Task completion evaluation method and system of a distributed dynamic task system

Abstract

This invention relates to a method and system for evaluating task completion in a distributed dynamic task system, belonging to the field of system health prediction technology. This invention can predict the failure rate of each device in the system over a future period based on the historical failure rate of each device, and then calculate the health of each device over that period. Based on the calculated health of each device and its weight, the system health is calculated, and the system delivery time affected by the health of each device is calculated. Finally, the system health and delivery time are combined to evaluate task completion. Therefore, this invention can accurately predict the health of each device and, in conjunction with task requirements, evaluate the impact of each device's health on task completion, accurately assessing task completion. It possesses high precision, dynamic adaptability, and versatility, and can be widely applied in logistics systems, intelligent manufacturing, and other fields.

Description

Technical Field

[0001] This invention relates to a method and system for evaluating task completion in a distributed dynamic task system, belonging to the field of system health prediction technology. Background Technology

[0002] Loosely coupled, dynamic, and complex systems (such as logistics systems and production lines) consist of multiple subsystems (such as elevators, transport vehicles, and cranes), with loose coupling between them and dynamic changes in their operating states over time. Traditional health prediction methods are mostly designed for single devices or static systems, making them ill-suited for complex scenarios involving multiple devices working together and experiencing dynamic changes. Existing methods (such as rule-based threshold judgments or simple statistical models) have the following shortcomings: insufficient dynamism, failing to capture the time-series characteristics of equipment failures; coarse health assessments, often using discrete classifications (such as healthy and unhealthy), lacking continuous quantification; and weak task correlation, failing to assess the actual impact of health in conjunction with system task requirements. To address this, some have proposed applying deep learning (such as LSTM) to time-series prediction. While this approach can effectively predict failure rates and health, it focuses on predicting the health or failure rate of a single device, lacking an overall evaluation of the distributed, dynamic task system. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for evaluating the task completion of a distributed dynamic task system, in order to solve the problem that current evaluations of distributed dynamic task systems are based on individual devices and lack an assessment of the overall task completion status of the distributed dynamic task system.

[0004] To address the aforementioned technical problems, this invention provides a method for evaluating task completion in a distributed dynamic task system. The method includes:

[0005] Based on the historical fault data of each device in the distributed dynamic task system, the failure rate of each device in the future period is predicted, and the health of each device in the future period is calculated based on the predicted failure rate.

[0006] The system's health is calculated based on the health of each device over a future period. When the system's health exceeds a set threshold, the system's delivery time is determined based on the health of each device. If the system's health exceeds the set threshold and the delivery time of the system under the influence of the health of each device meets the demand time for the future period, then the task can be completed.

[0007] Furthermore, the system's shipping time is the sum of the system's pickup time, transportation time, and delivery time. The pickup time is calculated based on the pickup time of various types of goods and the health status of the pickup equipment; the transportation time is calculated based on the transportation time and health status of the transportation equipment; and the delivery time is calculated based on the delivery time and health status of the delivery equipment.

[0008] Furthermore, the formula for calculating the pickup time is as follows:

[0009] ;

[0010] in For pickup equipment in the future health For the pickup time of the i-th type of goods, Let k be the number of goods of type i, and k be the number of pickup devices.

[0011] Furthermore, the formula used to calculate the transportation time is as follows:

[0012] ;

[0013] in For transportation equipment in the future health Let p be the transportation time of the transportation equipment, p be the total shipment, and l be the number of transportation equipment.

[0014] Furthermore, the delivery time is calculated using the following formula:

[0015] ;

[0016] in For delivery of equipment in the future health The delivery time for the delivered equipment.

[0017] Furthermore, the failure rate of each device in the future is predicted using an LSTM model. The LSTM model predicts the failure rate of the corresponding device in the future by capturing the dynamic changes in the historical failure rate sequence of each device.

[0018] Furthermore, the equipment health status is calculated based on the predicted equipment power rate, using the following formula:

[0019] ;

[0020] in For device d in the future time health For device d in the future time Failure rate, Maximum failure rate threshold, where x represents the length of a future period.

[0021] Furthermore, the system's health status over a future period is a weighted sum of the health statuses of all devices over the same period, calculated using the following formula:

[0022] ;

[0023] in To assess the system's health over a future period of time, For device d in the future time health The weight of device d is related to the criticality of the device, where D is the device domain and x represents the length of a future time period.

[0024] Furthermore, the method also includes segmented assessment of device health based on the predicted health of each device.

[0025] The present invention also provides a task completion evaluation system for a distributed dynamic task system, comprising a processor, wherein the processor is used to execute a computer program stored in a memory to implement the task completion evaluation method for the distributed dynamic task system of the present invention, the method comprising:

[0026] Based on the historical fault data of each device in the distributed dynamic task system, the failure rate of each device in the future period is predicted, and the health of each device in the future period is calculated based on the predicted failure rate.

[0027] The system's health is calculated based on the health of each device over a future period. When the system's health exceeds a set threshold, the system's delivery time is determined based on the health of each device. If the system's health exceeds the set threshold and the delivery time of the system under the influence of the health of each device meets the demand time for the future period, then the task can be completed.

[0028] Furthermore, the system's shipping time is the sum of the system's pickup time, transportation time, and delivery time. The pickup time is calculated based on the pickup time of various types of goods and the health status of the pickup equipment; the transportation time is calculated based on the transportation time and health status of the transportation equipment; and the delivery time is calculated based on the delivery time and health status of the delivery equipment.

[0029] Furthermore, the formula for calculating the pickup time is as follows:

[0030] ;

[0031] in For pickup equipment in the future health For the pickup time of the i-th type of goods, Let k be the number of goods of type i, and k be the number of pickup devices.

[0032] Furthermore, the formula used to calculate the transportation time is as follows:

[0033] ;

[0034] in For transportation equipment in the future health Let p be the transportation time of the transportation equipment, p be the total shipment, and l be the number of transportation equipment.

[0035] Furthermore, the delivery time is calculated using the following formula:

[0036] ;

[0037] in For delivery of equipment in the future health The delivery time for the delivered equipment.

[0038] Furthermore, the failure rate of each device in the future is predicted using an LSTM model. The LSTM model predicts the failure rate of the corresponding device in the future by capturing the dynamic changes in the historical failure rate sequence of each device.

[0039] Furthermore, the equipment health status is calculated based on the predicted equipment power rate, using the following formula:

[0040] ;

[0041] in For device d in the future time health For device d in the future time Failure rate, Maximum failure rate threshold, where x represents the length of a future period.

[0042] Furthermore, the system's health status over a future period is a weighted sum of the health statuses of all devices over the same period, calculated using the following formula:

[0043] ;

[0044] in To assess the system's health over a future period of time, For device d in the future time health The weight of device d is related to the criticality of the device, where D is the device domain and x represents the length of a future time period.

[0045] Furthermore, the method also includes segmented assessment of device health based on the predicted health of each device.

[0046] The beneficial effects of this invention are as follows: This invention can predict the failure rate of each device in the system for a future period based on the historical failure rate of each device, and then calculate the health status of each device for that period. Based on the calculated health status and device weights, the system health status is calculated, and the system delivery time affected by the health status of each device is calculated. Finally, the system health status and system delivery time are combined to evaluate task completion. Therefore, this invention can accurately predict the health status of each device, and also evaluate the impact of each device's health status on task completion based on task requirements, thus accurately assessing task completion. Attached Figure Description

[0047] Figure 1 This is a flowchart of the task completion evaluation method of the distributed dynamic task system of the present invention;

[0048] Figure 2 This is a schematic diagram of the segmented evaluation method used in the task completion evaluation method of the distributed dynamic task system of the present invention. Detailed Implementation

[0049] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0050] Implementation of Task Completion Evaluation Methods in Distributed Dynamic Task Systems

[0051] This invention first predicts the failure rate of each device over a future period based on historical fault data of each device in a distributed dynamic task system, and then calculates the health of each device over a future period based on the predicted failure rate. Next, it calculates the overall health of the system over a future period based on the health of each device. When the system's health exceeds a set threshold, the system's delivery time is determined based on the influence of each device's health. If the system's health exceeds the set threshold and the system's delivery time under the influence of each device's health meets the required time for the future period, then the task is considered complete. The implementation flow of this method is as follows: Figure 1 As shown below, a detailed explanation will follow.

[0052] 1. Predict the failure rate of each device over a future period of time.

[0053] This invention utilizes the acquisition module in a distributed dynamic task system to obtain historical failure rate data of each device in the system, and uses a prediction model and the acquired failure rate data of each device to predict the failure rate for a future period. The prediction model can employ a deep learning model; this embodiment uses an LSTM (Long Short-Term Memory) model. LSTM models can capture long-term dependencies in time-series data. Time-series data typically has temporal order and contains components such as trends, seasonality, periodicity, and noise. LSTM models selectively remember or forget past information through their internal gating mechanisms (forget gate, input gate, output gate) and cell states, thereby predicting future values.

[0054] In this embodiment, fault data of each device in the system is collected, including the fault rate f. d (t) (where d represents the device and t represents time), the collected failure rate data is standardized to obtain normalized data f. d' (t). The obtained normalized data f d' (t) is input into the LSTM model, and the LSTM model will output the failure rate of device d over a future period of time. In this embodiment, the future period of time refers to the next x hours. The failure rate for the next x hours output by the LSTM model is:

[0055]

[0056] When using the LSTM model for prediction, it is done on a device-by-device basis. Each device is predicted using the LSTM model to obtain the failure rate of each device in the next x hours.

[0057] As other implementation methods, feedforward neural networks (FNNs) and recurrent neural networks (RNNs) can also be used as prediction networks. Feedforward neural networks (FNNs) are suitable for static prediction with the current system state as input and are suitable for processing fixed feature sets. Recurrent neural networks (RNNs) are similar to long short-term memory networks (LSTMs) and can capture changes in system state over time (such as sequence data) to handle time dependencies.

[0058] 2. Calculate the health status of each device over a future period of time.

[0059] The health score of each device is calculated using the calculated failure rate. The health score is calculated by mapping the predicted failure rate to the health score. For device d, its corresponding health score is calculated based on the predicted failure rate, using the following formula:

[0060]

[0061] in For device d in the future time health For device d in the future time Failure rate, Maximum failure rate threshold. Each specific device has a corresponding maximum failure rate threshold, and the types and numbers of these thresholds vary.

[0062] 3. Calculate the system's health status based on the health status of each device.

[0063] The system's health is determined by the health of each device within the system. Therefore, the system's health over a future period is determined by the health of each device over that period. Considering the varying importance (i.e., criticality) of each device, a weight needs to be assigned to each type of device based on its importance within the system. The higher the importance, the higher the weight assigned. The system's health over a future period is thus determined by the health of each device over that period and the device weights. The formula for calculating system health is as follows:

[0064]

[0065] in To assess the system's health over a future period of time, For device d in the future time health The weight of device d is related to the criticality of the device; the more critical the device, the higher the weight. D is the device domain.

[0066] 4. Assess task completion based on shipping time and system health.

[0067] This invention evaluates task completion based on the system's health and the shipping time affected by the health. If the system's health is greater than the health threshold and the system's shipping time affected by the health meets the task's time requirement, it means that the task can be completed in the future; otherwise, it means that the task cannot be completed.

[0068] Therefore, this implementation method can first determine whether the health of the system is greater than the health threshold. If it is greater than the threshold, the system delivery time under the influence of health is calculated. Otherwise, there is no need to calculate the delivery time, and it can be directly stated that the task cannot be completed in the future.

[0069] The distributed dynamic task system in this embodiment is a logistics system, which includes pickup equipment, transportation equipment, and delivery equipment. Based on the previous steps, the health status of the pickup equipment, transportation equipment, and delivery equipment can be calculated for a future period of time. The formula used to calculate the delivery time affected by the health status of each piece of equipment is as follows:

[0070]

[0071] in This indicates the system's performance over time under the influence of health status. Shipping time, Indicates the system in time Pick-up time, Indicates the system in time The delivery time Indicates the system in time The delivery time and pickup time are calculated based on the pickup time and health status of the pickup equipment for each type of goods; the transportation time is calculated based on the transportation time and health status of the transportation equipment; and the delivery time is calculated based on the delivery time and health status of the delivery equipment.

[0072] The distributed dynamic task system in this invention refers to a system with the characteristics of a logistics system. Therefore, the completion status of the task can be evaluated by "shipping time". Alternatively, it can be evaluated by "quantity of goods delivered".

[0073] The formula for calculating pickup time is as follows:

[0074]

[0075] in For pickup equipment in the future health For the pickup time of the i-th type of goods, Let k be the number of goods of type i, and k be the number of pickup devices.

[0076] The formula used to calculate transportation time is:

[0077]

[0078] in For transportation equipment in the future health Let p be the transportation time of the transportation equipment, p be the total shipment, and l be the number of transportation equipment.

[0079] The delivery time is calculated using the following formula:

[0080]

[0081] in For delivery of equipment in the future health p represents the delivery time of the delivered equipment, and p represents the total quantity shipped.

[0082] The average shipping time for a future period is calculated based on the shipping times at various future points in time, influenced by equipment health. This implementation uses hours as the time node. Using the above method, the shipping time for the next 1 hour, 2 hours, ..., and x hours can be calculated. Averaging these x hours of shipping time yields the average shipping time for the future period. The specific calculation formula is as follows:

[0083]

[0084] The calculated average delivery time over a future period is compared with the preset demand time. If a comparison is made, and This indicates that the preset task can be completed. This is the minimum health threshold, which can be set according to actual needs; in this implementation, it can be set to 70. Gradient Boosting Tree (XGBoost) can be used to optimize the health prediction results.

[0085] Gradient Boosting Tree (XGBoost) is used to correct the LSTM prediction results, improving the accuracy of health prediction. The first stage is LSTM model training (trend prediction); the second stage is XGBoost model training (residual correction). XGBoost is responsible for learning the patterns that LSTM fails to capture (such as the nonlinear impact of sudden operating conditions on health, systematic biases, etc.).

[0086] To more intuitively illustrate the health status of each device, the method also includes segmented health assessments of the devices based on their predicted health status. This invention provides the health status values ​​for each device. Divide the system into intervals, with each interval representing a fault level and a health value. The smaller the value, the higher the fault level. Here, "segment" refers to the health status range, a system indicator that measures whether the system is functioning normally.

[0087] This implementation method uses the device's health value. Divided into 4 ranges: no risk, minor anomaly, potential risk, and serious failure, such as Figure 2 As shown, "no risk" refers to the health score. A score of 90-100 can be represented by green; slight abnormalities indicate a health level. A score of 70-89 can be represented in yellow; potential risk refers to the health score. A score of 50-69 can be represented by orange; a severe malfunction refers to a health score of [value missing]. The range is 0-49, and can be represented in red. As an alternative implementation, the number of levels and the color used for representation can be adaptively adjusted.

[0088] Through the above process, this invention can predict the failure rate of each device in the system for a future period based on the historical failure rate of each device, and then calculate the health of each device for that period. Based on the calculated health of each device and its weight, the system health is calculated, and the system delivery time affected by the health of each device is calculated. Finally, the system health and delivery time are combined to evaluate task completion. Therefore, this invention can accurately predict the health of each device and evaluate its impact on task completion based on task requirements, accurately assessing task completion. It possesses high precision, dynamic adaptability, and versatility, and can be widely applied in logistics systems, intelligent manufacturing, and other fields.

[0089] The method of the present invention will be described below with reference to a specific example:

[0090] Assume the logistics system in this example includes: n=4 elevators (used to carry transport vehicles, belonging to transportation equipment), each connecting m=4 warehouses, each warehouse having k=2 stacker trucks (belonging to picking equipment), and j=25 transport vehicles (belonging to delivery equipment); the number of goods types is l=5, the total number of goods is p=650, and the picking time is t. i = [1, 1.5, 2, 2.5, 3] minutes, elevator transport time t e =2 minutes, delivery time t v =5 minutes.

[0091] Failure rates: Elevator 0.05 times / hour, stacker truck 0.03 times / hour, transport vehicle 0.04 times / hour. These are the current values ​​in the historical failure rates. The failure rates of each device in the next 72 hours can be predicted based on the historical failure rates of each device in the past 24 hours.

[0092] 1) Input the historical failure rate of each device over 24 hours into the LSTM model and predict the failure rate of each device over the next 72 hours.

[0093] 2) Calculate the health status of each device based on its failure rate over the next 72 hours:

[0094]

[0095] Assuming the weights of each device are 0.5 for the elevator, 0.3 for the stacker truck, and 0.2 for the transport vehicle, calculate the health of the system.

[0096] In this example, the predicted system health (e.g., 85.2) is in the yellow zone, indicating that the system is operational but requires attention.

[0097] The calculated shipping time is approximately 650 minutes, which meets the task requirements. Task evaluation: Task completed.

[0098] Implementation of a Task Completion Evaluation System for Distributed Dynamic Task Systems

[0099] The task completion evaluation system of the distributed dynamic task system includes a memory, a processor, an internal bus, and a computer program stored in the memory. The processor and memory communicate and exchange data via the internal bus. The processor executes the computer program to implement the steps of the task completion evaluation method for the distributed dynamic task system of this invention. The processor can be a microprocessor (MCU), a programmable logic device (FPGA), or other processing devices; the memory can be various types of memory that store information using electrical energy, such as RAM or ROM, or other types of memory. The implemented task completion evaluation method for the distributed dynamic task system is as follows: Figure 1 As shown, the specific process includes the following:

[0100] 1. Predict the failure rate of each device over a future period of time.

[0101] This invention utilizes the acquisition module in a distributed dynamic task system to obtain historical failure rate data of each device in the system, and uses a prediction model and the acquired failure rate data of each device to predict the failure rate for a future period. The prediction model can employ a deep learning model; this embodiment uses an LSTM (Long Short-Term Memory) model. LSTM models can capture long-term dependencies in time-series data. Time-series data typically has temporal order and contains components such as trends, seasonality, periodicity, and noise. LSTM models selectively remember or forget past information through their internal gating mechanisms (forget gate, input gate, output gate) and cell states, thereby predicting future values.

[0102] In this embodiment, fault data of each device in the system is collected, including the fault rate f. d (t) (where d represents the device and t represents time), the collected failure rate data is standardized to obtain normalized data f. d' (t). The obtained normalized data f d' (t) is input into the LSTM model, and the LSTM model will output the failure rate of device d over a future period of time. In this embodiment, the future period of time refers to the next x hours. The failure rate for the next x hours output by the LSTM model is:

[0103]

[0104] When using the LSTM model for prediction, it is done on a device-by-device basis. Each device is predicted using the LSTM model to obtain the failure rate of each device in the next x hours.

[0105] 2. Calculate the health status of each device over a future period of time.

[0106] The health score of each device is calculated using the calculated failure rate. The health score is calculated by mapping the predicted failure rate to the health score. For device d, its corresponding health score is calculated based on the predicted failure rate, using the following formula:

[0107]

[0108] in For device d in the future time health For device d in the future time Failure rate, Maximum failure rate threshold.

[0109] 3. Calculate the system's health status based on the health status of each device.

[0110] The system's health is determined by the health of each device within the system. Therefore, the system's health over a future period is determined by the health of each device over that period. Considering the varying importance (i.e., criticality) of each device, a weight needs to be assigned to each type of device based on its importance within the system. The higher the importance, the higher the weight assigned. The system's health over a future period is thus determined by the health of each device over that period and the device weights. The formula for calculating system health is as follows:

[0111]

[0112] in To assess the system's health over a future period of time, For device d in the future time health The weight of device d is related to the criticality of the device; the more critical the device, the higher the weight. D is the device domain.

[0113] 4. Assess task completion based on shipping time and system health.

[0114] This invention evaluates task completion based on the system's health and the shipping time affected by the health. If the system's health is greater than the health threshold and the system's shipping time affected by the health meets the task's time requirement, it means that the task can be completed in the future; otherwise, it means that the task cannot be completed.

[0115] Therefore, this implementation method can first determine whether the health of the system is greater than the health threshold. If it is greater than the threshold, the system delivery time under the influence of health is calculated. Otherwise, there is no need to calculate the delivery time, and it can be directly stated that the task cannot be completed in the future.

[0116] The distributed dynamic task system in this embodiment is a logistics system, which includes pickup equipment, transportation equipment, and delivery equipment. Based on the previous steps, the health status of the pickup equipment, transportation equipment, and delivery equipment can be calculated for a future period of time. The formula used to calculate the delivery time affected by the health status of each piece of equipment is as follows:

[0117]

[0118] in This indicates the system's performance over time under the influence of health status. Shipping time, Indicates the system in time Pick-up time, Indicates the system in time The delivery time Indicates the system in time The delivery time and pickup time are calculated based on the pickup time and health status of the pickup equipment for each type of goods; the transportation time is calculated based on the transportation time and health status of the transportation equipment; and the delivery time is calculated based on the delivery time and health status of the delivery equipment.

[0119] The formula for calculating pickup time is as follows:

[0120]

[0121] in For pickup equipment in the future health For the pickup time of the i-th type of goods, Let k be the number of goods of type i, and k be the number of pickup devices.

[0122] The formula used to calculate transportation time is:

[0123]

[0124] in For transportation equipment in the future health Let p be the transportation time of the transportation equipment, p be the total shipment, and l be the number of transportation equipment.

[0125] The delivery time is calculated using the following formula:

[0126]

[0127] in For delivery of equipment in the future health p represents the delivery time of the delivered equipment, and p represents the total quantity shipped.

[0128] The average shipping time for a future period is calculated based on the shipping times at various future points in time, influenced by equipment health. This implementation uses hours as the time node. Using the above method, the shipping time for the next 1 hour, 2 hours, ..., and x hours can be calculated. Averaging these x hours of shipping time yields the average shipping time for the future period. The specific calculation formula is as follows:

[0129]

[0130] The calculated average delivery time over a future period is compared with the preset demand time. If a comparison is made, and This indicates that the preset task can be completed. This is the minimum health threshold, which can be set according to actual needs; in this implementation, it can be set to 70. Gradient Boosting Tree (XGBoost) can be used to optimize the health prediction results.

[0131] To more intuitively illustrate the health status of each device, the method also includes segmented health assessments of the devices based on their predicted health status. This invention provides the health status values ​​for each device. Divide the system into intervals, with each interval representing a fault level and a health value. The smaller the value, the higher the fault level.

Claims

1. A method for evaluating task completion in a distributed dynamic task system, characterized in that, The method includes: Based on the historical fault data of each device in the distributed dynamic task system, the failure rate of each device in the future period is predicted, and the health of each device in the future period is calculated based on the predicted failure rate. The system's health is calculated based on the health of each device over a future period. When the system's health exceeds a set threshold, the system's delivery time is determined based on the health of each device. If the system's health exceeds the set threshold and the delivery time of the system under the influence of the health of each device meets the demand time for the future period, then the task can be completed.

2. The task completion evaluation method for a distributed dynamic task system according to claim 1, characterized in that, The system's shipping time is the sum of the system's pickup time, transportation time, and delivery time. The pickup time is calculated based on the pickup time of various types of goods and the health status of the pickup equipment. The transportation time is calculated based on the transportation time and the health status of the transportation equipment. Delivery time is calculated based on the delivery time of the delivered equipment and the health status of the delivered equipment.

3. The task completion evaluation method for a distributed dynamic task system according to claim 2, characterized in that, The formula for calculating pickup time is as follows: ； in For pickup equipment in the future health For the pickup time of the i-th type of goods, Let k be the number of goods of type i, and k be the number of pickup devices.

4. The task completion evaluation method for a distributed dynamic task system according to claim 2, characterized in that, The formula used to calculate transportation time is: ； in For transportation equipment in the future health Let p be the transportation time of the transportation equipment, p be the total shipment, and l be the number of transportation equipment.

5. The task completion evaluation method for a distributed dynamic task system according to claim 2, characterized in that, delivery The formula used to calculate time is: ； in For delivery of equipment in the future health The delivery time for the delivered equipment.

6. The task completion evaluation method for a distributed dynamic task system according to claim 1 or 2, characterized in that, The failure rate of each device in the future is predicted by the LSTM model. The LSTM model predicts the failure rate of the corresponding device in the future by capturing the dynamic changes of the historical failure rate sequence of each device.

7. The task completion evaluation method for a distributed dynamic task system according to claim 1 or 2, characterized in that, Equipment health is calculated based on the predicted equipment power rate, using the following formula: ； in For device d in the future time health For device d in the future time Failure rate, Maximum failure rate threshold, where x represents the length of a future period.

8. The task completion evaluation method for a distributed dynamic task system according to claim 1 or 2, characterized in that, The system's health status over a future period is a weighted sum of the health statuses of all devices over the same period, calculated using the following formula: ； in To assess the system's health over a future period of time, For device d in the future time health The weight of device d is related to the criticality of the device, where D is the device domain and x represents the length of a future time period.

9. The task completion evaluation method for a distributed dynamic task system according to claim 1 or 2, characterized in that, The method also includes segmented assessment of device health based on the predicted health of each device.

10. A task completion evaluation system for a distributed dynamic task system, comprising a processor, characterized in that, The processor is used to execute a computer program stored in a memory to implement the steps of the task completion evaluation method of any one of claims 1 to 9 for a distributed dynamic task system.

Images