6);

[0167] -Scenario 8, where device 2 is expected to be used within the planned period of 70.

[0168] Usage log 11 can be decomposed into a set of usage sub-logs 111 for device 2. The usage sub-log 111 for level k represents the proportion of usage log 111 for device 2 during the runtime cycle encompassed between two consecutive previous maintenance operations 10 at levels k and k+1. Specifically, Figure 3 Highlight:

[0169] - Level 0 usage sub-log 111 (between Level 0 installation and previous Level 1 maintenance 10);

[0170] - Use sub-log 111 for level j-1 (between the last previous maintenance operation 10 for level j and the maintenance to be performed 6 for level j).

[0171] Log 11 includes all usage sub-logs 111 from level 0 to level j-1.

[0172] The sub-log 111 of level k represents the evolution of each of the usage conditions 110 between level k and level k+1, such as the evolution of temperature over time (e.g., Figure 3 (See in the middle).

[0173] In the usage sub-log 111 of device 2 at level k, the partial usage sub-log 1110 of device 2 at level (k,t) can be distinguished, that is, for time (t) included between maintenance at level k and maintenance at level k+1, the evolution of each of the previous maintenance 10 of level k and the usage conditions 110 between time (t). Figure 3As shown, within the context of planned use of device 2 between maintenance 6 (level j) to be performed and planned maintenance 7 (level j+1), scenario 8 represents the planned evolution of each of these planned use conditions 110. In scenario 8, partial scenarios 80 can be distinguished. For time (t) included between maintenance 6 to be performed at level j and planned maintenance 7 at level j+1, the partial scenario 80 at level (j,t) represents the planned evolution of each of the planned conditions 110 for device 2 between maintenance 6 to be performed at level j and time (t).

[0174] In addition, the steps will generate a status log 12 (or "log 12") of the device 2, which includes the material specifications 120 (or "specification 120") of the device 2.

[0175] In practice, the use of device 2 will cause aging, thus affecting the state 13 of device 2. The state 13 of device 2 at a given moment, characterized by material index 120, reflects the physical integrity of device 2, and the operational capability of the device depends on this physical integrity. All states 13 (or all material indexes 120) of device 2 that actually occur during cycle 5 constitute the state log 12 of device 2.

[0176] like Figure 4 As shown, this illustrates the case of device 2, such as a pump (with one of its material parameters 120 being flow rate (Q)). System 1 specifically includes a status log 12 of device 2 during cycle 5. Status log 12 can be decomposed into a set of status sub-logs 121. The status sub-log 121 of device 2 at level k represents the proportion of status log 12 during the operating cycle encompassed between two consecutive previous maintenance operations 10 at levels k and k+1. Specifically, Figure 4 Highlight:

[0177] - Level 0 status sub-log 121 (between Level 0 installation and previous Level 1 maintenance 10);

[0178] - Status sub-log 121 of level j-1 (between the last previous maintenance 10 of level j-1 and the maintenance to be performed of level j 6).

[0179] Log 12 includes all state sub-logs 121 from level 0 to level j-1. For example... Figure 4 As shown, the state sub-log 121 of device 2 at level k represents the evolution of the consecutive states 13 of device 2 between two previous maintenance operations 10 at levels k and k+1. The state sub-log 121 of device 2 at level k represents the evolution of each of the material indices 120 (e.g., the evolution of flow rate over time) between the two previous maintenance operations 10 at levels k and k+1. Figure 4As shown, partial state sub-logs 1210 can be distinguished in the state sub-log 121 of level k of device 2. The partial state sub-log 1210 of level (k,t) represents the evolution of each of the material indices 120 between the previous maintenance 10 of level k and time (t) (the time (t) between the maintenance of level k and the maintenance of level k+1).

[0180] Therefore, logs 9, 11, and 12 extend over time within cycle 5. They respectively include manufacturing and maintenance tasks 90, usage conditions 110, and material specifications 120.

[0181] The elements in logs 9, 11, and 12 indicate the designation of a computational domain in which measurements from device 2 have been continuously recorded over time.

[0182] Advantageously, in the first step, at least one correlation 14 is determined between at least one of the manufacturing and maintenance tasks 90 and / or at least one of the usage conditions 110 and at least one of the material indicators 120 of the state 13. The correlation 14 establishes at least one link between the causes and consequences of aging of the equipment 2.

[0183] In other words, the present invention chooses to approximate the behavior of device 2 from the perspective of the causes and consequences of aging.

[0184] Therefore, the present invention selects the manufacturing and maintenance log 9 of device 2 and the usage log 11 of device 2 to characterize the causes of aging of device 2. The present invention also selects the state 13 of device 2 caused by said logs 9 and 11 to characterize the consequences of aging of device 2.

[0185] Therefore, the present invention chooses to associate the state 13 of the device 2 with the manufacturing and maintenance log 9 and the usage log 11 of the device 2.

[0186] Therefore, the present invention anticipates associating the manufacturing and maintenance log 9 (hereinafter referred to as the “first item” of the correlation 14) and the usage log 11 (hereinafter referred to as the “second item” of the correlation 14) of a given device 2 in the series with the state 13 of the device 2 derived from these logs 9, 11 (hereinafter referred to as the “third item” of the correlation 14)—the three items of the correlation 14 constitute a triple 140.

[0187] For example, considering device 2, such as a pump, which is sensitive to the temperature of the fluid being delivered, the present invention correlates the pump's usage log (characterized by fluid temperature log and / or suction pressure log and / or pump speed log), as well as the pump's main manufacturing options (such as the type of impeller installed) and maintenance log (such as impeller replacement logs during various previous maintenance operations 10) with the pump's condition (characterized by its flow rate and / or delivery pressure).

[0188] Furthermore, the purpose of correlation 14 is only to consider relevant parameters that may affect the aging and operation of device 2; other parameters are not suitable for selection.

[0189] Therefore, a technical analysis of equipment 2 is conducted to identify key manufacturing and maintenance tasks 90 and usage conditions 110 that affect the condition 13 of equipment 2.

[0190] Therefore, through technical analysis of device 2, the manufacturing and maintenance task 90 is characterized by those tasks identified as critical, as defined below.

[0191] More precisely, the present invention selects at least one of the following methods to characterize the manufacturing and maintenance log 90 of device 2:

[0192] - For manufacturing process 4 of device 2, the present invention characterizes gestures set during manufacturing, installed components, or even adjustments adopted during manufacturing. If there are several possible protocol options for gestures for manufacturing process 4 of this series of devices 2, the gestures of the manufacturing process are characterized by protocol options for each manufacturing gesture. If there are several possible options for the components for manufacturing process 4 of this series of devices 2 (e.g., if the manufacturing of the pump series anticipates two possible types of pump bearings), the components installed during manufacturing are characterized by installation component options. If there are several possible values ​​for the adjustments for manufacturing process 4 of this series of devices 2 (e.g., the tightening torque of the pump gland), the adjustments adopted during manufacturing are characterized by the value of each adjustment.

[0193] - For maintenance procedures of device 2, this invention characterizes the gestures set, components installed, or even adjustments made during each maintenance procedure. Maintenance gestures include replacing a portion of another maintenance task 90 (e.g., retightening the electrical junction box connections). If several possible options exist for the components being maintained for this series of devices 2 (e.g., if the series of pumps anticipates two possible types of pump bearings), then the replacement of the component during the relevant maintenance is characterized by component selection. If several possible values ​​exist for the adjustments being made for the maintenance of this series of devices 2 (e.g., the tightening torque of the pump gland), then the adjustments made during the maintenance are characterized by the value of each adjustment.

[0194] Furthermore, according to one implementation, in correlation 14, the manufacturing and maintenance log 9 is simplified to a log of at least one critical task 90 in the form of a list of continuous values.

[0195] In other words, through technical analysis of device 2, the present invention selects to characterize the manufacturing and maintenance log 9 of device 2 by considering only critical tasks 90, namely, critical tasks 90 are identified as gestures, components and adjustments that determine the behavior of device 2 during operation (i.e., have an impact on aging and therefore on the state 13 of device 2).

[0196] Furthermore, the manufacturing and maintenance log 9 of equipment 2 (up to and including the last previous maintenance operation 10) is characterized by listing the values ​​that have been continuously used during the manufacturing process 4 and during each consecutive previous maintenance operation 10 throughout the life of equipment 2, up to the last previous maintenance operation 10 (the operation before the maintenance 6 to be performed) for each critical task 90. ​​The manufacturing and maintenance log 9 is thus characterized by a list of values. For example, considering the case of equipment 2 such as a pump, which includes bearings, and the placement or replacement of the pump bearings and their nature, this corresponds to the critical manufacturing and maintenance task 90 of the pump. In this example, the last previous maintenance 10 (the one before the maintenance 6 to be performed) is considered to be the seventh (level k = 7). The present invention characterizes the log of this critical manufacturing and maintenance task 90 by listing [A,0,0,B,0,0,B,0] to indicate the fact that a type A bearing was placed in manufacturing operation 4, the bearing was replaced with a new type B bearing in the third maintenance operation, the bearing was replaced with a new type B bearing in the sixth maintenance operation, and that no maintenance action was performed on the bearing in other previous maintenance operations 10. Therefore, the present invention chooses to characterize the manufacturing and maintenance log 9 of device 2 by means of a matrix of lists, each list being associated with a critical task 90 and listing consecutive values ​​characterizing that critical task 90 according to the manufacturing process 4 and then according to the various maintenance operations of device 2 up to and including the last previous maintenance operation 10. For example, considering the above-described case of device 2 such as a pump, where the manufacturing and maintenance log 9 can be simplified to a log of two critical tasks 90 "pump bearing placement or replacement" and "pump impeller placement or replacement", and after the seventh maintenance operation, there are lists [C,0,A,0,0,B,0,C] and [A,0,0,B,0,0,B,0] as corresponding log lists. The manufacturing and maintenance log 9 of the pump in question up to and including the seventh maintenance operation then corresponds to the matrix [[C,0,A,0,0,B,0,C]; [A,0,0,B,0,0,B,0]].

[0197] Furthermore, the use condition 110 is characterized by those conditions that the device 2 is sensitive to and exposed to during operation or when it is stopped.

[0198] In fact, through technical analysis of device 2, the present invention reduces condition 110 to environmental conditions and / or operating conditions (CA / CF) to which device 2 is sensitive and exposed during operation or when it is stopped.

[0199] Environmental conditions (CA) are understood to be the environmental conditions outside of device 2, and that device 2 is sensitive to and exposed to these environmental conditions (such as ambient air temperature, humidity, and irradiation rate) during operation or when it is stopped.

[0200] Run conditions (CF) should be understood as follows:

[0201] - Conditions inside device 2, and device 2 is sensitive to and exposed to these conditions during operation or when it is stopped (such as the temperature of the fluid being transported, the level of pump vibration);

[0202] and / or

[0203] - These parameters represent the power used by device 2 during operation, and these parameters affect the operating point of the device (such as the load of the transport vehicle, the speed of the truck engine).

[0204] and / or

[0205] - Other parameters necessary to characterize the sum of work performed by device 2 under the aforementioned environmental and operating conditions (CA / CF) (such as the total mileage of the truck, or typically the usage time of device 2).

[0206] For example, consider device 2, such as a pump, whose impeller is made of thermoplastic material and is therefore sensitive to and exposed to the temperature of the fluid being transported. The temperature of the fluid being transported is considered even more important as an operating condition compared to a pump whose impeller is made of metal.

[0207] Within the meaning of this invention, scenario 8 and planned use condition 110 are characterized by the same environmental and operating conditions (CA / CF) to which device 2 is sensitive and exposed during operation or when it is stopped.

[0208] Furthermore, the material state 13 of the device 2 is characterized by a material index 120 that is identified as representing the state 13 of the device 2.

[0209] Therefore, and through technical analysis of device 2, the present invention selects to characterize the state of device 2 at that moment by a set of performance, vibration behavior and other material indicators 120 (e.g., and in a non-limiting manner, material indicators typically measured by non-destructive testing techniques) that are deemed necessary and sufficient to represent the state 13 of device 2.

[0210] In order to characterize the logs of correlation 14 and, in particular, condition 110 as specifically as possible for the device 2 under discussion, the measured physical quantities or functions of the measured physical quantities were determined to characterize the conditions 110 to which the device 2 is sensitive and exposed during operation or when it is stopped, or to best characterize the logs of these usage conditions 110 (i.e., usage log 11).

[0211] These functions of physical quantities include several mathematical or algorithmic functions. In other words, each condition 110 has a physical quantity associated with it that characterizes the condition. Therefore, when relevant, the present invention characterizes the log of the condition through the log of that physical quantity.

[0212] For example, if device 2 is sensitive and exposed to ambient air temperature or, in the case of a pump, to the temperature of the fluid being pumped, then the log of the corresponding temperature is therefore considered. Furthermore, the present invention also selects to characterize the log of condition 110 as a function of the physical quantity of condition 110, if such characterization is more relevant than the log of the physical quantity.

[0213] For example, if device 2 is sensitive to ambient air temperature or the temperature of the fluid being transported (e.g., in the case of a pump) and is exposed to that temperature, the temperature log can be integrated into the time integral of the temperature over the duration between the installation of device 2 in facility 3 and that moment.

[0214] Furthermore, when the present invention chooses to characterize the log of the condition by means of a log of functions representing the physical quantity of condition 110, these functions may include, in a non-limiting manner, the calculation of the existence time of the physical quantity measured within at least one value range.

[0215] In practice, this invention chooses to consider the cumulative effects of abnormal transients on the aging of device 2 and thus on its behavior. Therefore, when relevant, this invention characterizes the log of condition 110 by counting the number of times the condition exists within the normal operating range, within the range close to failure, or even in the intermediate range between the first two ranges.

[0216] For example, considering the case of device 2, such as a pump, which is sensitive to and exposed to the temperature of the fluid being transported, the temperature log can then be characterized by the time integral of the temperature over the duration from when device 2 is installed in facility 3 until that time, thereby distinguishing the component of the integral within the normal operating range from the component within the range nearing destruction, and the component within the intermediate range between the two aforementioned domains. When the invention chooses to characterize the log of condition 110 by means of a log of functions representing the physical quantities of condition 110, these functions may also include, in a non-limiting manner, the calculation of at least one fluctuation of the measured physical quantity and / or the counting of said at least one fluctuation.

[0217] Specifically, this is the case when device 2 is sensitive to and exposed to changes in condition 110. In a non-restrictive manner, such a function can then correspond to a gradient function or a count of loops corresponding to the conditions under discussion.

[0218] In the first example, considering device 2, such as a pump, which is sensitive to and exposed to the temperature of the fluid being transported, particularly to sudden changes in said temperature, a log of condition 110 up to a given moment can be calculated during temperature transients (i.e., during the transients that cause sudden temperature changes) (e.g., an average change of 20°C / min "degrees per minute"). This average gradient can then be correlated with a count of similar transients (e.g., 2000 sudden temperature transients from installation up to said moment correlated with the average gradient of 20°C / min).

[0219] Instead of the average value of the logs for this temperature gradient, the median and standard deviation can also be obtained using any other statistical function.

[0220] In another example, consider the case of device 2, such as a metal boiler tank, which is sensitive to and exposed to the temperature of the fluid it contains, especially to temperature cycling (i.e., experiencing large-amplitude temperature changes during heating or cooling, e.g., between 200°C and 80°C) and is sensitive to these cycles. These cycles can then be counted (e.g., 20 temperature cycles from installation up to a certain point in time), and this count of such cycles can be correlated with the average cycle amplitude in the tank's log (e.g., an average cycle amplitude of 150°C over 20 cycles).

[0221] Instead of the average value of the log, any other statistical function, such as the median and standard deviation, may be used. When the invention selects a log that represents a function of the physical quantity of condition 10 to characterize the log of condition 110, these functions may also include, in a non-limiting manner, the count of the at least one fluctuation.

[0222] This is especially true when the device is sensitive to shutdown or startup transients. Counting such transients is then useful for characterizing the usage log 11. Similarly, counting the duration of the shutdown while the device 2 continues to age is also useful for characterizing the usage log. Likewise, in order to characterize the correlation 14, particularly the state 13 of the device 2 at a given moment, and as specifically as possible for the device 2 in question, also determine the measured physical quantities or functions of these measured physical quantities that characterize the material state 13 of the device 2 at that moment.

[0223] For example, considering device 2, such as a centrifugal pump, the pump's condition can then be characterized by instantaneous flow rate and instantaneous delivery pressure. In this step, and through technical analysis of device 2, the present invention has thus identified key manufacturing and maintenance tasks 90 specific to device 2 for optimally characterizing the manufacturing and maintenance log 9 of device 2. Similarly, usage conditions 110 specific to device 2, and physical quantities or functions of physical quantities optimally characterizing said conditions 110, as well as the usage log 11 of device 2, have been identified.

[0224] Similarly, the material index 120 that best characterizes the state 13 of device 2 at any given time, and the physical quantity or physical quantity function that best characterizes the material index 120 at that time, have been determined.

[0225] Therefore, this technical analysis of device 2 enables the determination of correlation 14 in a form specific to device 2. Thus, the determined correlation 14 is indeed specific to the physical phenomenon carried out by each device 2 in the series. Therefore, according to one embodiment, in correlation 14, the function of the measured physical quantity using condition 110 includes the calculation of the duration of the measured physical quantity in at least one value range; and / or the calculation of at least one fluctuation representing the measured physical quantity; and / or the counting of said at least one fluctuation.

[0226] Previously, in Relevance 14, the present invention chose to characterize the manufacturing and maintenance log 9 of the equipment 2 (up to and including the last previous maintenance operation 10) by listing values ​​continuously used throughout the life of the equipment 2 during manufacturing and various consecutive previous maintenance operations 10 (up to and including the last previous maintenance operation 10) for at least one (preferably each) critical manufacturing and maintenance task 90. ​​The log for each critical task 90 is thus characterized by a list of values. For example, considering the above-described case of the equipment 2 such as a pump including bearings, and for this case, the placement or replacement of the pump bearing and its nature correspond to the critical task 90 for manufacturing and maintaining the pump. In this example, the last previous maintenance 10 (the one before the maintenance 6 to be performed) is considered to be the seventh (level k = 7). The present invention characterizes the log of this critical manufacturing and maintenance task 90 by a list [A,0,0,B,0,0,B,0] to indicate the placement of a type A bearing in manufacturing operation 4, the replacement of the bearing with a new type B bearing in the third maintenance operation, the replacement of the bearing with a new type B bearing in the sixth maintenance operation, and the fact that no maintenance action was performed on the bearing in other maintenance operations.

[0227] Therefore, the manufacturing and maintenance log 9 of device 2 is a matrix that groups together the corresponding value lists for each critical task 90. ​​Furthermore, in order to reduce the matrix characterizing the manufacturing and maintenance log 9 (up to the last previous maintenance operation 10) to the necessary and sufficient information and to have an information level independent of the last previous maintenance operation 10, the present invention employs the following principle, referred to herein as the "persistent parameter principle." According to one embodiment, in correlation 14, the manufacturing and maintenance log 9 has been simplified to a critical task log 90 in the form of at least one continuous list of values, in each list, only persistent values ​​are selected as the values ​​adopted in the last maintenance operation during the execution of the task in question 90, and only persistent values ​​are retained in the log of the critical task 90. ​​In other words, in the manufacturing and maintenance log 9 and for each critical manufacturing and maintenance task 90:

[0228] - In the final manufacturing or maintenance step of performing the task 90 (either in the manufacturing step or in the previous maintenance step 10), the persistent value of the task 90 is considered to be the value adopted for the task 90;

[0229] - Only the persistent values ​​of the tasks mentioned are retained in the manufacturing and maintenance log 9.

[0230] In its final form, the manufacturing and maintenance log 9 is thus simplified to a list of values, that is, consisting of persistent values ​​for critical tasks 90 (or one persistent value for each critical task 90). In this case, the invention simplifies the manufacturing and maintenance log 9 of device 2 by characterizing the log solely by the values ​​of the critical tasks 90, which determine the behavior of device 2 during operation following the last previous maintenance operation 10 in question. In practice, in the manufacturing and maintenance log 9:

[0231] - The value representing each manufacturing and maintenance task 90 performed during the last previous maintenance operation 10 rewrites the series of values ​​representing the task 90 in the log (from manufacturing process 4 until the maintenance prior to the last previous maintenance operation 10);

[0232] -Starting from manufacturing process 4 and the previous maintenance operation 10 prior to the last previous maintenance operation 10, for each task 90 that has not been executed in the last previous maintenance operation 10, only the value representing the last execution of the task 90 is retained.

[0233] For example, considering the case of equipment 2 such as a pump, the tasks of "placing or replacing the pump bearing" and "placing or replacing the pump impeller" are the only two critical manufacturing and maintenance tasks 90.

[0234] If the log of the "pump bearing placement or replacement" task after the seventh previous maintenance operation 10 is characterized by the list [A,0,0,B,0,0,B,0], then the persistent value of task 90 is "B", which is the value of task 90 used during the last maintenance operation in which task 90 in question was performed (in this example, during the sixth maintenance operation).

[0235] If the log of the "pump impeller placement or replacement" task after the seventh previous maintenance operation 10 is characterized by the list [C,0,A,0,0,B,0,C], then the persistent value of task 90 is "C", which is the value of task 90 used during the last maintenance operation in which task 90 in question was performed (in this example, during the seventh maintenance).

[0236] Then, the pump manufacturing and maintenance log 9 after the seventh maintenance operation is written as [B; C], where the first item indicates the persistence value of the task "Placement or replacement of pump bearings" and the second item indicates the persistence value of the task "Placement or replacement of pump impeller".

[0237] In other words, the present invention thus simplifies the manufacturing and maintenance log 9 of device 2 (up to and including the last previous maintenance operation 10) to the configuration of device 2 after the last previous maintenance operation 10. In this case, the present invention represents the manufacturing and maintenance log 9 (up to the last previous maintenance operation 10) by a vector whose magnitude is independent of the level of said last previous maintenance operation 10. It should be noted that the present invention may take into account the timing of recorded values, for example, in the form of data timestamps.

[0238] In the characterization of manufacturing and maintenance log 9, this timing may lead to the addition of timer 91 at each critical persistence value of task 90, which timer 91 characterizes the duration since task 90 was performed on device 2. In this step, the correlation 14 previously determined in a form specific to device 2 (and therefore specific to any standard device in this series) is then determined in a format suitable for computer processing, particularly through machine learning algorithms, by means of a technical analysis of device 2.

[0239] Once a technical analysis of device 2 was performed, a correlation 14 specific to this series was identified. The following were then determined:

[0240] - In the first item of correlation 14: the key manufacturing and maintenance tasks 90 of the manufacturing and maintenance log 9 of equipment 2 are best characterized;

[0241] - In the second item of correlation 14: the physical quantity or function of the physical quantity of the usage log 11 of condition 110 and device 2 is best characterized;

[0242] - In the third item of correlation 14: the physical quantity or function of the log 12 of the material index 120 and the state 13 of the device 2 best characterizes the material index 120 and the state 13 of the device 2.

[0243] The key task 90, namely the physical quantity or function of physical quantity identified in correlation 14, indicates the raw data (measured and recorded for each device 2 in the series) to be initially extracted from logs 9, 11, and 12 of device 2. Therefore, the present invention anticipates accessing this accessible information from all devices 2 in the series under discussion. For all devices 2 in the series, data associated with these tasks 90 and data associated with physical quantities or functions of physical quantities relating to usage conditions 110 and material parameters 120, as identified in correlation 14, are recovered and extracted to form data triplets 140 and obtain the raw dataset 150. For each device 2 in the series and for each moment in their corresponding log 11, the three items of the triplet 140 associated with that moment are:

[0244] -The manufacturing and maintenance collection 95 of the device 2, that is, a portion of the manufacturing and maintenance log 9 of the last maintenance 10 up to the time mentioned above;

[0245] - The usage set 115 of the device 2, that is, the portion of the usage log 11 up to the time mentioned;

[0246] - The state 13 of device 2 at that time. As for the functions of the physical quantities identified in correlation 14, they indicate transformations 141 to be applied subsequently to each of the triples 140 of the original dataset 150, in order to form the final dataset 15 to be considered, so as to model correlation 14 via virtual model 16. For example, considering the case of device 2 such as a pump, for this case, the terms of correlation 14 specific to the pump are therefore characterized as:

[0247] - The manufacturing and maintenance log 9 is characterized by critical tasks related to the pump's bearings and impellers 90;

[0248] -The features of the usage log 11 up to that moment are the first time integral (INT1(t)) of temperature over the duration between installation and that moment, the second time integral (INT2(t)) of suction pressure over the duration between installation and that moment, and the average value of temperature gradient (VMG(t)) over the duration between installation and that moment, which is related to the number of sudden temperature transients (NTB(t)) within the same duration.

[0249] - The state at that moment is characterized by the instantaneous values ​​of flow rate and delivery pressure 13.

[0250] The data to be extracted from logs 9, 11, and 12 of each pump in the series under discussion, and for each moment in their corresponding usage log 11, is as follows:

[0251] - Regarding the first item of correlation 14: the bearing type and impeller type that were in place at that moment, to be extracted from manufacturing and maintenance log 9;

[0252] - Regarding the second item of correlation 14: physical quantities to be extracted from usage log 11: values ​​of temperature and inhalation pressure at the time of consideration;

[0253] - Regarding the third item of correlation 14: the physical quantities to be extracted from status log 12: the values ​​of delivery pressure and flow rate at the considered time. Once extracted, this data enables the construction of three items in a triplet 140 associated with each pump and each time(t) of their corresponding usage log 11, namely:

[0254] - The manufacturing and maintenance set 95 of the pump, that is, the portion of the manufacturing and maintenance log 9 during the duration between manufacturing process 4 and the last maintenance 10 prior to said time (t). In this example, the manufacturing and maintenance set 95 is a log of tasks 90 (related to the pump's bearings and impellers) performed continuously during said duration, or a matrix of two lists, one associated with bearings and the other with the pump's impellers, each listing consecutive values ​​characterizing the critical task 90 performed during said period (type of in-place bearing; type of in-place impeller).

[0255] - Pump usage set 115, that is, in this example, temperature and suction pressure values ​​recorded from installation until the time (t);

[0256] - The pump's state 13 at time (t), that is, in this example, the flow rate and delivery pressure at time (t).

[0257] Therefore, as many triplets 140 as the time period under consideration and the pumps in the series are obtained: all these triplets 140 form the original dataset 150. Next, for each pump in the series and for each triplet 140 associated with a time period, a set 95 of pump manufacturing and maintenance up to that time period is transformed into a vector, which is associated with that time period and indicates the persistent value of each critical task 90. ​​Then, for each time period under consideration, the corresponding vector is [the type of bearing in place at that time; the type of impeller in place at that time], each item of which is associated with its own timer 91. Similarly, for each pump in the series and for each triplet 140 associated with a time period, a set 115 of pump usage up to that time period is transformed. In this example, since the physical quantities identified in correlation 14 are functions of the integrals of temperature and inhalation pressure over time (INT1(t), INT2(t)) and the average of the temperature gradient (VMG(t)) associated with the number of sudden temperature transients (NTB(t)), the transformation 141 to be applied to the raw data, temperature, and inhalation pressure extracted from usage log 11 is:

[0258] - Calculate the time integral of temperature and suction pressure over the duration between installation and that moment;

[0259] - Calculate the average temperature gradient over the same duration and the number of sudden temperature transients over the same duration.

[0260] For each pump in the series and for each moment, the usage set 115 of the pump between installation and that moment is thus converted into a numerical vector [INT1(t), INT2(t), VMG(t), NTB(t)] associated with that moment. Next, for each pump in the series and for each moment, the pump's status 13 at that moment is indicated in the form of a numerical vector [flow rate, delivery pressure] associated with that moment. Finally, each of the triples 140 (associated with the various pumps (k) belonging to the series and their respective usage logs 11 at various moments (t)) is thus converted into a vector [Mj(t)(k); Vu(t)(k); Status(t)(k)], where:

[0261] -Mj(t)(k) represents the vector of the manufacturing and maintenance set 95 of pump(k) up to this moment, in the form of [bearing type; wheel type], with each item in the vector associated with its timer 91;

[0262] -Vu(t)(k) represents the vector of the set 115 used by pump(k) between installation and this moment, in the form of [INT1(t),INT2(t),VMG(t),NTB(t)];

[0263] -Status(t)(k) represents the vector of the state 13 of pump (k) at this moment, in the form of [flow rate, delivery pressure].

[0264] Therefore, the original dataset 150 is transformed into the form of the final dataset 15, namely: a vector set [Mj(t)(k); Vu(t)(k); Status(t)(k)] of any pump(k) belonging to the series and their corresponding usage log 11 at any time (t). In short, System 1 anticipates recovering and processing only the data identified as relevant in correlation 14. Optionally, as described above, System 1 may recover all data relevant to logs 9, 11, 12 of device 2 so that only those relevant data are extracted and then processed. Thus, System 1 transforms the original dataset 150 into a dataset 15 associated with correlation 14 specific to the series in question. System 1 uses this dataset 15 as input to machine learning project 142: then at least one virtual model 16 is trained based on dataset 15. Similar to correlation 14, the obtained virtual model 16 is then specific to device 2 of the series in question. It also possesses the accuracy of an empirical model (data-driven model).

[0265] Figure 5 and Figure 6 The process of collecting data and training model 16 is illustrated. Figure 5 In this context, device 2, which is waiting to perform its maintenance 6, is considered, along with its logs 9, 11, 12 and a given time (t) during cycle 5.

[0266] Figure 5 The diagram shows three items of a triple 140 linked by correlation 14: the manufacturing and maintenance set 95 of the device, the usage set 115, and the state 13 of the device at time (t), where:

[0267] - Manufacturing and maintenance set 95 includes manufacturing and maintenance decisions from level 0 to level k of the last maintenance prior to said time (t);

[0268] - Use set 115 to include the use of sub-log 111 up to time (t) during period 5;

[0269] - The state of device 2 at time (t) is 13. Figure 15 The extraction and processing of data associated with the device 2 are also illustrated, with particular examples showing:

[0270] - Extraction of data associated with manufacturing and maintenance set 95, usage set 115, and state 13 at said time (t);

[0271] - The composition of triple 140 (set 95; set 115; state 13 at time (t)) is as follows Figure 5 As shown;

[0272] - The formation of as many similar triples as time(t) recorded during period 5;

[0273] - Inject the triples into the original dataset 150, such as Figure 5 As shown;

[0274] - The composition of the original dataset 150 for each device in facility 3 that is of the same type as device 2.

[0275] Figure 6 It is depicted schematically:

[0276] - The original dataset 150 extracted from several facilities 3 that operate the same series of equipment as device 2 is recovered by transmission in order to merge them, and then the data is transformed 141 to form dataset 15;

[0277] - The dataset 15 is presented as input to the machine learning project 142, and a virtual model 16 is trained based on the dataset 15 to model the correlation 14. By modeling the behavior of device 2 (or the behavior of the series of devices) in this way, the invention correlates the observed behavior (i.e., state 13) of each device 2 with the manufacturing and maintenance log 9 and the usage log 11 of the device 2 at the series scale. Therefore, the invention uses a feedback mechanism that derives its superior properties from the functional correlation of the modeled correlation 14, especially because it is specific to the physical phenomena of each device 2 carried by the series.

[0278] According to one implementation, the training of Model 16 falls within the field of artificial intelligence and is capable of machine learning. In other words, the present invention anticipates the use of computer technologies involving artificial intelligence, particularly the field of machine learning, which is based on mathematical and statistical methods to endow computers with the ability to learn from data, i.e., to improve their performance in solving tasks without explicitly programming for each of them.

[0279] Specifically, the monitoring system 1 includes a virtual model 16 generated by this learning. Furthermore, the training of model 16 may involve any type of machine learning, that is, in a non-restrictive manner, supervised, semi-supervised, unsupervised, reinforced, or even transfer-based machine learning. Moreover, machine learning can implement any class of training methods, which can be combined, that is, in a non-restrictive manner: neural networks (including deep learning methods), k-nearest neighbor methods (“KNN”), genetic algorithms, genetic programming, or other methods, such as, in particular, Bayesian networks, support vector machines (SVM), Q-learning, decision trees, statistical methods, logistic regression, and linear discriminant analysis.

[0280] Prior to this, the training of model 16 involves supervised, regression-based, and deterministic machine learning problems (model 16 preferentially determines vectors of quantitative and continuous data from a set of labeled data 15, which are data vectors that have been selected to be integrated into the planned state 130 of device 2 in this invention).

[0281] According to other implementation schemes, Model 16 can be of any type.

[0282] The present invention envisions considering new manufacturing, maintenance, usage, and status data generated since the first training of model 16 within the context of machine learning project 142, and updating said virtual model 16. To this end, according to one embodiment, the previous operations are repeated periodically:

[0283] - New data is repeatedly recovered from at least one manufacturer, maintenance technician, and / or operator. Preferably, new data related to all devices 2 in the series is recovered from a group of operators implementing the series of devices;

[0284] Then, the extraction and transformation of the new data 141 is performed to obtain the complete dataset 15, and the training of the model 16 is subsequently updated based on the complete dataset 15.

[0285] It should be noted that this update is performed periodically, and the periodicity is determined by the designer, manufacturer, maintenance technician, and operator based on the criteria they deem most relevant or from a “data science” perspective.

[0286] For example, the periodicity criterion could correspond to a 10% rate of the duration during which the original dataset 150 is compiled in the context of the first training. If the data is compiled for more than ten years, the period used to update the virtual model 16 could correspond to twelve months.

[0287] Therefore, the present invention enables the use of system feedback in a continuous or quasi-continuous manner.

[0288] According to the present invention, once the correlation 14 is determined and training of model 16 is performed, the present invention applies model 16 to device 2.

[0289] Therefore, advantageously, during the maintenance 6 to be performed on the device 2, values ​​are submitted to the model 16. These values ​​correspond to at least one of the tasks 90 in the manufacturing and maintenance log 9 and at least one of the usage conditions 110 in the usage log 11. Furthermore, these values ​​correspond to at least one of the tasks 90 in the maintenance 6 to be performed and the planned usage conditions 110 of scenario 8.

[0290] In fact, model 16 has learned to associate the state 13 of device 2 at a given moment with the manufacturing and maintenance set 95 (i.e., the last previous maintenance 10 up to the time mentioned above) and the usage set 115 up to the time mentioned above.

[0291] Model 16 is thus able to associate the planned state 130 of device 2 at a given moment with the modified manufacturing and maintenance log 9 of the maintenance task 90 to be performed, as well as the usage log 11 up to said moment, which is followed by partial scenario 80.

[0292] In the application of Model 16, the following input data is therefore submitted to Model 16:

[0293] - The overall manufacturing and maintenance log, including the manufacturing and maintenance log 9 of equipment 2 (up to the last previous maintenance 10 before the maintenance 6 to be performed), and modified with the help of task 90 related to the maintenance decision considered in the maintenance 6 to be performed;

[0294] - The overall usage log includes usage log 11 for device 2 (up to the day before maintenance 6 to be performed), followed by the partial scenario 80 discussed (from the start of the planned operation cycle 70 until the moment considered during the planned operation cycle 70).

[0295] Therefore, the present invention uses the model 16 as a simulator customized for device 2.

[0296] In fact, model 16 enables the simulation of a planned state 130 of device 2 at a given moment, for a given maintenance decision regarding task 90 of maintenance 6 to be performed on device 2 and for the usage of device 2 up to that moment in the context of scenario 8. Furthermore, this simulation is tailored to device 2 because the manufacturing and maintenance logs 9 up to the day before the maintenance 6 to be performed and the usage logs 11 up to the day before the maintenance 6 to be performed, considered when performing the simulation, are specific to device 2, i.e., specific to the device 2 in question.

[0297] like Figure 1 As can be seen, applying the monitoring system 1 to device 2 awaiting maintenance 6 (level j) allows data related to device 2 to be considered to determine the planned state 130 of device 2 at a future time corresponding to the day before planned maintenance 7 (level j+1):

[0298] - Manufacturing and maintenance log 9 and usage log 11 of the equipment;

[0299] - and tasks 90 regarding maintenance decisions to be performed (maintenance 6) and use cases (use scenarios 8).

[0300] Figure 1 Specifically, model 16 receives the aforementioned data as input and generates the planned state 130 of device 2 at the end of the planning period 70 as output.

[0301] Furthermore, once these values ​​are submitted, the model 16 generates the planned state 130 of the device 2 after the maintenance 6 to be performed.

[0302] In other words, based on the data submitted as input, model 16 is able to predict the planned state 130 of device 2 for a given time, that is, the value of each of the material parameters 120 for that planned state 130 at that time. The planned state 130 is then compared with the minimum state 17 identified as required for the operation of device 2.

[0303] In other words, for each of the material parameters 120 in the program state 130, the value predicted by model 16 will be compared with the minimum required for the operation of device 2—for example, the minimum performance of device 2 (or, depending on the circumstances, with the maximum required for the operation of device 2—for example, the maximum wear of device 2).

[0304] If the simulated planned state 130 for the considered time period meets the operating criteria (that is, if the value predicted by model 16 for each material index 120 of planned state 130 is greater than (or less than) the minimum (or maximum) value required for the operation of equipment 2), then the predicted planned state 130 corresponds to the correct operating state of equipment 2.

[0305] For example, considering the case of device 2, such as a pump, its state 13 can be considered to be characterized at a certain moment (t) by two material parameters 120, namely flow rate (Q(t)) and delivery pressure (Pref(t)), where Q min and Pref min These are the minimum values ​​required for the operation of the pump.

[0306] The pump's lowest state 17 is therefore a set of values ​​[Q] min ,Prefmin The pump's operating standard at time (t) is therefore that the planned state 130 is greater than the minimum state 17, meaning that the flow rate Q(t) and the delivery pressure Pref(t) satisfy two conditions: Q(t) > Q. min and Pref(t)>Pref min .

[0307] To determine whether the planned pump state 130, as predicted for the time considered in model 16, corresponds to the correct operating state of the pump, state 130 is then compared with the lowest state 17. That is, the predicted values ​​of the flow rate Q(t) and delivery pressure Pref(t) material index 120 for that time are compared with their corresponding minimum values ​​Q. min and Pref min Compare them.

[0308] In this example, Figure 7 The projections for state 130 of the program are shown, along with a comparison to the lowest state 17.

[0309] Figure 7 The previous example of device 2, such as a pump, is repeated, where its state 13 at time (t) can be characterized by two material parameters 120: flow rate (Q(t)) and delivery pressure (Pref(t)), where Q... min and Pref min The material index 120 is used as the minimum value for the minimum state 17, which is the minimum value required for the operation of the pump.

[0310] Figure 7 The evolution of the planned state 130 of the pump (i.e., continuous values ​​of material parameters 120Q and Pref) calculated by model 16 during the planning cycle 70 is shown.

[0311] Because the curve for planned state 130 decreases over time due to aging. Figure 7 The comparison of the predicted state 130 at time (t) with the minimum state 17 is also shown: In order to identify whether the planned state 130 of the pump corresponds to the correct operating state of the pump, the planned state 130 is then compared with the minimum state 17, that is, the predicted values ​​of the flow rate (Q(t)) and delivery pressure (Pref(t)) material parameters 120 at said time are compared with their corresponding minimum values ​​Q. min and Pref min Compare them.

[0312] According to one implementation, a given maintenance decision (a decision about task 90 to be performed on device 2) regarding maintenance 6 to be performed is determined, followed by a given scenario 8, whether the maintenance decision in question will be sufficient, i.e., whether it will give device 2 a sufficient probability of zero failures during operation, within the context of said current scenario 8 and until the end of the planned cycle 70, or until the day before the planned maintenance 7. To this end, according to one implementation, at least one change is performed on at least one value of the task 90 of maintenance 6 to be performed. Then, when these values ​​are submitted to the model 16, the changed values ​​are introduced.

[0313] In all changes, at least one adequate decision is made for the planned state 130 of the device 2 to perform maintenance, and the planned state is greater than or equal to the lowest state 17 when the planned maintenance is performed.

[0314] In other words, and when planned maintenance 7 corresponds to the next maintenance step after maintenance 6 to be performed, a set of possible maintenance decisions (regarding task 90) are considered for maintenance 6 to be performed, and the intention is to determine the appropriate maintenance decision among the latter. For example, considering the case of equipment 2 such as a pump, for which two maintenance tasks 90 are identified as critical:

[0315] - Replace the pump bearing (with type A and type B as possible bearing types);

[0316] - Replace the pump impeller (with type A, type B, and type C as possible impeller types).

[0317] Then, the possible maintenance decisions for maintenance 6 to be performed are formed by twelve combinations of the following decisions:

[0318] - Three possible decisions regarding the bearing: do nothing (leave the bearing in place) or replace the bearing with a type A or type B bearing;

[0319] - Four possible decisions regarding the impeller: do nothing (leave the impeller in place) or replace the impeller with a type A, type B, or type C impeller.

[0320] Among these twelve possible maintenance decisions, the aim is to determine the most appropriate maintenance decision. Furthermore, if, among these possible decisions for the maintenance 6 to be performed, one maintenance decision is characterized by the vector [0; A] (i.e., the bearing is not replaced and the impeller is replaced by a type A impeller), and another maintenance decision is characterized by the vector [0; 0] (i.e., neither the bearing nor the impeller is replaced), these two maintenance decisions for the maintenance 6 to be performed vary with respect to the impeller (more precisely, the change in the value of task 90 of the maintenance 6 to be performed is related to the change in the "replace" and "not replace" values ​​of the impeller). For each possible maintenance decision for the maintenance 6 to be performed, the aforementioned customized simulation is performed by model 16 to determine the planned state 130 of equipment 2 at the end of the planned cycle 70 (i.e., the day before the planned maintenance 7) (taking into account the usage of equipment 2 after the maintenance 6 to be performed in the context of scenario 8 up to the time corresponding to the day before the planned maintenance 7, and taking into account the manufacturing and maintenance log 9 of equipment 2 on the day before the maintenance 6 to be performed, and taking into account the usage log 11 of equipment 2 on the day before the maintenance 6 to be performed). If the planned state 130 (which is therefore simulated by model 16 for the moment corresponding to the end of the planned period 70) meets the operating criteria of device 2 compared to the minimum state 17 required for the operation of device 2, then the maintenance decision is considered sufficient (referred to as the "sufficient maintenance decision") to give device 2 a sufficient probability of zero failures during operation until the end of the planned period 70 in the context of scenario 8.

[0321] For example, considering the case of device 2, such as a pump, its state 13 can be characterized at a certain moment by two material parameters 120 of state 13, namely flow rate (Q) and delivery pressure (Pref), the minimum values ​​required for the operation of the pump having (Q) min ) and (Pref min If the maintenance decision allows planned state 130 (for the moment corresponding to the end of planned cycle 70) to meet the pump's operating criteria, that is, for the moment corresponding to the end of planned cycle 70, the two conditions regarding material index 120 are: (Q>Q) min ) and (Pref>Pref min If the maintenance decision is deemed sufficient, then the simulation is repeated for each possible maintenance decision (with respect to task 90) for maintenance 6 to be performed. Therefore, the corresponding plan state 130 simulated by model 16 for the moment corresponding to the end of the plan period 70 allows for the classification of possible maintenance decisions for maintenance 6 to be performed between sufficient and insufficient maintenance decisions.

[0322] Therefore, in all variations of at least one value of the value of the maintenance task 90 to be performed, during the planned maintenance 7, for the planned state 130 of the device 2 which is greater than or equal to the minimum state 17, at least one sufficient decision is made to select the maintenance 6 to be performed.

[0323] According to one implementation scheme, for a given maintenance decision (regarding task 90 of maintenance 6 to be performed) and for each of the planned usage conditions 110 of scenario 8 of device 2 within the planning period 70, during the planning period 70 and until the end of the planned period, or until the day before the next planned maintenance 7 after maintenance 6 to be performed, the maximum possible limit 18 for maintaining zero failures of device 2 in the context of scenario 8 (i.e., the limit for uninterrupted operation of device 2 without failure risk) is determined.

[0324] Therefore, and for a given maintenance decision, the value of at least one of the planned usage conditions 110 in scenario 8 is modified. When the value is submitted to model 16, the modified value and the value of the maintenance decision are introduced. Then, during planned maintenance 7, a limit is calculated for at least one of the planned conditions 110 for planned state 130 of device 2, which is equal to the minimum state 17. In other words, scenario 8 of device 2 within planning period 70 is characterized by at least one controllable planned usage condition 110 whose value (“idle”) varies idly, and other possible planned usage conditions 110 (controllable and / or uncontrollable) are set to their corresponding values ​​for scenario 8. Model 16 is then used to solve, via an iterative method, the equation in which planned state 130 of device 2 at the end of planning period 70 corresponds to the minimum state 17, relative to the controllable planned usage condition 110 that remains idle.

[0325] Given a maintenance decision (regarding task 90 of maintenance 6 to be performed) and the controllable planned usage conditions 110 discussed, a possible limit (preferably the maximum limit 18) is thus determined, which remains in compliance with zero failures of device 2 (i.e., the limit of uninterrupted operation of device 2 with no risk of failure) until the end of the planned cycle 70 (other planned usage conditions 110 remain set to their respective values ​​in scenario 8).

[0326] For example, in the case where facility 3 is a truck and monitoring device 2 is the engine of the truck, considering scenario 8 between maintenance 6 to be performed and the next scheduled maintenance 7, the following planned usage conditions 110 are characterized:

[0327] -Controllable plan usage conditions 110: Average load to be transported (two tons = 2T), distance to be traveled (100,000 kilometers = 100,000 kilometers), average speed (90 kilometers per hour)

[0328] =90 km / h);

[0329] -Uncontrollable planned usage conditions 110: Ambient air temperature (20 degrees Celsius = 20℃), average slope of the route used (5% = 5%).

[0330] For scenario 8, the intention is to determine the maximum limit 18 of the average load to be transported, which conforms to the zero failure of the planned maintenance 7 until the maintenance to be performed 6, while other planned usage conditions 110 remain set to their respective target values ​​for scenario 8.

[0331] Using model 16, an iterative method is employed to solve the equations relative to the average load to be transported, where the planned state 130 of device 2 at the end of the planning period 70 corresponds to the minimum state 17. Thus, the maximum limit 18 of 2.5T is given by way of example (for the target value of 2T for the average load to be transported in scenario 8). Figure 8 In the above example of device 2, such as a truck engine, the usage conditions 110 are the load P to be transported, the mileage d to be traveled, the average speed V, the ambient air temperature T°, and the average gradient A of the route used. Figure 8 It is also assumed that the planned state 130 is defined by the material index 120 corresponding to the average compression pressure U of the cylinder, wherein the value of U in the lowest state 17 is... min This is the minimum value required for the proper operation of device 2. Figure 8 The calculation shows the maximum limit 18 (Plim) of the controllable usage conditions 110P associated with the fixed logs 9 and 11 of device 2 for a given maintenance decision related to maintenance 6 to be performed. For usage scenario 8 within the planned period 70, it is limited by the corresponding target average values ​​(Pc, dc, Vc, T°c, Ac) of the controllable (P, d, V) and uncontrollable (T°, A) usage conditions 110 (Pc = 2T; dc = 100,000km; Vc = 90km / h; T°c = 20℃; Ac = 5%, as described in the example above).

[0332] For the other four usage conditions 110 (d, V, T°, A) set as their respective average values ​​for scenario 8 (dc = 100,000 km, Vc = 90 km / h, T° = 20°, Ac = 5%), the maximum limit 18 of the controllable usage condition 110 (P) is determined iteratively by means of model 16 (in this example, it is determined to be Plim = 2.5T).

[0333] For the maximum limit 18Plim, the planned state 130 at the end of the planned period 70 corresponds to the required minimum state 17, that is, the material condition index 120 corresponds to the required minimum value (U = Umin ).

[0334] Similarly, Figure 9 The calculation of the maximum limit 18 (dlim) of the controllable operating condition 110d is shown. For the other four operating conditions 110 (P, V, T°, A) set as their respective average values ​​in scenario 8 (Pc = 2T; Vc = 90km / h; T°c = 20℃; Ac = 5%), the maximum limit 18 of the controllable operating condition 110(d) is determined iteratively using model 16 (in this example, it is determined to be dlim = 110,000km). For this maximum limit 18 (dlim), the planned state 130 at the end of the planned cycle 70 corresponds to the required minimum state 17, that is, the material condition index 120 corresponds to the required minimum value (U = U min For each of the other controllable planned usage conditions 110 in Scenario 8, the same operation is performed to determine its possible maximum limit 18, which remains consistent with zero failures of device 2 (i.e., the limit of uninterrupted operation of device 2 without failure risk) in the context of Scenario 8 during the planning period 70 and until the end of the planning period (i.e. until the day before the next planned maintenance 7 after maintenance 6 to be performed), and this is done for a given maintenance decision (regarding task 90 of maintenance 6 to be performed). In this way, for a given maintenance decision, the maximum limit 18 defines the scope of use for zero failures of device 2 in the context of Scenario 8 during the planning period 70 and until the end of the planning period: thus it is known that as long as device 2 is running, where the given controllable planned usage condition 110 remains below or equal to the value of its thus determined maximum limit 18, and other expected usage conditions 110 remain below or equal to their respective target values ​​in Scenario 8, then device 2 is likely to achieve zero failures. Then, the target value of the planned use condition 110 in scenario 8, as well as the value of the maximum limit 18 previously calculated for each of the controllable planned use conditions 110, can be graphically represented by a network mapping graph, with each axis assigned to the planned use condition 110. Thus, a polygon with the same number of vertices as the planned use condition 110 is obtained.

[0335] Therefore, a first polygon 180, referred to as "target polygon 180" (representing the target value of each of the planned use conditions 110 in scene 8), can be superimposed with a second polygon 181, referred to as "limit polygon 181" (representing the maximum limit 18 value of each of the controllable planned use conditions 110 and the target value of scene 8 for each of the uncontrollable planned use conditions 110), and then the target polygon 180 is enclosed by the limit polygon 181. Figure 10Consider the above example of device 2, such as a truck engine, where the usage conditions 110 are the load P to be transported, the distance d to be traveled, the average speed V, the ambient air temperature T°, and the average gradient A of the route used. Figure 10 It is also assumed that the planned state 130 is defined by the material state index 120 corresponding to the average compression pressure U of the cylinder, wherein the value of the minimum state 17 U min This is the minimum value required for the proper operation of device 2. Figure 10 The diagram illustrates the target polygon 180 and limit polygon 181 of device 2 associated with fixed logs 9 and 11 of the device for a given maintenance decision related to maintenance 6 to be performed. For usage scenario 8 within the planned period 70, these polygons are limited by the corresponding target average values ​​(Pc, dc, Vc, T°c, Ac) of controllable (P, d, V) and uncontrollable (T°, A) usage conditions 110 (Pc = 2T; dc = 100,000km; Vc = 90km / h; T°c = 20℃; V = 5%, as described in the example above).

[0336] Figure 10 The target polygon 180 is shown in solid lines along the corresponding axis:

[0337] - The corresponding target average values ​​(Pc, dc, Vc) of the controllable plan using condition 110 (P, d, V) in scenario 8 are represented by the opened padlock;

[0338] -The uncontrollable plan in scenario 8 uses the corresponding target average value (T°c, Ac) of condition 110 (T°, A), represented by a closed padlock.

[0339] Figure 10 "Limited Polygon 181" is shown as a dashed line along the corresponding axis:

[0340] - The values ​​of the corresponding maximum limits 18 (Plim, dlim, Vlim) of the controllable plan use condition 110 (P, d, V) in scenario 8 are represented by the opened padlock;

[0341] -The uncontrollable plan in scenario 8 uses the corresponding target average value (T°c, Ac) of condition 110 (T°, A), represented by a closed padlock.

[0342] According to one implementation, the maximum limit 18 of the planned use condition 110 of scenario 8 is repeated for each of the maintenance decisions determined to be sufficient (i.e., maintenance decisions regarding task 90 to be performed on device 2 during maintenance 6, which, in the context of scenario 8, during the planned cycle 70 and until the end of the planned week or until the day before planned maintenance 7).

[0343] Therefore, according to one implementation, when these values ​​are submitted to the model 16, the selected values ​​for the adequate maintenance decision are introduced. During planned maintenance 7, for the planned state 130 of the equipment 2, which is equal to the minimum state 17, the maximum limit 18 of the planned use condition 110 is calculated for the adequate maintenance decision.

[0344] For each maintenance decision determined to be sufficient for maintenance 6 to be performed and scenario 8 under discussion, when the maximum limit 18 of the controllable planned use condition 110 has been calculated, the target use (i.e., the target value of the planned use condition 110 in scenario 8) in the form of target polygon 180 and the limit use (i.e., the value of the maximum limit 18 of the controllable planned use condition 110 and the target value of the uncontrollable planned use condition 110 in scenario 8) in the form of limit polygon 181 are graphically represented in a similar network map, and this is accomplished by superimposing two polygons.

[0345] According to one implementation, a usage margin is determined for at least one of the planned usage conditions 110 of the scenario 8, as the deviation 182 between the corresponding value and the corresponding maximum limit 18.

[0346] In fact, for all controllable plan usage conditions 110 in scenario 8, the present invention quantifies the deviation 182 between their target values ​​and their corresponding maximum limits 18.

[0347] In other words, the representation and superposition of the target polygon 180 and the limit polygon 181 of the maximum limit 18 in the same network map of scenario 8 makes it possible to graphically represent the deviation 182 between the target value and the value of the maximum limit 18 of the controllable plan using condition 110 for each controllable plan.

[0348] For example, in the above case of a truck engine, for each controllable plan usage condition 110, the deviation 182 between the target value and the maximum limit 18 is therefore graphically represented as:

[0349] - For the average load to be transported: the deviation between the target value of 2T and the maximum limit value of 2.5T is 182;

[0350] - For the distance to be traveled: the deviation between the target value of 100,000km and the maximum limit value of 110,000km is 182;

[0351] - For average speed: the deviation between the target value of 90 km / h and the maximum limit value of 100 km / h is 182.

[0352] Figure 10For each controllable plan usage condition 110, a deviation 182 is shown between the target value and the maximum limit 18 of the controllable plan usage condition 110. Preferably, and for comparing comparable deviations 182, the axes of a network map of the target polygon 180 and the limit polygon 181 of the maximum limit 18, on which scene 8 is shown and overlaid, are normalized. In other words, for each controllable plan usage condition 110, the target value and its maximum limit 18 are represented on a scale, where each value is normalized (i.e., represented by the ratio between the value and the target value).

[0353] For example, in the case of the truck engine described above, and on the network map where its axis is normalized, the deviation 182 between the target value and the maximum limit 18 can thus be graphically represented for each controllable plan using condition 110:

[0354] - For the average load to be transported: the target value of 2T is represented by the ratio 1 (2T / 2T), while the maximum limit value of 18 is represented by the ratio 1.25 (2.5T / 2T), that is, there is a 25% deviation between the two values;

[0355] - For the distance to be traveled: the target value is represented by ratio 1 (100,000km / 100,000km), while the maximum limit 18 is represented by ratio 1.1 (110,000km / 100,000km), meaning there is a 10% deviation between the two values;

[0356] - For average speed: the target value is represented by ratio 1((90km / h) / (90km / h)), while the maximum limit 18 is represented by ratio 1.11((100km / h) / (90km / h)).

[0357] This indicates an 11% deviation between the two values. This normalization facilitates comparison between the target value and the maximum limit 18: it allows identification of whether the deviation 182 between the target value and the maximum limit 18 is a significant or acceptable controllable planned use condition 110, and whether the deviation 182 is not significant or acceptable under controllable planned use conditions 110. This is equivalent to identifying planned use conditions 110 that tolerate acceptable or unacceptable low or high margins before failure.

[0358] For example, in the above-described case of a truck engine, the deviation 182 between the target value and the maximum limit 18 is much larger for the average load to be transported (25%) than for the mileage to be traveled and the average speed (10% and 11%, respectively). Therefore, the average load to be transported tolerates a larger deviation 182 relative to its target value compared to the mileage to be traveled and the average speed (however, without causing a zero-failure problem). In other words, the average load has a greater confidence in its use before failure compared to the mileage to be traveled and the average speed, in order to benefit from zero failures until the end of the planning cycle 70. Therefore, during the planning cycle 70, the operator should pay closer attention to the target values ​​of the mileage to be traveled and the average speed in scenario 8 compared to the average load transported. Therefore, the present invention defines the pre-failure usage margin (MUBF) as the average of the normalized deviations 182 between the target value and the maximum limit 18 for each controllable planned usage condition 110, or the average deviation between the target polygon 180 of scenario 8 and the limit polygon 181 of the maximum limit 18.

[0359] In the example above, the pre-failure margin of use (MUBF) averaged over the three normalized deviations 182: 25%, 10%, and 11% is approximately 15%. Therefore, for maintenance decisions determined to be adequate for maintenance 6 and scenario 8 to be performed, the maximum limit 18 of the controllable planning condition 110 has been calculated, and the suitability of the maintenance decision (regarding the anticipated task 90 of maintenance 6 to be performed) is primarily quantified in scenario 8 and in the complete log of device 2 by the pre-failure margin of use (MUBF) metric associated with the maintenance decision.

[0360] For example, considering device 2 such as a truck engine, seven adequate maintenance decisions have been identified for this device regarding the maintenance to be performed, and a margin (MUBF) has been calculated for each maintenance decision. These decisions are:

[0361] - First decision (D1): Replace the diesel filter to achieve 3% MUBF;

[0362] - Second decision (D2): Replace the injector and make it possible to obtain 7% MUBF;

[0363] - Third decision (D3): Replace the injectors and diesel filter, and make it possible to achieve 9% MUBF;

[0364] - Fourth decision (D4): Replace the injection pump to enable 11% MUBF;

[0365] - Fifth decision (D5): Replace the injection pump and diesel filter, which will enable a 13% MUBF.

[0366] - Sixth decision (D6): Replace the injection pump and injectors, and enable a 20% MUBF gain;

[0367] - Decision 7 (D7): Replace all components (injection pump, injectors, and diesel filter) to achieve 25% MUBF.

[0368] For device 2 and its logs 9 and 11, note that decision D7 proved to be more suitable for scenario 8 than other decisions (especially decision D1). In fact, decision D1 enabled a 3% MUBF: to benefit from zero failures until the end of the planning cycle 70, the average deviation of planned use condition 110 from the target value of scenario 8 should be 3% of the average deviation 182.

[0369] Decision D7, in itself, allows 25% MUBF: In order to benefit from zero failures until the end of the planning cycle 70, the deviation 182 of the planned use condition 110 should be much greater than the average of 25% relative to the target value of scenario 8.

[0370] The use of margin (MUBF) thus makes it possible to quantify the suitability of each maintenance decision for the same planned use case 8.

[0371] According to one implementation, once a usage margin is obtained, the optimal maintenance decision is selected from the sufficient maintenance decisions as either a sufficient decision that enables an acceptable usage margin, or a sufficient decision 182 that enables at least one of the acceptable deviations. In other words, for each decision identified as sufficient for the maintenance 6 to be performed and the scenario 8 in question, the suitability of the sufficient maintenance decision (regarding the anticipated task 90 of the maintenance 6 to be performed) for scenario 8, quantified by the pre-failure usage margin (MUBF), has been calculated based on the maximum limit 18 of the controllable planned usage condition 110. Therefore, the invention preferentially selects the optimal maintenance decision as the decision that allows the anticipated usage margin. Thus, the optimal decision regarding the amount of the task 90 of the maintenance 6 to be performed is preferentially determined. In short, this optimal decision is tailored to the complete log of device 2, suitable for scenario 8 (to allow zero failures until the end of the planned cycle 70 and the anticipated pre-failure usage margin).

[0372] For example, considering the above situation with the truck engine, seven adequate maintenance decisions (D1 to D7) have been identified for the maintenance to be performed, and a margin (MUBF) has been calculated for each maintenance decision.

[0373] The operator may consider a 10% deviation of 182 between the target value in scenario 8 and the maximum limit value in scenario 18 to be sufficient:

[0374] - For the load to be transported (i.e., the operator estimates that the average load to be transported between the maintenance to be performed 6 and the next scheduled maintenance 7 will not exceed 10% of the target of 2T);

[0375] - For the mileage to be driven until the next maintenance (i.e., the operator estimates that the mileage to be driven between the maintenance to be performed 6 and the next planned maintenance 7 will not exceed 10% of the target of 100,000 km);

[0376] - For average speed (i.e., the average speed between the maintenance to be performed 6 and the next scheduled maintenance 7 is estimated by the operator to be no more than 10% of the target of 90 km / h).

[0377] Therefore, the optimal maintenance decision for maintenance 6 to be performed, which is associated with a margin greater than 10% (MUBF) and minimizes maintenance constraints, is decision D4, which has a margin of 11% (MUBF). Specifically, decisions D5, D6, and D7 also seem to allow for a margin greater than 10% (MUBF), but for the operator's intention, this would involve unnecessary over-maintenance with higher maintenance constraints. Therefore, for the operator, the optimal maintenance decision for maintenance 6 to be performed is decision D4 (i.e., the decision that includes replacing the jet pump).

[0378] The present invention also envisions selecting the optimal maintenance decision to be applied during the maintenance 6 to be performed, as a decision on the expected deviation 182 between the target value of scenario 8 and the value of the maximum limit 18, and performing the decision for one of the planned use conditions 110.

[0379] For example, considering the above situation with the truck engine, where seven sufficient maintenance decisions (D1 to D7) have been identified regarding the maintenance to be performed, and where the operator:

[0380] - It is estimated that the mileage driven the day before the next scheduled maintenance 7 should not exceed the target of 100,000 kilometers in scenario 8;

[0381] - It is estimated that the average speed on the day before the next planned maintenance 7 should not exceed the target of 90km / h in scenario 8;

[0382] - This means a deviation of at least 10% 182 between the target value (2T) and the maximum limit 18 of the average load to be transported only.

[0383] In this example, and for each of the seven adequate maintenance decisions, the deviation 182 of the load to be transported between the target value (2T) in scenario 8 and the value of the maximum limit 18 calculated by simulation through model 16 is then considered:

[0384] - Decision D1 allows a 5% deviation for the load to be transported.

[0385] - Decision D2 allows for an 11% deviation of 182 for the load to be transported;

[0386] - Decision D3 allows a 15% deviation for the load to be transported;

[0387] - Decision D4 allows for an 18% deviation of 182 for the load to be transported;

[0388] - Decision D5 allows for a 21% deviation of 182 for the load to be transported;

[0389] - Decision D6 allows a deviation of 33% for the load to be transported;

[0390] - Decision D7 allows for a 41% deviation of 182 for the load to be transported.

[0391] Considering the operator's assumptions and intentions, in this case, the optimal maintenance decision for maintenance 6 to be performed is no longer decision D4, but decision D2 (with an 11% deviation 182 between the target value and the maximum limit 18 of the load to be transported). Specifically, decisions D3 through D7 also result in deviations greater than 10% 182, but for the operator's intentions, this would involve unnecessary over-maintenance with higher maintenance constraints. Therefore, for the operator, the optimal maintenance decision for maintenance 6 to be performed in this case is decision D2 (i.e., the decision including replacing the injector).

[0392] The present invention may also select the best maintenance decision to be applied during maintenance 6 to be performed, as a decision on the expected average of the deviation between the target value of scenario 8 and the value of the maximum limit 18, and perform the decision for the expected portion of the planned use condition 110.

[0393] For example, considering the above situation with truck engines, seven adequate maintenance decisions (D1 to D7) have already been identified regarding the maintenance to be performed.

[0394] -Among them, the operator estimates that the average speed on the day before the next maintenance should not exceed the target of 90 km / h, and

[0395] -Among them, the average deviation of the operator's intention regarding the load to be transported and the mileage to be traveled is 10% (between the target value in scenario 8 and the value of the maximum limit 18).

[0396] In this example, and for each of the adequate maintenance decisions, then for the two planned use conditions 110 (load to be transported and mileage to be traveled), the average deviation 182 between the target value of scenario 8 and the maximum limit 18 is considered, where the maximum limit 18 is calculated by simulation allowed by model 16.

[0397] This example also assumes:

[0398] - Decision D1 allows for a deviation of 5% and 2% for the load to be transported and the mileage to be traveled, respectively, or an average deviation of 3%.

[0399] - Decision D2 allows for 11% and 5% deviation, or an average deviation of 8%;

[0400] - Decision D3 allows for a deviation of 15% and 6% (182), or a mean deviation of 10%;

[0401] - Decision D4 allows for a deviation of 18% and 7% (182), or an average deviation of 13%;

[0402] - Decision D5 allows for a deviation of 21% and 8% (182), or an average deviation of 15%;

[0403] - Decision D6 allows for a deviation of 33% and 13% (182), or an average deviation of 23%;

[0404] - Decision D7 allows for a deviation of 41% and 16% 182, which is an average deviation of 29%.

[0405] Considering the operator's assumptions and intentions, the maintenance decision for maintenance 6 to be performed is no longer decision D2 or decision D4 in this case, but rather decision D3 with a 10% average deviation. Specifically, decisions D4 through D7 also allow for an average deviation greater than 10%, but for the operator's intention, this involves unnecessary over-maintenance with higher maintenance constraints. Therefore, for the operator, the optimal maintenance decision for maintenance 6 to be performed is decision D3 in this case (i.e., the decision that includes replacing the injectors and diesel filter).

[0406] Therefore, for device 2 that continues operating until maintenance 6 is to be performed, and for scenario 8 that is to be performed after maintenance 6, the simulation allowed by model 16 enables the determination of:

[0407] - Among a set of sufficient maintenance decisions, the optimal maintenance decision is: the decision regarding the nature of the maintenance task 90 to be performed is tailored to the complete log of device 2, suitable for scenario 8 to allow zero failures until the end of the planned cycle 70 and to allow for a margin of availability (MUBF) before anticipated failures.

[0408] - The associated limit of use, defined by the maximum limit 18 of the controllable planned use condition 110 (which is associated with the best maintenance decision), i.e., the maximum possible use (complying with zero failures until the next planned maintenance 7 after the maintenance to be performed 6).

[0409] For example, considering the above situation with truck engines, seven adequate maintenance decisions (D1 to D7) have been identified for the maintenance to be performed 6, and the operator expects a mean time between failures (MUBF) of 10% for this set of controllable planned use conditions 110 (average load to be transported, mileage to be driven until the next maintenance, and average speed).

[0410] In this example (and considering the truck engine manufacturing and maintenance log 9 and usage log 11, as well as scenario 8 during the planning cycle 70), among the sufficient maintenance decisions regarding the maintenance 6 to be performed, the best maintenance decision is decision D4 (i.e., the decision that includes replacing the injection pump).

[0411] To benefit from zero engine failures until the end of the planned cycle 70, the limit usage to be observed during the planned cycle 70 is the limit usage associated with the optimal maintenance decision (i.e., the limits defined by the maximum limits 18 already calculated for the planned usage condition 110 and for the maintenance decision D4: 2.3T, with an 18% deviation 182 from the target value of 2T; 107,000 km, with a 7% deviation 182 from the target value of 100,000 km; 97 km / h, with an 8% deviation 182 from the target value of 90 km / h).

[0412] In other words, considering the mode used to calculate the maximum limit 18, as long as the engine is running, where the given controllable condition 110 remains below or equal to the value of its maximum limit 18 determined therefrom, and other conditions 110 remain below or equal to their respective target values ​​in scenario 8, it can be ensured that the truck engine will have zero failures until the end of the expected cycle 70.

[0413] Constructing the limit polygon 181 according to the maximum limit 18 of the planned use condition 110 in order to quantify its average deviation from the target polygon 180 of scene 8 (i.e., the failure-before-use margin (MUBF)) is a conceptual means envisioned in this invention.

[0414] In fact, for the same target scenario 8 and in each case of the full maintenance decision 6 to be performed, the limit polygon 181 and margin (MUBF) calculated according to the same method always make it possible to evaluate the applicability of each maintenance decision for target scenario 8 (taking into account the manufacturing and maintenance log 9 and the usage log 11 of device 2): the limit polygon 181 and margin (MUBF) thus make it possible to compare the maintenance decisions among them.

[0415] Furthermore, this invention anticipates and deduces all usage scenarios that meet the requirement of zero failures until the next maintenance date.

[0416] In practice, considering the pattern used to calculate the maximum limit 18, the limit polygon 181 and its maximum limit 18 define the limit of use that conforms to zero failures for device 2 until the end of the planned cycle 70. In this sense, zero failures are possible as long as device 2 is running, where the given controllable planned use condition 110 remains below or equal to its maximum limit 18, provided that other planned use conditions 110 remain below or equal to their respective target values ​​in scenario 8. Specifically, the limit polygon 181 cannot infer whether the use of device 2 still conforms to the expected zero failures if two controllable planned use conditions 110 can exceed their respective target values ​​while adhering to their respective possible maximum limits 18. For example, considering the above situation with truck engines, target scenario 8 is defined between the maintenance to be performed 6 and the subsequent planned maintenance 7 as follows:

[0417] - Controllable plan usage conditions 110: The average load to be transported is 2T, the distance to be traveled is 100,000km, and the average speed is 90km / h;

[0418] -Uncontrollable planned usage conditions 110: Ambient air temperature is 20℃, and the average slope of the route used is 5%.

[0419] In this example, it is also assumed that the maximum limit 18 is for an average load of 2.5T to be transported, a distance of 110,000km to be traveled, and an average speed of 100km / h.

[0420] In this example, as constructed, the limit polygon 181 indicates the following use case 8 as conforming to zero failures until the next planned maintenance 7:

[0421] - The average transport volume between two maintenance operations is 2.5T, the average mileage exceeds 100,000 kilometers, and the average speed is 90km / h;

[0422] - The average transport volume between two maintenance operations is 2 tons, the average mileage is over 110,000 kilometers, and the average speed is 90 km / h;

[0423] - The average transport volume between two maintenance operations is 2 tons, the average mileage exceeds 100,000 kilometers, and the average speed is 100 km / h.

[0424] However, as constructed, the limiting polygon 181 cannot in any way determine whether an average transport volume of 2.3T, an average mileage exceeding 105,000 km, and an average speed of 95 km / h between two maintenance operations constitute a usage scenario that meets the expected zero-failure requirement until the end of the next planned maintenance 7. According to one embodiment, the invention therefore anticipates different determinations of all usage scenarios that meet the zero-failure requirement until the date of the next maintenance. For this purpose, system 1 includes a graphical representation in the form of a graph, wherein at least one curve indicates the functional relationship between the maximum limit 18 of the first planning condition 110 and at least a second planning condition of said planning condition 110. This graph enables the determination of usage scenarios that meet the zero-failure requirement within the planned period 70 between the maintenance 6 to be performed and the next planned maintenance 7.

[0425] In other words, the present invention envisions constructing at least one graphical representation of the type of nomogram, for example, in the form of a set of curves on a graph, hereinafter referred to as a "graph", thereby allowing for richer information regarding extreme scenarios. For a given scenario 8 (defined by the target values ​​of controllable and uncontrollable planned use conditions 110), the constructed graph graphically represents the set of extreme uses that meet zero failures until the next planned maintenance 7, i.e., all n-tuples of the values ​​of controllable planned use conditions 110, each n-tuple indicating the maximum limit 18 of the nth controllable planned use condition 110 that meets zero failures until the next planned maintenance 7 for (n-1) controllable planned use condition 110 values ​​(the uncontrollable planned use conditions 110 of scenario 8 are locked at their respective values). For example, considering the above situation with truck engines, scenario 8 is considered between the maintenance to be performed 6 and the next planned maintenance 7, i.e., the scenario defined by the following planned use conditions 110:

[0426] - Controllable plan usage conditions 110: The average load to be transported is 2T, the distance to be traveled is 100,000km, and the average speed is 90km / h;

[0427] -Uncontrollable planned usage conditions 110: Ambient air temperature is 20℃, and the average slope of the route used is 5%.

[0428] In this context, and using a graphical representation of the distance to be traveled as a function of the average load to be transported, the graph is a set of curves, each associated with a given average speed between two consecutive maintenance operations. Each constant-speed curve is a set of triplets of values ​​for three controllable planned usage conditions 110 (average load to be transported; distance to be traveled; average speed), each triplet defining the limit of use that meets zero failures until the date of planned maintenance 7 (uncontrollable planned usage conditions 110 of scenario 8 (ambient air temperature and average gradient of the route used) are locked at their respective values ​​of 20°C and 5%). Therefore, for a given value of the average load to be transported (e.g., 2.3T), the curves associated with an average speed of 95 km / h make it possible to infer the maximum limit 18 of the distance to be covered until the next planned maintenance 7 (101,000 km in this example), or thus the maximum possible value that meets zero failures until the date of planned maintenance 7 (uncontrollable planned usage conditions 110 (ambient air temperature and average gradient of the route used) are locked at their respective target scenario 8 values ​​(20°C and 5%)). To this end, in order to construct a curve associated with a given value of controllable planned use condition 110, the present invention calculates, through an iterative method and simulation implemented by model 16, the maximum limit 18 of the controllable planned use condition 110 observable between the maintenance to be performed 6 and the planned maintenance 7 for different values ​​of the remaining controllable planned use condition 110 (uncontrollable planned use conditions 110 remain locked at their respective target scenario 8 values). In the aforementioned example of the truck engine, there are three controllable planned use conditions 110. The graph represents a set of curves, each associated with a value of the first controllable planned use condition 110 (e.g., average speed).

[0429] To construct a curve associated with the value of the first controllable planned condition 110 (e.g., an average speed of 100 km / h), for different values ​​of the third planned condition 110 (in this example, the average load to be transported), the maximum limit 18 of the second controllable planned condition 110 (e.g., the mileage to be traveled) is calculated through simulation by model 16, where the uncontrollable planned usage conditions 110 (ambient air temperature and the average gradient of the route used) are always locked at their respective target scenario 8 values ​​(i.e., 20°C and 5% in this example, respectively). Therefore, for example, the graph indicates that the triplet (2.3T; 101,000 km; 95 km / h) is one of the possible extreme usage limits. Figure 11 In the above example of device 2, such as a truck engine, the usage conditions 110 are the load P to be transported, the mileage d to be traveled, the average speed V, the ambient air temperature T°, and the average gradient A of the route used. Figure 11It is also assumed that the planned state 130 is defined by the material state index 120 corresponding to the average compression pressure U of the cylinder, wherein the value of the minimum state 17 U min This is the minimum value required for the proper operation of device 2. Figure 11 This illustrates an example of the use of such a graphical representation of device 2 associated with fixed logs 9 and 11 for a given maintenance decision related to maintenance 6 to be performed, for usage scenario 8 within the planned period 70, limited by the corresponding target average values ​​(Pc, dc, Vc, T°c, Ac) of controllable (P, d, V) and uncontrollable (T°, A) usage conditions 110 (Pc = 2T; dc = 100,000km; Vc = 90km / h; T°c = 20℃; Ac = 5%, as described in the example above).

[0430] exist Figure 11 The graph shown is a set of three curves that graphically represent the functional relationship between the distance to be traveled (d) and the average load to be transported (P). The curves are correlated with three average speeds (90 km / h, 95 km / h, and 100 km / h) given between two consecutive maintenance operations. For the two expected target average values ​​of P and V under controlled operating condition 110 (Pc = 2.3T; Vc = 95 km / h), the graph calculated by model 16 indicates the value of the maximum limit 18 (dlim = 101,000 km) of the final controlled operating condition 110 (d). Therefore, the usage scenario defined by the average of the three expected targets of the controllable planned usage condition 110 (Pc = 2.3T, dlim = 101,000km and Vc = 95km / h) and the two target values ​​of the uncontrollable planned usage condition 110 (T℃ = 20℃ and Ac = 5%) corresponds to the extreme scenario of expected zero failures until the end of the planned cycle 70, that is, the planned state 130 (U = Umin) of the device 2 at the end of the expected cycle 70 corresponding to the minimum state 17 required for correct operation.

[0431] According to one embodiment, once maintenance 6 has been performed and during the planned cycle 70, the invention refreshes the planned state 130 of device 2 at any time to its state at the end of planned cycle 70 or the day before the next planned maintenance 7. In other words, at a given moment during the planned cycle 70 in which scenario 8 of device 2 is being performed, considering the actual use of device 2 from the beginning of planned cycle 70, it determines whether device 2 maintains sufficient potential for zero failures during operation at the considered moment, in the context of the remainder of scenario 8, during the remainder of the process, and until the end of planned cycle 70 or until the day before the next planned maintenance 7 after maintenance 6 to be performed. However, the invention has anticipated selecting the optimal maintenance decision among sufficient maintenance decisions the day before maintenance 6 to be performed, which allows the planned state 130 at the end of planned cycle 70 (and considering scenario 8) to conform to the operation of device 2. However, the invention anticipates refreshing the planned state 130 to its state at the end of planned cycle 70 at any time during the planned cycle 70, as such refreshing is considered potentially necessary.

[0432] In reality, during the past period between the start of the planned operating cycle 70 and the time under consideration, the actual use of device 2 may differ from the anticipated use in scenario 8.

[0433] Furthermore, during the past portion of the planned cycle 70, device 2 may have experienced abnormal operational transients, particularly intrusions into areas nearing failure, or even into intermediate areas between these near-failure areas and areas of normal operation. However, the present invention is selective for aging, so the future behavior of the device will be affected by the sum of these abnormal transients. Therefore, the present invention anticipates refreshing the planned state 130 during operation to its state at the end of the planned cycle 70, including the sum of any abnormal transients that may have occurred from the start of the planned operating cycle 70 up to that point. Similarly, if the present invention anticipates determining the optimal decision for performing maintenance 6 and the associated limit usage (i.e., the maximum utilization rate that may be observed during zero-failure operation after the maintenance 6 until the next planned maintenance 7) the day before the maintenance 6 to be performed, then the present invention anticipates refreshing this limit usage during operation, as potentially necessary.

[0434] In reality, during operation, due to the actual use of device 2 (which is more restrictive for device 2 than initially envisioned scenario 8), the maximum use of device 2 is likely to shrink during the past part of the planned period 70, and due to any abnormal operational transients.

[0435] By refreshing the limit of zero-failure use during operation to meet the requirements until the next planned maintenance 7, the present invention anticipates allowing users to thus identify the maximum limit 18, to which they can push the use of the equipment without compromising zero failure or where the originally intended use must be limited in order to benefit from zero failure. For example, considering the above-described case of a truck engine, where the decision D6 of maintenance 6 to be performed (i.e., replacement of the diesel filter and injection pump) is adopted, and where, at the end of half of the planned cycle 70, it is proven that the average load of transport correctly conforms to the target value of 2T, and the mileage (50,000 km) correctly conforms to the proportion of the target value of 100,000 km at the end of the planned cycle 70, but the average speed is 105 km / h, exceeding the target value of 90 km / h in scenario 8 and the value of 100 km / h in the maximum limit 18.

[0436] Under these conditions (and apart from any anomalous transients during the past portion of planned cycle 70), continuing truck operation in the context of initial scenario 8 with its target values ​​(2T, 100,000km, 90km / h) would not meet the zero-failure target until the next planned maintenance 7. Therefore, it is recommended to refresh the calculations of the maximum limits 18 for various planned usage conditions 110, which the operator will have to observe in order to benefit from zero failures until the end of planned cycle 70. To this end, updated values ​​are submitted to model 16 after maintenance 6 has been performed. As previously stated, these values ​​correspond to at least one of the tasks 90 in manufacturing and maintenance log 9 and at least one task 90 of maintenance 6 performed.

[0437] These values ​​also correspond to at least one of the usage conditions 110 in the usage log 11 since the maintenance 6 was performed. These values ​​also correspond to the planned usage condition 110 of scenario 8.

[0438] Therefore, based on these values, model 16 refreshes the planned state 130 of device 2. Furthermore, the planned state 130 is compared with the minimum state 17 identified as required for the operation of device 2. In short, according to one embodiment, these values ​​are submitted to model 16 after maintenance 6 has been performed.

[0439] - At least one of the tasks 90 in the manufacturing and maintenance log 9 performed, and at least one of the tasks 90 in the maintenance log 6 performed.

[0440] and at least one of the following: usage conditions 110 of usage log 11 since the execution of maintenance 6

[0441] - and the planned usage condition 11 described in scenario 8,

[0442] The model 16 refreshes the planned state 130 of the device 2, which is compared with the minimum state 17 identified as required for the operation of the device 2. In other words, at a given moment during the next planning cycle 70, the predicted planned state 130 of the device 2 is refreshed to the state at the end of the planning cycle 70 or the day before the planned maintenance 7. For this purpose, and because maintenance 6 was performed and the device 2 is operational at the considered moment, the model 16 is used as a simulator customized for the device 2 by adjusting the independent variables submitted as input to the model 16, i.e.:

[0443] - Manufacturing and maintenance log of equipment 2, number 9;

[0444] - Regarding maintenance task 6, 90;

[0445] -Usage log 11 up to the time under consideration, followed by the past part of the planned run cycle 70 up to the time under consideration;

[0446] - The anticipated use of device 2 in the remainder of scenario 8, from the stated time and for the remainder of the planned operating cycle 70 until the day before the next planned maintenance 7. Therefore, the possibility of refreshing the expected state 130 of device 2 at the end of the planned operating cycle 70 after the future execution of the entire scenario 8 (taking into account the actual execution of task 90 for maintenance 6, and considering the actual use of facility 3 and device 2 from the start of the planned operating cycle 70 to the considered time) thus allows for the execution of various applications such as model 16 anticipated by the present invention the day before maintenance 6.

[0447] Specifically, the use of model 16 enables it to determine at any time during the planning cycle 70 whether the continuation of scenario 8 remains in compliance with zero failures until the end of the planning cycle 70 by comparing the planned state 130 with the minimum state 17 required for the operation of device 2.

[0448] Specifically, the use of Model 16 allows for updating the maximum limit 18, limit polygon 181, pre-failure usage margin (MUBF), and graphs of all extreme scenarios at any point during the planning cycle 70 to achieve zero failures until the next planned maintenance 7. This allows the user to identify limits to which the user can push the use of Device 2 without compromising zero failures, or where the user must limit the initially envisioned use to benefit from zero failures. At a given moment during the planning cycle 70, the invention updates the planned state 130 of Device 2 to its state at the end of the planning cycle 70 and after the execution of scenario 8. Similarly, at a given moment during the planning cycle 70, for the period between said moment and the end of the planning cycle 70, the invention updates the planned state 130 of Device 2 to its state at the end of the planning cycle 70 and after the execution of a scenario different from the initially envisioned use scenario 8.

[0449] In the independent variables submitted as input to model 16, instead of submitting the usage of device 2 in the context of scenario 8, from that time and for the remainder of the planned operating period 70 until the day before planned maintenance 7, the usage of device 2 in the context of an unforeseen scenario is actually submitted to model 16, from that time and for the remainder of the planned operating period 70 until the day before planned maintenance 7. Similarly, at a given moment during the planned period 70, and in order to perform a scenario different from the initially foreseen usage scenario 8, the invention refreshes the maximum limit 18, limit polygon 181, pre-failure usage margin (MUBF), and graphs of all extreme scenarios that conform to zero failures until the next maintenance 7 between that time and the end of the planned period 70: this allows the user to identify the limits to which the user can push the use of the device without compromising zero failures, or the user must limit the initially envisioned use to benefit from zero failures, and do so in order to perform a scenario different from the initially foreseen usage scenario 8. Therefore, the present invention allows operators to determine, during the planning cycle 70, the use of the equipment 2 that they can or should make in place of the initially foreseen scenario 8 in the context of both the initially foreseen scenario 8 and the unforeseen scenario 8, maintaining that use in accordance with the zero-failure requirement until the next maintenance date.

[0450] Advantageously, the simulation via Model 16 also enables the progressive determination of adequate maintenance decisions for continuous planned maintenance operations 7 throughout the lifespan of equipment 2, taking into account the operator's anticipated target use in the short and long term (i.e., continuous scenario 8). Therefore, System 1 enables the determination of various possible adequate maintenance schedules for anticipated use conditions throughout the entire lifespan of equipment 2. To this end, according to one embodiment, at least the following steps are performed.

[0451] First, it is assumed that at least the sufficient decision for maintenance 6 to be performed has been performed, and that scenario 8 has been performed up to the planned maintenance 7 following maintenance 6 to be performed.

[0452] Then, at least one change is made to at least one value of the task 90 of the planned maintenance 7. When these values ​​are submitted to the model 16, the changed values ​​and the values ​​of the next scenario 81 anticipated for the subsequent planning cycle 710 (i.e., the running cycle between the planned maintenance 7 and the subsequent planned maintenance 71) are introduced.

[0453] In all variations, during planned maintenance 71 following planned maintenance 7, for planned state 130 of the device 2 which is greater than or equal to the lowest state 17, at least one sufficient decision is selected for planned maintenance 7.

[0454] Once the adequate decision has been selected, the above steps are repeated for every planned maintenance operation throughout the equipment's lifespan. In short, once an adequate maintenance decision has been defined for maintenance 6 to be performed, a series of adequate maintenance decisions are redefined for planned maintenance operation 70 following maintenance 6. In fact, the day before maintenance 6 is to be performed (considering the maintenance decisions determined to be adequate for maintenance 6 and assumed to have been performed, taking into account scenario 8), model 16 enables the determination of adequate maintenance decisions for the next planned maintenance 7 following maintenance 6. In this sense, model 16 allows for repetition to progressively determine adequate maintenance decisions from one planned maintenance operation 7 to the next.

[0455] For this reason, and similarly before, for each possible maintenance decision following the planned maintenance 7 to be performed, the planned state 130 of device 2 is simulated as the state at the end of the next planned cycle 710 after the subsequent scenario 81, that is, the day before the next planned maintenance 71 (i.e., after the planned maintenance 7 in question). To this end, and taking these assumptions into account, model 16 is used as a simulator tailored for device 2 by adjusting the independent variables submitted as input, i.e.:

[0456] - The manufacturing and maintenance log 9 of device 2 (up to the day before the maintenance 6 to be performed) is modified using task 90 which assumes that maintenance 6 to be performed has already been performed;

[0457] - Regarding the task 90 concerning the possible maintenance decisions discussed in the planned maintenance 7;

[0458] - Use log 11 (up to the day before maintenance 6 to be performed), followed by scenario 8, which assumes it has already been performed during the planned cycle 70;

[0459] - The next scenario 81 for device 2 (during the next planned cycle 710 and until the day before the next planned maintenance 71). If the planned state 130 of device 2 (thus simulated by model 16 for the moment corresponding to the end of the next planned cycle 710 (i.e., the day before the next planned maintenance 71)) meets the operating criteria of device 2 compared to the minimum state 17 required for the operation of device 2, then the maintenance decision regarding the planned maintenance 7 after the maintenance to be performed 6 can be considered sufficient to give device 2 a sufficient probability of zero failure during the operation in the context of the next scenario 81, during the next planned cycle 710, and until its end. Thus, repetition is established. In fact, the day before the maintenance to be performed (level j maintenance) (considering that the maintenance decision is deemed sufficient for the maintenance to be performed and is assumed to have been performed, and considering scenario 8 which is assumed to have been performed during the next planned operating cycle 70), model 16 has enabled the determination of sufficient maintenance decisions for planned maintenance 7 (level j+1 maintenance), i.e., a zero-failure decision authorized for the next scenario 81 during the next planned cycle 710 and until its end. Furthermore, the present invention enables the determination of sufficient decisions for the maintenance to be performed 6. Therefore, the simulation of model 16 enables the determination of a sufficient sequence of maintenance decisions for any planned maintenance 7 by iterative step-by-step determination, and doing so until the expected planned maintenance level 7 (level j+n maintenance) during the lifespan of device 2. Figure 12 and Figure 13 In the process, the device associated with logs 9 and 11 of device 2 was considered. Figure 12 In the planning cycle of 70, the use case 8 is (in) Figure 13 (In the next planning cycle 710, the next scenario 81) Figure 12 and Figure 13 The stepwise simulation of the planned state 130 of equipment 2 the day before each planned maintenance operation 7, 71 is shown, along with the repeated determination of sufficient maintenance decisions regarding the planned maintenance operations 7, 71: Model 16 enables the determination of sufficient maintenance decisions regarding the maintenance operation 6 to be performed (at level j), and Model 16 then enables the determination of sufficient maintenance decisions regarding the next planned maintenance 7 at level j+1. First, Figure 12As shown, for possible maintenance decisions regarding the maintenance 6 to be performed, model 16 enables the prediction of the planned state 130 of device 2 at the end of the planning cycle 70: specifically, model 16 thus enables the determination of whether the possible decisions are sufficient to allow for the expected zero failures until the end of the planning cycle 70 (depending on whether the planned state 130 of device 2 at the end of the planning cycle 70 is sufficient relative to the minimum state 17). Model 16 clearly enables the determination of a sufficient maintenance decision regarding the maintenance 6 to be performed among the possible maintenance decisions. In the second step, Figure 13 As shown, for a maintenance operation at level j (in this case, for maintenance 6, which is assumed to have already been performed) identified as a sufficient maintenance decision, and for the next maintenance at level j+1 (in the upcoming case, for the next planned maintenance 7), model 16 enables the prediction of the planned state 130 of device 2 at the end of the next planned cycle 710: model 16 thus enables the determination of whether the latter possible decision is sufficient to allow for the expected zero failures until the end of the next planned cycle 710 (based on whether the planned state 130 of device 2 at the end of the next planned cycle 710 is sufficient relative to the minimum state 17). Therefore, Figure 13 As shown, for maintenance operations at level j that are identified as sufficient maintenance decisions, Model 16 enables the determination of sufficient maintenance decisions among the possible maintenance decisions for the next maintenance at level j+1.

[0460] like Figure 12 and Figure 13 As shown, system 1 thus enables the determination of sufficient maintenance decisions over the lifespan of equipment 2 by repetition, from one maintenance operation to another, i.e., progressively from maintenance operation of level k to maintenance operation of level k+1. Specifically, Figure 12 and Figure 13 The diagram illustrates how model 16 is used to determine the planning status 130 of device 2 during planning cycle 70, and to determine the planning status 130 of device 2 during the next planning cycle 710. Figure 12 and Figure 13 This actually illustrates the construction of the input independent variables for Model 16 in both cases. Specifically, the stepwise simulation of the planned state 130 the day before each planned maintenance 7 is due to the fact that the simulation of the planned state 130 of the next planned cycle 710 after maintenance at level j+1 (as in...) Figure 13 (As can be seen in the text) The integration of this simulation has enabled the identification of maintenance at level j as sufficient for maintenance decisions (such as in...). Figure 12(As can be seen in the text). Similarly, the present invention progressively determines the continuous planned state 130 of device 2 during two planning cycles 70 and 710. To this end, two maintenance decisions are considered, respectively, regarding the maintenance to be performed 6 (level j maintenance) and the planned maintenance 7 (level j+1 maintenance), as well as two usage scenarios 8 and 81 for each of the planning cycles 70 and 710. In this respect, Figure 14 An example of two curves for the planned state 130 of device 2, such as a pump, is shown (in this example simplified to its “flow rate Q” material index 120).

[0461] For each moment of planned cycle 70 (from maintenance level j to maintenance level j+1) and then for each moment of the next planned cycle 710 (from maintenance level j+1 to maintenance level j+2), model 16 enables the prediction of the planned state 130 of the pump. Specifically, the simulation of model 16 highlights the necessity of operating the pump during maintenance at level j+1 in order to benefit from the expected zero failures (otherwise, the pump will fail before maintenance at level j+2, as shown by the extended dashed line of the curve).

[0462] Figure 14 Particularly noteworthy is the disconnection of flow on either side of maintenance at level j+1 due to the maintenance decision performed on the pump (on the day of maintenance at level j+1). This maintenance decision at level j+1 is particularly sufficient because it allows for sufficient flow up to the day before maintenance at level j+2. Similarly and more generally, the invention progressively determines the continuous planned states 130 of the device 2 in as many consecutive planned cycles 70, 710 as possible, which will be taken into account over the service life of the device 2. For this purpose, a series of maintenance decisions are considered regarding the maintenance 6 to be performed (maintenance at level j) and then regarding the various planned maintenance operations 7, 71 (maintenance at levels j+1 to j+k) prior to the planned cycles 70, 710 in question, and a series of usage scenarios 8, 81 are considered for each of the planned cycles 70, 710 in question. In this respect, Figure 15 Examples of two general curves for the planned state 130 of device 2, such as a pump, are shown (in this example simplified to its “flow rate Q” material index 120).

[0463] Model 16 enables the prediction of the planned state 130 of the pump at various moments in the cycle between maintenance at level j and maintenance at level j+k, and at various moments in the cycle between maintenance at level j+k and maintenance at level j+n. These curves are obtained by simulating each operating cycle between two consecutive maintenance operations using Model 16. Figure 15 The continuous planned state 130 of the pump (defined by the vertical dashed lines in the middle) is generated by the prediction.

[0464] Therefore, it is important to emphasize that the pump needs to be operated during level j+k maintenance in order to benefit from the expected zero-failure period during each operation between level j and level j+n maintenance. Similarly, Figure 15 In particular, it highlights the disconnection of flow on either side of maintenance at level j+k due to maintenance decisions made on the pump (on the day of maintenance described at level j+k).

[0465] Therefore, the present invention enables the determination of the planned state 130 of the device 2, that is, the state at any moment during any subsequent planned cycle 710 of the device 2's lifespan (in the context of a given next scenario 81 and considering a maintenance decision that is sufficient for the planned maintenance 7).

[0466] Because of this predictive capability, and similarly to the previous invention, the present invention enables the identification, for each subsequent planned cycle 710 in the lifespan of the device 2, of the limits to which the use of the device can be pushed to without compromising the expected zero failures, or the limits to which the initially envisioned use should be limited in order to benefit from zero failures.

[0467] In fact, for each subsequent planned cycle 710 in the lifespan of device 2 (after a given next scenario 81 and considering that a maintenance decision is deemed sufficient for the corresponding service maintenance 7), the present invention:

[0468] - Similar to before, determine the maximum limit 18 of the controllable plan usage condition 110;

[0469] -Therefore, during the next planned cycle 710, the limit of zero-failure use in the next scenario 81 is defined, and this is done until the day before the next planned maintenance 71;

[0470] - Similar to and prior to this, the calculated maximum limits 18 of the planned use conditions 110 quantify the applicability of the adequate maintenance decision (regarding task 90 of the planned maintenance 7) to the next scenario 81 by means of the failure-before-use margin (MUBF) or by means of the deviation 182 between the target value of the next scenario 81 and at least one of the maximum limits 18 of the controllable planned use conditions 110 of the next scenario 81.

[0471] Similar to before, a graph is constructed indicating the set of extreme scenarios valid for the next planned maintenance cycle 710. Therefore, the simulation of model 16 enables the determination of sufficient maintenance decisions for any planned maintenance 7 by iterative, stepwise determination, and doing so until the expected planned maintenance level 7 is reached. Furthermore, for each of the next planned maintenance cycles 710, the simulation allowed by model 16 enables the determination of the maximum limit 18, limit polygon 181, pre-failure service margin (MUBF), and extreme scenario graphs for each controllable planned use condition 110 corresponding to the sufficient maintenance decisions 81.

[0472] According to one implementation, the optimal decision is selected from at least the sufficient decisions for the corresponding maintenance. In other words, for each planned maintenance step 7 during the life of device 2, similar to the previous selection for maintenance 6 to be performed, the optimal maintenance decision is chosen from the maintenance decisions that have already been determined to be sufficient. The optimal decision is selected preferentially with respect to the pre-failure margin of use (MUBF) allowed by each considered sufficient maintenance decision, or with respect to at least one deviation 182 between the target value of scenario 8 and the value of the maximum limit 18 for at least one planned use condition 110 for said scenario 8.

[0473] Therefore, the present invention anticipates optimizing each planned maintenance operation 7 during the lifespan of device 2 by determining the optimal maintenance decision for each maintenance step during the lifespan of device 2 and considering successive scenarios 8 during the lifespan of device 2. This decision regarding the nature of the task 90 of the planned maintenance 7 is tailored to the complete log of the device, suitable for scenario 8 to allow zero failures until the end of the planned cycle 70, and allows for the operator's expected pre-failure margin of use (MUBF). In other words, the present invention determines this series of optimal maintenance decisions for each of the planned maintenance operations 7 for the remaining lifespan of device 2. Furthermore, for each of the planned cycles 70 during the lifespan of device 2, the present invention determines the maximum limit 18 (for each controllable planned use condition 110 corresponding to scenario 8), the limit polygon 181, the pre-failure margin of use (MUBF), a graph of the limit scenarios, and does so by considering maintenance decisions identified as optimal. Thus, for each of the planned cycles 70 during the lifespan of device 2, the present invention makes it possible to identify the limits to which the use of the device can be pushed without compromising zero failures, or the limits to which the initially envisioned use should be limited to benefit from zero failures. In other words, the present invention determines the series of extreme uses of the device 2 by repeatedly determining each of the planned cycles 70 in the remaining lifespan of the device 2.

[0474] As previously stated, for maintenance 6 to be performed, model 16 enables the distinction between sufficient and insufficient maintenance decisions. In the case of a sufficient maintenance decision, the maintenance decision is suitable for use scenario 8 because it is sufficient to allow zero failures until the end of the planned period 70. In the case where the maintenance decision is determined to be insufficient, the operation of device 2 in the context of use scenario 8 results in a failure of device 2 before the end of the planned period 70. In the latter case, the calculation of the failure date 19 is reasonable. Therefore, according to one embodiment, when the maintenance decision is insufficient and the planned state 130 (during the planned period 70) is less than the minimum state 17, the failure date 19 is determined for the insufficient maintenance decision.

[0475] Failure date 19 corresponds to the moment when planned state 130 equals minimum state 17. This failure date 19 can be determined through simulation using model 16. Therefore, and for a given inadequate maintenance decision regarding maintenance 6 to be performed, the following procedure is performed for planned period 70 and its scenario 8. A given moment is considered during the process of planned period 70 and scenario 80, corresponding to that moment. By adjusting the independent variables submitted as input to model 16, model 16 is used as a simulator customized for device 2, i.e.:

[0476] - Manufacturing and maintenance log of equipment 2, number 9;

[0477] - Inadequate maintenance decisions regarding the maintenance to be performed;

[0478] - Device 2 Usage Log 11;

[0479] - The use of device 2 is anticipated within the context of partial scenario 80, which is defined up to the time of the considered planning period 70. Model 16 is then used to solve the equations relative to said time, in which the planned state 130 of device 2 at said time corresponds to the minimum state 17. For the considered inadequate maintenance decision and scenario 8, the failure date 19 is determined as the solution to said equations.

[0480] For inadequate maintenance decisions, assuming maintenance 6 to be performed, the same process can be applied to determine the failure date 19 of equipment 2 during the next planning cycle 710, which is associated with the inadequate maintenance decision regarding planned maintenance 7 and with the next scenario 81. Figure 16The study considers device 2 such as pumps (associated with their logs 9 and 11), scenario 8 during the planned cycle 70 (including between the maintenance 6 (level j) to be performed and the next planned maintenance 7 (level j+1)), scenario 8 and part of scenario 80 associated with a certain moment, and insufficient maintenance decisions regarding the maintenance 6 to be performed. Figure 16 The curve of the planned state 130 of the pump is shown (simplified in this example to its "flow rate Q" material index 120), and model 16 enables predictions for each moment of the planned cycle 70.

[0481] The curve for planned state 130 decreases over time due to aging until it reaches the lowest planned state 130, equal to the lowest planned state 17 before the planned maintenance date 7. In the case shown, the possible maintenance decision is actually an inadequate one: it does not allow planned state 130 to exceed the lowest planned state 17 until the end of the planned cycle 70. The intercept of the curve with the lowest planned state 17 corresponds to the pump failure date 19.

[0482] According to one implementation, for maintenance 6 to be performed without any maintenance decision being identified as sufficient, and for at least one given insufficient decision regarding said maintenance 6, the end-of-life date of device 2 is determined as the failure date 19 associated with said maintenance decision. In other words, if all possible maintenance decisions for maintenance 6 to be performed prove to be insufficient (in order to allow for zero failures during the planned cycle 70 and until its end, in the context of scenario 8), then device 2 is at the end of its life, and the maintenance 6 to be performed corresponds to the last maintenance in the life of device 2. When the maintenance 6 to be performed is therefore identified as the last maintenance in the life of device 2 and for a given maintenance decision, the end-of-life date of device 2 is determined as the failure date 19 associated with said maintenance decision.

[0483] When planned maintenance operation 7 is identified as the last maintenance in the life of device 2, the same process is applied to determine the end date of the life of device 2 associated with the given maintenance decision and the next scenario 81.

[0484] According to one implementation, and for maintenance 6 (or planned maintenance 7) identified as the last maintenance to be performed during the life of equipment 2, a maintenance decision is selected from the possible maintenance decisions that optimizes any combination of the constraints between the failure date 19, the last use margin (MLU), and the task 90 of the last maintenance decision.

[0485] In other words, for maintenance 6, which is identified as the last maintenance to be performed during the lifespan of device 2, all possible maintenance decisions were considered. For each of these maintenance decisions, the minimum operating duration (D) for device 2 during the planned cycle 70 was considered. MIN The minimum operating duration is determined based on the constraints of task 90 related to the maintenance decision (i.e., based on the nature and quantity of task 90).

[0486] For example, considering the case of device 2 such as a truck engine, five possible maintenance decisions have been identified regarding the last maintenance during the life of device 2, and a minimum expected operating duration (D) has been determined for each of these five possible maintenance decisions. MIN ),Right now:

[0487] - Decision E1: Replace the diesel filter (with an expected minimum operating duration equal to 3 days D). MIN );

[0488] - Decision E2, replace the injectors and diesel filter (with an expected minimum operating duration of 1.1 months D) MIN );

[0489] - Decision E3, replace the injection pump and injector (with an expected minimum operating duration equal to 2 months D) MIN );

[0490] - Decision E4: Replace the injection pump, injectors, and diesel filter (with an expected minimum operating duration of 2.1 months). MIN );

[0491] - Decision E5: Replace the injection pump, injectors, diesel filter, and cylinder sections (with an expected minimum operating duration of 12 months). MIN Among possible maintenance decisions, then based on the service life extension (SLE) associated with said maintenance decision.

[0492] (That is, the duration between the last maintenance during the lifespan of device 2 and the date of the end of the lifespan of device 2) whether it meets the requirements for considering the maintenance decision (i.e., SLE>D). MIN The expected minimum operating duration (D) MIN To select each decision.

[0493] For example, considering the above-mentioned situation with truck engines, five possible maintenance decisions were identified regarding the last maintenance during the engine's lifespan, with each decision determining the expected minimum operating duration (D). MIN and Service Life Extension (SLE), namely:

[0494] - Decision E1, replace the diesel filter (with a minimum expected operating time of 3 days D) MIN And a one-month extension of SLE, then the injector is the limiting factor);

[0495] - Decision E2, replace the injectors and diesel filter (with an expected minimum operating duration of 1.1 months D) MIN And a 1.5-month extension of SLE, then the injection pump is the limiting factor;

[0496] - Decision E3, replace the injection pump and injector (with a minimum expected operating duration of 2 months D) MIN And a 0.5-month lifespan extension for SLE, then the diesel filter is the limiting factor;

[0497] - Decision E4: Replace the injection pump, injectors, and diesel filter (with an expected minimum operating duration of 2.1 months). MIN And a 6-month extension of SLE, followed by segmentation, is a limiting factor;

[0498] - Decision E5, replace the injection pump, injectors, diesel filter, and cylinder sections (with a minimum expected operating duration of 12 months). MIN And a 7-month extended SLE, then the crankshaft is the limiting factor).

[0499] Decision E3 is not allowed to be related to the expected minimum running duration D. MIN (2 months) Compatible lifespan extension SLE (0.5 months). The same applies to decision E5. Therefore, the chosen decisions are decisions E1, E2, and E4, each of which, conversely, allow for a lifespan greater than the expected minimum operating duration D. MIN The lifespan is extended by SLE. For each possible maintenance decision selected for the last maintenance in the lifespan of device 2, the present invention determines for each controllable planned usage condition 110 of scenario 8:

[0500] - The final use limit is defined as the minimum expected operating duration D of device 2 after the maintenance is performed. MIN The invention calculates this final use limit in a manner similar to the maximum limit 18. Therefore, the final use limit defines the maximum possible limit D for zero failures up to the expected minimum operating duration of device 2. MIN The limits of use that must be adhered to during the 70-day planned period at the end of the project;

[0501] - For the controllable plan usage condition 110, the deviation 182 between the target value in scenario 8 and the value of the final usage limit.

[0502] For each possible maintenance decision selected, the margin (MLU) is finally used, and thus for all controllable plans in scenario 8, condition 110 is derived as the average of the deviations (i.e., in a manner similar to the calculation of the margin before failure (MUBF)).

[0503] For example, considering the above situation with truck engines, three possible maintenance decisions (E1, E2, and E4) were selected for the last maintenance during the equipment's lifespan, and the final margin of use (MLU) for each decision was calculated, namely:

[0504] - Decision E1: Replace the diesel filter (with a 3% margin MLU);

[0505] - Decision E2, replace the injectors and diesel filter (with a 4% margin MLU);

[0506] - Decision E4: Replace the injection pump, injectors, and diesel filter (with a 12% margin MLU). Then, among the previously selected final maintenance decisions, the optimal final maintenance decision is determined in a non-restrictive manner:

[0507] -It optimizes service life extension (SLE);

[0508] -or it optimizes the last-use margin (MLU);

[0509] - or its optimized Last Maintenance Efficiency (LME), which is defined as the lifespan extension SLE versus the minimum operating duration D. MIN The difference between the two values, which is related to the lifespan extension SLE (that is, LME = 1 - D). MIN / SLE);

[0510] -or it may optimize any other criteria, combining at least two of the following criteria under the operator's judgment: Service Life Extension (SLE), Last Use Margin (MLU), and Last Maintenance Efficiency (LME);

[0511] - Alternatively, in the absence of a maintenance decision that meets the above criteria, the decision is to replace equipment 2 with a new one.

[0512] For example, considering the above situation with the truck engine, three possible maintenance decisions (E1, E2, E4) were selected for the last maintenance during the lifespan of device 2 (with the condition SLE>D allowed). MIN ),Right now:

[0513] - Decision E1: Replace the diesel filter (with 1-month extended SLE, 90% efficiency LME, and 3% margin MLU);

[0514] - Decision E2, replace the injectors and diesel filter (with a 1.5-month extended SLE, 26% efficiency LME and 4% margin MLU);

[0515] - Decision E4, replace the injection pump, injectors and diesel filter (with 6-month extended SLE, 65% efficiency LME and 12% margin MLU).

[0516] Furthermore, in this example, it is selected as the best decision for the last maintenance based on the criteria of simultaneously optimizing service life extension (SLE), last-use margin (MLU), and last-maintenance efficiency (LME).

[0517] From an operational perspective, decisions E1 and E2 are meaningless (the lifespan extension SLE is only 1 month and 1.5 months, respectively). Furthermore, these decisions provide excessively low usage confidence before failure (with margins of 3% and 4% MLU). Conversely, decision E4 is more operationally meaningful with a 6-month lifespan extension SLE, acceptable usage confidence before failure (margin MLU > 10%), and a very acceptable efficiency LME (65%).

[0518] Therefore, on the date of the last maintenance in the engine's life, it is reasonable not to replace the engine with a new one and to perform the final maintenance decision E4 (i.e., the maintenance decision including replacing the injection pump, injectors, and diesel filter). Thus, the present invention anticipates indicating an extension of engine life. For maintenance 6 to be performed, which is identified as the last maintenance in the life of device 2, the present invention therefore selects from the possible maintenance decisions an optimization of the lifespan extension SLE (which takes into account the failure date 19), the final use margin (MLU), and the final maintenance efficiency LME (which takes into account the duration D). MIN And therefore, maintenance decisions are made considering any combination of constraints between the last maintenance task 90. ​​Thus, the present invention anticipates and instructs an extension of the service life of device 2, and thus determines the optimal service life of device 2. It should be noted that this instruction for extending the service life of the device takes into account both the manufacturing and maintenance log 9 and the usage log 11 of device 2, as well as the usage scenario 8 of device 2 in the context of its extended service life.

[0519] The same process can be applied to determine the best decision regarding the last maintenance of any planned maintenance 7 that is identified as the last maintenance during the lifespan of device 2 and is associated with the next scenario 81.

[0520] According to one embodiment, the present invention determines the optimal decision 4 for manufacturing equipment 2, namely the critical task 90 of manufacturing process 4, which optimizes the planned state 130 of equipment 2 at the end of planning cycle 70 and for scenario 8. For this reason, and for at least two virtual equipment of the same series of equipment 2 associated with individual manufacturing decisions...

[0521] - The simulation is performed by submitting the at least two manufacturing decisions and at least one planned use scenario 8 to the model 16;

[0522] Model 16 generates at least one planned state 130 for each of the two virtual devices;

[0523] - Select one of the at least two manufacturing decisions based on the planning status 130 of the two virtual devices;

[0524] - The selected manufacturing decisions are accessible to the designer / manufacturer. In other words, the invention considers a variety of possible manufacturing decisions 4 (i.e., various possible combinations of key tasks 90 for said manufacturing process 4). A virtual device in this series is associated with each combination.

[0525] For example, considering the case of device 2 such as a pump, the key task 90 of manufacturing process 4 is simplified to the following task 90:

[0526] - "Placement and selection of pump bearings" has two possible options for bearing type: Type A and Type B;

[0527] - "Place and select pump impeller", with three possible options for impeller type: Type A, Type B and Type C.

[0528] The various possible combinations of the critical task 90 of manufacturing process 4 (i.e., the various possible decisions of manufacturing process 4) are therefore six combinations [bearing type; impeller type]: [A; A][A; B][A; C][B; A][B; B][B; C]. Therefore, considering various virtual devices, each virtual device is associated with manufacturing decision 4 from the various possible combinations of manufacturing 4 running in the first operating cycle 5 (simplified to a single operating cycle, i.e., excluding any previous maintenance 10) within the context of assuming it has already been performed in facility 3 and assuming it has already been performed in usage log 11 (the same for all virtual devices considered). In addition to the maintenance decision regarding the maintenance 6 to be performed (the same for all virtual devices considered), scenario 8 (the same for all virtual devices considered) is also considered.

[0529] For each virtual device, the present invention then uses the model 16, customized for the virtual device, as a simulator to determine the planned state 130 of the virtual device at the end of the planned period 70 by adjusting the independent variables submitted as input to the model 16.

[0530] - Manufacturing and maintenance log 9 (simplified to manufacturing decisions 4 specific to the virtual device);

[0531] - Maintenance decisions regarding Maintenance 6 to be performed (common to all virtual devices considered);

[0532] - Use log 11 (which is common to all virtual devices considered);

[0533] -Scenario 8 (common to all virtual devices under consideration).

[0534] Therefore, for all virtual devices associated with their corresponding manufacturing decisions 4 (and any other parameters also considered), the planning status 130 of the virtual devices is compared at the end of the planning cycle 70. After comparing the planning status 130, the optimal planning status 130 indicates the optimal manufacturing decision 4 (the best combination of critical manufacturing tasks 90).

[0535] According to one implementation, the simulation of model 16 determines the optimal manufacturing decision 4, namely, the task 90 of optimizing the optimal lifecycle of equipment 2 (the optimal lifecycle includes a series of optimal maintenance decisions for each of the maintenance to be performed 6 and the planned maintenance operations 7, 71 during the lifecycle of equipment 2, a series of extreme uses (i.e., maximum limits 18) for each of the planned cycles 70, 710 during the lifecycle of equipment 2, and the optimal service life of equipment 2), and does so for a given usage scenario (a set of consecutive scenarios 8, 81 of the various planned cycles 70, 710 during the lifecycle of equipment 2). For this purpose, for at least two virtual devices of the same series of equipment 2 associated with the two separate manufacturing decisions 4:

[0536] - For each of the at least two manufacturing decisions 4, repeating simulations are performed in a similar manner to determine a series of optimal maintenance decisions, a series of maximum limits 18, and the optimal lifespan of the virtual device;

[0537] - Select the optimal manufacturing decision based on the simulation results;

[0538] - The selected optimal manufacturing decision 4 is accessible to the designer / manufacturer. In other words, since model 16 is able to simulate the planned state 130 of the same series of virtual devices in the same manner as the actual device 2 (as previously described), the present invention determines the optimal lifespan of the virtual device in the same manner as the actual device 2. Preferably, as many optimal lifespan simulations as virtual devices are performed (i.e., as many simulations as manufacturing decision 4). The optimal manufacturing decision 4 is then selected based on the results of the simulations, that is, the manufacturing decision that authorizes the optimal lifespan. Thus, system 1 enables the determination of manufacturing decision 4 that optimizes the complete lifespan of the series of devices 2 for the benefit of the manufacturer. According to another embodiment, model 16 also enables the evaluation of the sensitivity of the behavior of device 2 (particularly planned state 130, optimal lifespan) for usage conditions 110 and the identification of optimal values ​​for each usage condition 110: this indicates that there are many possible methods to enhance the protection of the device against usage conditions 110, or even optimize the operating point of the device.

[0539] According to one embodiment, the present invention provides operators with overall oversight of all equipment 2 in a fleet (e.g., a group of vehicles) incorporated into their facility 3. For this purpose, system 1 is applied to a group of several devices 2 belonging to a single operator's fleet. The results obtained for each of the devices 2 are combined, and the results are then accessible at least to the operator.

[0540] In other words, System 1 delivers and continuously updates information related to the optimal usage cycle of each device 2 in their group to a given operator for one or more anticipated usage scenarios (i.e., all usage scenarios 8, 81 in the short and long term of the lifespan of each device 2), specifically:

[0541] - The schedule for the best future maintenance of each device 2 (i.e., a series of best maintenance decisions regarding the maintenance to be performed 6 and a set of planned maintenance operations 7, 71 during the service life of each device 2);

[0542] -A series of possible extreme uses for each device 2 over the long term of its service life (i.e., for each device 2 during its service life and for each planned operating cycle 70, 710 in the context of scenarios 8, 81);

[0543] - The optimal lifespan of each device 2. Then, system 1 enables timely prediction and updating at all times:

[0544] - For the benefit of the participants in the "supply chain", the nature of the components required for each maintenance operation in the maintenance operations to be performed during the service life of each device 2 (i.e., the maintenance to be performed 6 and all future planned maintenance operations 7, 71 during the service life of each device 2).

[0545] - In order to protect the interests of technicians, the scheduling of industrial maintenance (for the maintenance to be performed 6 and all future planned maintenance operations 7, 71 to be performed 90 during the service life of each piece of equipment 2);

[0546] - For the benefit of the operator, all constraints (including total cost) of future maintenance operations during the life of each device 2 (for maintenance 6 to be performed and for all future planned maintenance operations 7, 71 during the life of each device 2);

[0547] - For the benefit of the operator, the scheduling of possible future use of the equipment over the long term of the service life of each equipment 2, and this is about the previously determined limit of use.

[0548] Therefore, system 1 specifically enables:

[0549] -The remaining value of device 2 is constantly assessed and refreshed throughout the lifespan of each device 2;

[0550] - Determine the objective price of each asset constituted by each piece of equipment 2, and thus appropriately inform the decision of whether to retain or resell equipment 2. Consider an example of a group of equipment 2 (such as trucks). System 1 enables the answering of the following questions for the operators of the truck fleet.

[0551] Between two consecutive maintenance operations, for different operating modes (e.g., based on the load to be transported, the mileage to be driven, the average speed, the average gradient of the route used, and the ambient air temperature) associated with different geographical regions (e.g., Russia, Senegal, France), what is the maximum possible utilization rate of several new trucks by the operator?

[0552] - An operator has several trucks that have reached half their service life and intends to scrap them. Considering their corresponding logs 9 and 11, and their region-specific operating conditions worldwide, what is the remaining potential of each truck (in terms of load to be transported, mileage to be traveled, average speed, and service life), and what is the total planned maintenance cost 7 over their remaining service life? What price can the operator demand for each truck?

[0553] The operator owns several trucks currently operating in Senegal, and they are nearing the end of their service life. Considering the standard usage conditions specific to Senegal and those specific to France, and taking into account their respective usage and maintenance logs, is it better to continue operating the trucks in Senegal or to repatriate them to France?

[0554] - For each truck in the operating fleet, considering its logs 9 and 11, the optimal end-of-life date, the maximum possible usage before the end of its life, and the total cost of future maintenance, when does using that truck cease to be profitable?

[0555] This invention enables the updating of optimal lifecycle information for each truck in a fleet at a predetermined periodicity or even in real time to answer questions such as: therefore, with regard to the management of the assets represented by these trucks, this invention allows for optimized decisions to be made at the right time. Thus, system 1 enables the updating of optimal lifecycle information for each piece of equipment 2 in the operator's fleet at any time, or even in real time and at a predetermined periodicity: this allows for optimized and well-planned decisions, thereby benefiting operators and maintenance technicians.

[0556] This invention enables significant improvements in the design, manufacture, maintenance, and use control of a given series of devices 2, and provides long-term visibility into the service life of each device 2. Therefore, for the benefit of designers, manufacturers, maintenance technicians, and operators, this invention achieves the following benefits in terms of control, safety, and confidence in use, as well as economic benefits.

[0557] When these decisions involve the short term (i.e., they involve maintenance 6 to be performed and usage during the planning cycle 70), the present invention unlocks access to control over optimal operation and maintenance decisions for a given device 2.

[0558] On the one hand, the present invention enables the determination of the optimal maintenance decision to be performed for maintenance 6: the decision is tailored to the complete log of device 2 (manufacturing and maintenance log 9 and usage log 11), suitable for the expected use of device 2 after the maintenance (scenario 8), to allow zero failures until the next scheduled maintenance 7, while pushing back the expected zero failure usage limit within the planned cycle 70 as much as possible.

[0559] On the other hand, the present invention eliminates failure uncertainty in a controlled manner. During operation, the present invention effectively provides the operator with real-time updated information on the limits of use that conform to zero failure until the next maintenance: the operator then knows the limits to which they can push the use of device 2 without compromising the expected zero failure, or the limits to which they must limit use to benefit from the expected zero failure. Thus, the present invention unlocks access to the security and use confidence of device 2 and facility 3.

[0560] In fact, by eliminating operational uncertainty in a controlled manner, this invention represents a paradigm shift. Operators no longer need to know whether they will suffer a failure or whether their failure predictions are correct. Now, operators not only determine the failure date (e.g., the day of the next scheduled maintenance 7), but also the pre-failure usage margin (or the margin between the expected usage in scenario 8 and the limit of zero-failure usage): now operators control the use of device 2 to maintain it in line with expected zero failures.

[0561] Therefore, the present invention provides a practical response that meets the stringent requirements of the availability of device 2 (or the security of facility 3 when the availability of device 2 is a prerequisite for the security of facility 3). Thus, the present invention offers short-term economic advantages.

[0562] During the planned cycle 70, in the maintenance 6 steps to be performed and during the operation of equipment 2, the present invention eliminates the uncertainty about failures during the operation of equipment 2, and does so in a controlled manner (by accessing the best maintenance decisions and then by refreshing the knowledge that conforms to the limit of zero failures).

[0563] On the one hand, the present invention thus achieves the full production time of facility 3 as expected by the operator. On the other hand, the present invention thus achieves excellent reliability guarantees in terms of maintenance and operation: the present invention thus makes it possible to further optimize insurance contracts for maintenance technicians and operators. The present invention also unlocks access to controls involving long-term operation and maintenance decisions.

[0564] In fact, with regard to the long-term use of device 2, the present invention enables the timely and constant determination of the potential of device 2, and the prediction of the possible use of device 2 over the long term of its remaining service life (by means of updated information on the extreme use of device 2 and its optimal service life).

[0565] Regarding the long-term maintenance of equipment 2, the present invention also enables timely prediction, both throughout the long service life of equipment 2 and for every possible use of the equipment, of the industrial tasks for various future maintenance operations, the total cost of future maintenance, and the required and sufficient inventory of spare parts (and thus reduces responsiveness constraints on the supply chain due to unforeseen spare parts needs).

[0566] Furthermore, since the present invention enables the maximum remaining potential as the total cost of planned maintenance 7 remaining during the life of the equipment 2 to be constantly assessed and refreshed throughout the life of the equipment 2, the present invention also enables the determination of the remaining value of the equipment throughout the life of the equipment 2, and the notification of decisions to keep or resell the equipment in a relevant manner.

[0567] By refreshing the optimal lifespan information of device 2 or a group of devices at anticipated periodicity or even in real time, optimal decisions can be made at the right time regarding the management of the assets represented by these devices. This invention also enables better control over the design and manufacturing choices of the series of devices 2 by optimizing the designer's / manufacturer's market strategy and by achieving excellent reliability guarantees in design and manufacturing. This increased control brings economic benefits due to its ability to optimize the designer's / manufacturer's insurance contracts. In one aspect, this invention enables the optimization of the designer's / manufacturer's market strategy (market positioning).

[0568] For usage scenarios associated with the market segment targeted by Device 2, this invention effectively enables the characterization of the optimal lifespan of Device 2 by generating relevant metrics (margin between target and limit usage, optimal service life, optimal manufacturing cost, and maintenance schedule within its service life). These metrics enable the assessment of the suitability of Device 2 for the segment under discussion. Therefore, this invention allows designers / manufacturers to:

[0569] - Benefits in market strategy: Designers / manufacturers are thus able to prioritize the target of a part, in which device 2 optimizes the applicability of that part;

[0570] Benefits in market positioning: Designers / manufacturers are thus better able to assert the suitability of device 2 for the needs of customers and competitors, and for supporting model metrics. On the other hand, the invention also provides excellent reliability guarantees in design and manufacturing. The invention effectively enables the determination of optimal manufacturing parameters suitable for a given use case (e.g., suitable for a given market segment intersecting with a given geographic area). The invention also enables the determination of possible methods for optimizing the design for a given use case. Therefore, the invention enables further optimization of insurance contracts for designers / manufacturers. Thus, the invention, especially its model 16, is disruptive in consideration of conventional machine learning projects by systematizing functional assessment and data cross-processing, and by improving considerations for functional assessment when processing data. In this context, the invention is particularly disruptive relative to the prior art, especially for predictive maintenance techniques, because it is more relevant, more powerful, and more widely applicable than simple fault prediction.

Claims

1. A digital monitoring system (1) executed by at least one computing terminal for monitoring the operation and maintenance of at least one industrial device (2) within a monitoring facility (3), comprising at least the following steps: - Install at least one industrial device (2) produced by the manufacturing process (4) and representing a series within the facility (3), and then operate the device (2) for at least the period (5) up to the maintenance step (6) to be performed. - After at least one scenario (8) where the device (2) is scheduled to be used within the planned operating cycle (70), at least one planned maintenance (7) is specified after the maintenance (6) to be performed. Its features The manufacture (4), installation, operation and maintenance of the equipment (2) include at least: - Manufacturing and maintenance log (9), the manufacturing and maintenance log includes: The task (90) is used to manufacture the at least one device (2) up to the installation. The device (2) has at least one previous maintenance (10) task (90); - The usage log (11) of the device (2) during the cycle (5) between the installation and the maintenance (6) to be performed, the usage log (11) including the usage conditions (110) of the device (2) during the cycle (5). - A log (12) of the status (13) of the device (2), the log (12) of the status (13) including the material specifications (120) of the device (2); in Through technical analysis of the device (2), at least one correlation (14) is determined between at least one of the tasks (90) and / or at least one of the usage conditions (110) and at least one of the material indicators (120) of the state (13), the correlation (14) establishing at least one link between the cause and effect of aging of the device (2). In the aforementioned correlation (14): - The task (90) is characterized by those tasks identified as critical, and / or the usage conditions (110) are characterized by those conditions to which the device (2) is sensitive and exposed during operation or when stopped, and the task (90) and the usage conditions (110) discussed affect the state (13) of the device (2). - The material state (13) of the device (2) is characterized by the index (120) identified as representing the state (13) of the device (2). Then, in the correlation (14), the following is determined: -A physical quantity or a function of the measured physical quantity that characterizes the device (2) to which it is sensitive and exposed during operation or when it is stopped, based on the conditions of use (110); -A measured physical quantity or a function of the measured physical quantity that characterizes the material state (13) of the device (2) at a given time; Then For other devices in the series, data associated with these tasks (90) and data associated with the physical quantities or functions of the physical quantities relating to these usage conditions (110) and these material parameters (120), the data being identified and determined in the correlation (14), are recovered and extracted in order to obtain a dataset (15). Train at least one virtual model (16) based on the dataset (15); And among them During the maintenance (6) to be performed on the device (2), the following values ​​are submitted to the model (16): -At least one of the tasks (90) in the manufacturing and maintenance log (9) and at least one of the usage conditions (110) in the usage log (11), - and at least one of the tasks (90) of the maintenance (6) to be performed and at least one of the planned usage conditions of the scenario (8); The model (16) generates a planned state (130) of the device (2) after the maintenance (6) to be performed, and the planned state (130) is compared with the minimum state (17) required for the operation of the device (2).

2. The monitoring system (1) according to claim 1, characterized in that... - Make at least one change to at least one of the values ​​of the task (90) of the maintenance (6) to be performed: - When the value is submitted to the model (16), the changed value is introduced; - In all the changes, at least one sufficient decision is made for the planned state (130) of the device (2) to perform the maintenance (6), the planned state being greater than or equal to the lowest state (17) when the planned maintenance (7) is performed.

3. The monitoring system (1) according to claim 2, characterized in that... - For a given maintenance decision, modify the value of at least one of the planned usage conditions in scenario (8); - When the value is submitted to the model (16), the modified value and the value of the maintenance decision are introduced; - During the planned maintenance (7), the limit is calculated for at least one of the planned usage conditions for the planned state (130) of the device (2), the planned state being equal to the minimum state (17).

4. The monitoring system (1) according to claim 3, characterized in that - When the value is submitted to the model (16), the selected value is used to make sufficient maintenance decisions; - During the planned maintenance (7), for the planned state (130) of the equipment (2) equal to the minimum state (17), the maximum limit (18) of the planned use conditions is calculated for the sufficient maintenance decision.

5. The monitoring system (1) according to claim 4, characterized in that... - For at least one of the planned usage conditions for the scenario (8), a usage margin is determined as the deviation (182) between the corresponding value and the corresponding maximum limit (18).

6. The monitoring system (1) according to claim 5, characterized in that... - Select the best decision from the sufficient decisions, which has an acceptable margin of use or at least one of the following deviations (182).

7. The monitoring system (1) according to claim 6, characterized in that... The regulatory system includes - A graphical representation in the form of a chart, wherein at least one curve is associated with at least one of the planned use conditions, which is a function of the second planned use condition among the planned use conditions. The chart determines the maximum limit (18) of the conditions for use of the first plan.

8. The monitoring system (1) according to any one of claims 1 to 7, characterized in that... - In the correlation (14), the manufacturing and maintenance log (9) is simplified to a log of at least one critical task (90) in the form of a list of consecutive values, each value in the list representing the task (90) discussed in a given maintenance operation. In each list, only the persistent value is selected as the value used in the last maintenance operation during the execution of the task (90) discussed in discussion. - Only the persistent value is retained in the log (9) of the critical task (90).

9. The monitoring system (1) according to any one of claims 1 to 7, characterized in that... In the correlation (14), the function of the measured physical quantity of the usage condition (110) includes - Calculation of the duration of existence of the measured physical quantity within at least one value range; and / or - Calculation of at least one fluctuation in the measured physical quantity: and / or - The count of at least one fluctuation.

10. The monitoring system (1) according to any one of claims 1 to 7, characterized in that... In the correlation (14), periodically - Repeatedly restore new data from at least one manufacturer, maintenance technician, and / or operator. -Then extract the new data to obtain the complete dataset (15). -Then the training of the model (16) is updated based on the complete dataset (15).

11. The monitoring system (1) according to any one of claims 1 to 7, characterized in that... After the maintenance (6) has been performed, the following values ​​are submitted to the model (16). - At least one of the tasks (90) of the manufacturing and maintenance log (9), and at least one task (90) of the maintenance (6) performed. and at least one of the usage conditions (110) in the usage log (11) since the maintenance (6) was performed. - and the planned usage conditions of the scenario (8), The model (16) refreshes the planned state (130) of the device (2) and compares it with the minimum state (17) identified as required for the operation of the device (2).

12. The monitoring system (1) according to any one of claims 5 to 7, characterized in that... Perform at least the following steps: - Assume that at least the sufficient decision for the maintenance (6) to be performed has been performed, and assume that the scenario (8) has been performed up to the planned maintenance (7) after the maintenance (6) to be performed. -Then, at least one change is made to at least one of the values ​​of the task (90) of the planned maintenance (7); - When the value is submitted to the model (16), the value of the change is introduced and the value of the next scenario (81) is foreseen for the planned operation cycle (710) after the planned maintenance (7); - In all changes, during maintenance (71) after the planned maintenance (7), for the planned state (130) of the device (2) that is greater than or equal to the minimum state (17), at least one sufficient decision is selected for the planned maintenance (7); - Determine at least one of the maximum limits (18) and the sufficient maintenance decision made in connection with such selection and the usage margin associated with the next scenario (81); Then, after the planned maintenance during the lifespan of the device (2), the steps are repeated in a cyclical manner every other time.

13. The monitoring system (1) according to any one of claims 2 to 7, characterized in that... - When maintenance decisions are insufficient and the planned state (130) is less than the minimum state (17), the failure date (19) is determined for the corresponding maintenance decision.

14. The monitoring system (1) according to claim 13, characterized in that... - For the maintenance (6) to be performed or for any planned maintenance (7, 71) that is not identified as sufficient, and for at least one given insufficient decision for the maintenance (6) to be performed, the end of life date of the device (2) is determined as the failure date (19) associated with the maintenance decision.

15. The monitoring system (1) according to claim 14, characterized in that... - For the maintenance (6) to be performed or for the planned maintenance identified as the last maintenance during the lifespan of the equipment (2), Among the possible maintenance decisions, choose any combination of the constraints between optimizing the failure date (19), the last use margin, and the last maintenance task (90).

16. The monitoring system (1) according to claim 1, characterized in that... For at least two virtual devices of the same series as the device (2) associated with a separate manufacturing decision. - The simulation is performed over the assumed lifespan of the two virtual devices by submitting at least two manufacturing decisions and at least one planned use scenario (8) to the model (16); - The model (16) generates at least one planning state (130) for each of the two virtual devices. - Select one of the at least two manufacturing decisions based on the planning status (130) of the two virtual devices; - The chosen manufacturing decisions are accessible to the designer / manufacturer.

17. The monitoring system (1) according to claim 16, characterized in that... - For at least two virtual devices of the same series of the device (2) associated with two separate manufacturing decisions; - Repeated simulations are performed in a similar manner for each of the at least two manufacturing decisions to determine the optimal lifespan associated with each of the manufacturing decisions: a series of optimal maintenance decisions, a series of maximum limits (18) and usage margins associated with these optimal maintenance decisions, and the associated optimal lifespan of the virtual device; - Select the optimal manufacturing decision based on the simulation results; - The selected optimal manufacturing decisions are accessible to the designer / manufacturer.

18. The monitoring system (1) according to any one of claims 1 to 7, characterized in that... -Applied to a group of several devices belonging to a single operator in the series (2); - For each of the devices in the device (2), the combined result; The results are at least accessible to the operator.

19. The monitoring system (1) according to any one of claims 1 to 7, characterized in that... - The training of the model (16) belongs to the field of artificial intelligence and can be machine learning.

Images