EAM-based asset full life cycle management method and system

Abstract

The invention discloses an EAM-based asset full-life-cycle management method and system, and the method comprises the steps: obtaining key performance indexes of assets in purchasing, operation, maintenance and scrapping stages, carrying out the dynamic smoothing of the data of each stage based on an EAM algorithm, constructing an asset state trend curve, extracting feature change points of each stage according to the asset state trend curve, and carrying out the calculation of the feature change points. And calculating a weight coefficient of each stage based on the feature change points, generating a comprehensive score of the full life cycle of the assets based on the weight coefficients, dividing an optimization priority of the assets according to the score, comparing the optimization priority with a preset threshold, and judging whether an operation strategy of the assets needs to be adjusted or a maintenance stage needs to be entered. Therefore, the management model is trained through the historical data, the asset management strategy is verified according to the comprehensive score and the optimization priority, the model parameters are adjusted when the strategy is abnormal, and the overall efficiency and precision of asset management are improved.

Description

Technical Field

[0001] This invention relates to the technical field of EAM asset management, and in particular to a method and system for EAM asset lifecycle management. Background Technology

[0002] In modern enterprise asset management, full lifecycle management of assets (including procurement, operation, maintenance, and disposal stages) is crucial for improving management efficiency and accuracy. The EAM (Exponential Moving Average) algorithm, with its advantages in dynamic data smoothing and trend prediction, has become an important tool for asset status monitoring and optimized maintenance. However, existing EAM-based asset management methods often focus on specific aspects (such as resource allocation and performance configuration), failing to effectively cover the entire asset lifecycle, thus limiting management efficiency and accuracy.

[0003] Specifically, existing technologies have significant limitations. For example, patent CN111026677B proposes a method for dynamically adjusting the cache based on storage volume IOPS using EAM to optimize storage performance. However, the application of this solution is strictly limited to the storage resource management stage and cannot be extended to other key stages of the asset lifecycle (such as procurement evaluation, in-depth operation and maintenance, and obsolescence decisions). In addition, the effectiveness of its strategy is highly dependent on the real-time nature of IOPS data. When data is delayed or the collection frequency is low, caching decisions will lag, thus affecting the overall effect.

[0004] On the other hand, patent CN105513648B generates the optimal performance configuration for chip startup through EAM training, achieving adaptive optimization. However, the application scope of this solution is mainly limited to the field of chip performance configuration and is difficult to transfer to other types of asset management scenarios. At the same time, its strong dependence on the EAM algorithm makes the configuration results susceptible to bias or noise interference from the initial training data, thus affecting the actual operating performance of the chip.

[0005] In summary, current EAM-based asset lifecycle management methods mainly suffer from the following problems: First, the application is limited to specific stages (such as cache management and chip configuration), lacking a global perspective. Second, it fails to cover the entire process of asset procurement, operation, maintenance, and disposal, resulting in incomplete coverage. Finally, the algorithm's performance is sensitive to data quality (real-time performance and accuracy), exhibiting insufficient robustness. Therefore, there is an urgent need to develop a method that covers the entire asset lifecycle and optimizes the application strategies of EAM algorithms at different stages to achieve dynamic management and intelligent decision-making throughout the entire process, thereby comprehensively improving the overall efficiency of modern enterprise asset management. Summary of the Invention

[0006] The present invention aims to at least partially solve one of the technical problems in the related art.

[0007] Therefore, the purpose of this invention is to propose an asset lifecycle management method and system based on EAM (Enterprise Asset Management). This method trains the management model using historical data, verifies the asset management strategy based on comprehensive scoring and optimization priorities, and adjusts the model parameters when the strategy is abnormal, thereby improving the overall efficiency and accuracy of asset management.

[0008] To achieve the above objectives, this invention proposes an asset lifecycle management method and system based on EAM (Enterprise Asset Management). On one hand, it provides an EAM-based asset lifecycle management method, including: S1, acquiring key performance indicators of assets at each stage of procurement, operation, maintenance, and disposal, and dynamically smoothing the data at each stage based on the EAM algorithm to construct an asset status trend curve; S2, extracting characteristic change points at each stage based on the asset status trend curve, and calculating the weight coefficients for each stage based on these characteristic change points, wherein the formula for calculating the weight coefficients is: S3. Generate a comprehensive score for the entire asset lifecycle based on weighted coefficients, and prioritize assets according to the score; S4. Compare the optimization priorities with preset thresholds to determine whether the asset's operating strategy needs adjustment or if it needs to enter the maintenance phase; S5. If it is determined that the operating strategy needs adjustment, repeat steps S1 to S4 until the closed-loop management of the entire asset lifecycle is completed; S6. Determine that the training of the asset lifecycle management model is complete, and use the trained model parameters to dynamically monitor and make intelligent decisions for new assets, where the model parameters include time decay parameters. and normalization factor The optimal value.

[0009] On the other hand, it provides an asset lifecycle management system. By dynamically smoothing the data at each stage of the asset and combining the weight allocation of feature change points, it realizes global management of the entire asset lifecycle. Before actual application, the management model is trained with historical data, the asset management strategy is verified based on comprehensive scoring and optimization priority, and the model parameters are adjusted when the strategy is abnormal, thereby improving the overall efficiency and accuracy of asset management.

[0010] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0011] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating the asset lifecycle management method based on EAM in an embodiment of the present invention. Figure 2This is a flowchart of the asset lifecycle management method based on EAM in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the process of constructing the asset status trend curve in an embodiment of the present invention; Figure 4 This is a structured diagram illustrating the formula for calculating the weight coefficients of feature change points in an embodiment of the present invention. Figure 5 This is a block diagram of the main modules of the asset lifecycle management system in this embodiment of the invention; Figure 6 This is a schematic diagram illustrating the interaction between model parameter training and dynamic monitoring in an embodiment of the present invention. Detailed Implementation

[0012] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention. Rather, embodiments of the invention include all variations, modifications, and equivalents falling within the appended spirit and scope.

[0013] The following description, in conjunction with the accompanying drawings, illustrates the EAM-based asset lifecycle management method and system of this invention.

[0014] like Figure 1-6 As shown in the embodiment of the present invention, the asset lifecycle management method and system based on EAM (Enterprise Asset Management) acquires key performance indicators (KPIs) and non-key performance indicators (NKPIs) of assets at each stage of procurement, operation, maintenance, and disposal through various means such as sensors, monitoring equipment, historical databases, and real-time performance data in step S1. Key performance indicators refer to indicators that have a decisive impact on the operating status, and may include, but are not limited to, asset operating efficiency, failure rate, energy consumption level, and maintenance costs. NKPIs refer to secondary indicators, such as the degree of wear and tear. To ensure the accuracy and reliability of the data, after acquiring the raw data, the EAM algorithm is used to dynamically smooth the data at each stage. The core idea of ​​the EAM algorithm is to assign different weights to time-series data, making recent data have a greater impact on the results, thereby more sensitively reflecting the changing trends of asset status.

[0015] Specifically, the formula for the EAM algorithm is: ,in, This represents the smoothed value at the current moment. This represents the actual value at the current moment. It is a smoothing factor and satisfies 0 < <1, This represents the smoothed value from the previous moment. The smoothed value calculated using the above formula can effectively reduce noise interference, thereby more accurately constructing the asset status trend curve.

[0016] In step S2, characteristic change points for each stage are extracted based on the asset status trend curve generated in step S1. Characteristic change points refer to time points where the asset status undergoes significant changes, typically manifested as abrupt changes in the curve slope or numerical jumps. To identify these characteristic change points, the system uses differential analysis to calculate the trend curve point by point. When the difference value at a point exceeds a preset threshold, that point is marked as a characteristic change point. Subsequently, the weighting coefficients for each stage are calculated based on the extracted characteristic change points.

[0017] The formula for calculating the weighting coefficient is: The formula for calculating the weighting coefficient is: The weighting coefficients calculated using the above formula can quantify the impact of each feature change point on the overall asset status, thus providing a basis for the generation of subsequent comprehensive scores.

[0018] In step S3, a comprehensive score for the entire asset lifecycle is generated based on the weighting coefficients calculated in step S2. The formula for calculating the comprehensive score is: ,in, This indicates the overall score. This represents the weight coefficient for the i-th feature change point. This represents the key performance indicator value associated with the i-th feature change point. The comprehensive score calculated using the above formula can fully reflect the asset's performance level throughout its entire lifecycle, and the optimization priority of the asset can be determined based on the score.

[0019] The priority weights of each dynamically changing feature are calculated based on historical transaction data and real-time performance data. The formula for calculating the priority weights is as follows: ,in, This represents the priority weight of the i-th dynamically changing feature. This indicates the magnitude of performance variation of this feature in historical transaction data. This indicates the frequency of fluctuation of this feature in real-time performance data. This indicates the time span of the feature. , and These are the weighting coefficients for performance changes, fluctuation frequency, and time span, respectively. The significance of this formula lies in comprehensively considering the performance of dynamic change characteristics across different dimensions, thereby determining their importance in asset management.

[0020] The specific optimization priority classification rules are as follows: when the comprehensive score is lower than a preset first threshold, the asset is classified as high optimization priority; when the comprehensive score is between the first and second thresholds, the asset is classified as medium optimization priority; and when the comprehensive score is higher than the second threshold, the asset is classified as low optimization priority. Through this classification method, the system can quickly identify assets requiring focused attention and formulate corresponding management strategies.

[0021] In step S4, the optimization priorities generated in step S3 are compared with preset thresholds to determine whether the asset's operating strategy needs to be adjusted or whether it needs to enter the maintenance phase. For example, for assets classified as high optimization priority, the system will further analyze the changing trends of their key performance indicators. If a certain indicator is found to be continuously deteriorating, an alarm will be issued immediately and maintenance will be recommended. For assets with optimization priority, the system will periodically monitor their status changes and provide optimization suggestions when necessary. For assets with low optimization priority, the existing operating strategy will be maintained and routine monitoring will be performed. In this way, the system can automatically adjust the management strategy according to the status of different assets, thereby improving the overall efficiency and accuracy of asset management.

[0022] In step S5, if it is determined that the asset's operating strategy needs to be adjusted, steps S1 to S4 are repeated until closed-loop management of the asset's entire lifecycle is completed. For example, after a maintenance is completed, the system will re-collect the asset's key performance indicators and update the asset status trend curve, then recalculate the weighting coefficients and comprehensive score, and finally determine the new optimization priority. Through this iterative process, the system can track changes in asset status in real time and dynamically adjust management strategies to ensure that asset management is always in an optimal state.

[0023] In step S6, it is determined that the training of the asset lifecycle management model is complete, and the trained model parameters are used to dynamically monitor and make intelligent decisions for new assets. The model parameters include time decay parameters. and normalization factor The optimal values ​​for these parameters are obtained through training on historical data. Specifically, the system selects a certain amount of historical asset data as the training set and continuously adjusts it. and The value of minimizes the model's prediction error, thus obtaining the optimal parameter combination. After training, the system will use the optimal parameters to dynamically monitor new assets, collect key performance indicators in real time, and generate a comprehensive score, thereby achieving intelligent asset management.

[0024] Furthermore, this invention also provides an asset lifecycle management system based on the above-described method. For example... Figure 5The diagram shows the main module structure of the asset lifecycle management system in this embodiment of the invention. The system includes a data acquisition module, a data processing module, a feature extraction module, a weight calculation module, a comprehensive scoring module, an optimization decision-making module, and a model training module. The data acquisition module is responsible for acquiring key performance indicators of assets from multiple sources; the data processing module performs dynamic smoothing of the data using the EAM algorithm; the feature extraction module identifies feature change points in the asset status trend curve; the weight calculation module calculates the weight coefficients for each stage based on the feature change points; the comprehensive scoring module generates a comprehensive score for the entire asset lifecycle; the optimization decision-making module prioritizes optimizations and formulates management strategies based on the comprehensive score; and the model training module trains model parameters using historical data to improve the system's prediction accuracy. Through the collaborative work of these modules, the system can achieve global perspective management of the entire asset lifecycle, significantly improving the overall efficiency and accuracy of asset management.

[0025] The optimization decision-making module includes a comparative inspection and evaluation unit, a verification and evaluation unit, and a priority setting and management unit. The comparative inspection and evaluation unit and the verification and evaluation unit calculate the asset status consistency parameter. The asset status consistency parameter characterizes the conformity between the generated inspection and evaluation values ​​and the actual operating data; it is defined as the average deviation between the two. When the asset status consistency parameter is less than or equal to the first preset consistency parameter, it indicates a significant overall deviation between the inspection and evaluation values ​​and the actual operating data, at which point the operation of the asset lifecycle simulation process is deemed abnormal. Further, the optimization module determines the deviation correction factor for each dynamic change characteristic and calculates the comprehensive deviation evaluation parameter based on these factors. The comprehensive deviation evaluation parameter characterizes the error of each dynamic change characteristic. The formula for calculating the comprehensive deviation evaluation parameter is: ,in, This represents the comprehensive deviation evaluation parameter. This represents the priority weight of the i-th dynamically changing feature. This indicates the deviation value of the feature. When the comprehensive deviation evaluation parameter is less than or equal to the preset deviation evaluation parameter, although there are certain errors in each dynamic change feature, the overall error is small, and the asset life cycle simulation process is considered to be running successfully. Conversely, if the comprehensive deviation evaluation parameter is greater than the preset deviation evaluation parameter, it indicates that the errors of some dynamic change features are too large. In this case, it is necessary to adjust the priority weight formula to improve the prediction accuracy.

[0026] In summary, the EAM-based asset lifecycle management method and system of this invention achieves global-view management of the entire asset lifecycle by introducing the EAM algorithm to dynamically smooth data at each stage of the asset lifecycle and combining it with weight allocation of feature change points. Before practical application, the system trains the management model using historical data, verifies the asset management strategy based on comprehensive scoring and optimization priorities, and adjusts model parameters when the strategy is abnormal, thereby improving the overall efficiency and accuracy of asset management.

[0027] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method for full lifecycle asset management based on EAM, characterized in that, Includes the following steps: S1. Obtain key performance indicators of assets at each stage of procurement, operation, maintenance and disposal, and dynamically smooth the data at each stage based on the EAM algorithm to construct asset status trend curves; S2. Extract the characteristic change points for each stage based on the asset status trend curve, and calculate the weight coefficient for each stage based on the characteristic change points. The formula for calculating the weight coefficient is as follows: ; S3. Generate a comprehensive score for the entire life cycle of an asset based on weighting coefficients, and prioritize the optimization of assets according to the score; The formula for calculating priority weight is: ,in, This represents the priority weight of the i-th dynamically changing feature. This indicates the magnitude of performance variation of this feature in historical transaction data. This indicates the frequency of fluctuation of this feature in real-time performance data. This indicates the time span of the feature. , and These are the weighting coefficients for performance change, fluctuation frequency, and time span, respectively. S4. Compare the optimization priority with the preset threshold to determine whether it is necessary to adjust the asset's operating strategy or enter the maintenance phase. S5. If it is determined that the operation strategy needs to be adjusted, repeat steps S1 to S4 until the closed-loop management of the entire asset life cycle is completed. S6. Determine that the training of the asset lifecycle management model is complete, and use the trained model parameters to dynamically monitor and make intelligent decisions for new assets. The model parameters include time decay parameters. and normalization factor The optimal value.

2. The asset lifecycle management method based on EAM according to claim 1, characterized in that, The formula for the EAM algorithm in step S1 is: ,in, This represents the smoothed value at the current moment. This represents the actual value at the current moment. It is a smoothing factor and satisfies 0 < <1, This represents the smoothed value from the previous time step.

3. The asset lifecycle management method based on EAM according to claim 1, characterized in that, The key performance indicators in step S1 include, but are not limited to, asset operating efficiency, failure rate, energy consumption level, and maintenance cost.

4. The asset lifecycle management method based on EAM according to claim 1, characterized in that, The feature change points in step S2 are identified by differential analysis. When the difference value of a point exceeds a preset threshold, the point is marked as a feature change point.

5. The asset lifecycle management method based on EAM according to claim 1, characterized in that, In step S3, the formula for calculating the comprehensive score is: ,in, This indicates the overall score. This represents the weight coefficient for the i-th feature change point. This represents the key performance indicator value associated with the i-th feature change point.

6. The asset lifecycle management method based on EAM according to claim 1, characterized in that, In step S4, the optimization priority is divided according to the following rules: when the comprehensive score is lower than the first preset threshold, the asset is classified as high optimization priority; when the comprehensive score is between the first preset threshold and the second preset threshold, the asset is classified as medium optimization priority. When the overall score is higher than the second preset threshold, the asset is classified as having a low optimization priority.

7. The asset lifecycle management method based on EAM according to claim 1, characterized in that, In step S6, the optimal values ​​of the model parameters are obtained through training with historical data. The training process includes selecting several historical asset data sets as the training set and continuously adjusting them. and The value of minimizes the model's prediction error.

8. The asset lifecycle management system based on the EAM asset lifecycle management method according to any one of claims 1 to 7, characterized in that, It includes a data acquisition module, a data processing module, a feature extraction module, a weight calculation module, a comprehensive scoring module, an optimization decision-making module, and a model training module. The data acquisition module is used to obtain key performance indicators of assets from multiple sources; The data processing module is used to dynamically smooth the data using the EAM algorithm; The feature extraction module is used to identify characteristic change points in the asset status trend curve; The weight calculation module is used to calculate the weight coefficients for each stage based on the feature change points; The comprehensive scoring module is used to generate a comprehensive score for the entire lifecycle of an asset; The optimization decision-making module is used to prioritize optimizations and formulate management strategies based on comprehensive scores. The model training module is used to train model parameters using historical data to improve the system's prediction accuracy.

9. The asset lifecycle management system according to claim 8, characterized in that, The model training module obtains the time decay parameter. and normalization factor The optimal value is determined, and the trained model parameters are used to dynamically monitor and make intelligent decisions for new assets.

Images