Progress evaluation method and system based on work order task

Abstract

The invention discloses a progress evaluation method and system based on work order tasks, and the method comprises the steps: obtaining each work order task of a complete work order, and obtaining the state of the work order task; evaluating the current progress of the work order task, and predicting other task schedules which are not executed; and synthesizing the current task progress obtained by evaluation and the other task schedules to generate prediction of completion of the work order. The accuracy and reliability of work order management are effectively improved, a user is helped to quickly judge the credibility of a prediction result in a complex task environment, the decision process is optimized, the management risk caused by prediction errors is reduced, and the work order execution efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of task progress evaluation, and in particular to a method and system for progress evaluation based on work order tasks. Background Art

[0002] With the development of digitalization and intelligence, enterprises and organizations are increasingly relying on work order systems to manage and schedule tasks. However, as the complexity of work order tasks increases, how to accurately evaluate the progress of work orders, predict the time required for unfinished tasks, and evaluate the reliability of the prediction results has become one of the key challenges in improving the efficiency of work order management.

[0003] Traditional work order progress assessment methods usually rely on simple linear predictions or empirical rules, lacking comprehensive consideration of multiple factors (such as task complexity, historical data reference, equipment failure rate, etc.). This method cannot accurately reflect the actual progress of work order tasks, especially in complex and changing task environments, which can easily lead to deviations in prediction results.

[0004] Existing work order progress forecasting methods usually ignore the evaluation of the reliability of the forecast results. The forecast results are only provided to users as reference data, and it is difficult for users to judge the credibility of these forecast results, which may lead to wrong decisions.

[0005] When evaluating the progress of work order tasks, many systems are unable to update prediction results in real time or dynamically adjust prediction strategies. This can easily lead to delayed evaluation results when the task status changes frequently, thereby affecting the overall management efficiency of the work order. Summary of the invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problem solved by the present invention is that the existing task progress evaluation method has the problems of lack of reliability analysis, lack of dynamics and real-time performance, etc.

[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: a progress evaluation method based on work order tasks, comprising:

[0009] Get each work order task of the complete work order, and get the status of the work order task;

[0010] Evaluate the current progress of the work order task and predict the schedule of other tasks that have not yet been executed;

[0011] The evaluated current task progress and the other task schedules are combined to generate a prediction for the completion of the work order.

[0012] As a preferred solution of the progress evaluation method based on work order tasks described in the present invention, the work order tasks include: outputting the content of the work order through the front end, identifying all the work order tasks in the content, and obtaining the work order tasks of the operation instructions and the work order tasks of the process approval;

[0013] When the work order task is process approval, upload the content of the operation instruction to be executed to the approver; when the work order task is an operation instruction, perform the operation according to the content of the operation instruction;

[0014] The status of the work order task in the review process includes: among all work order tasks, the status of process approval includes: approved, unapproved, and waiting for approval; the status of operation instructions includes: pending execution, pending approval, and executed;

[0015] Among them, waiting for approval means that after completing the previous process, it reaches the work order task for process approval, and at the same time, the next operation instruction generates a status of pending approval.

[0016] As a preferred solution of the progress evaluation method based on work order tasks described in the present invention, the state of the work order tasks in the review process also includes arranging each work order task in sequence according to the complete work order to obtain a sequence R = {R q1 , R q2 , ..., R qn};

[0017] Among them, q represents the type of work order task, process approval is 0, and operation instruction is 1; n represents the number of work order tasks; R qn Indicates the nth work order task with work order task type q;

[0018] Assume R qi and R qj , are two adjacent work order tasks, and i<j; when work order i is a process approval, if it is approved, it will enter the next work order task; if it is not approved, the process will be terminated; when work order i is an operation instruction, the operation completion progress of work order i is updated in real time. If the completion progress of work order i reaches the preset progress, it will enter the next work order task of the task process;

[0019] Split the sequence R according to the type of work order task to obtain the process approval sequence SP = {SP1, SP2, ..., SP u} and the sequence of operation instructions CZ={CZ1,CZ2,...,CZ v};

[0020] Among them, SP u Indicates the work order task for the u-th process approval, CZ vrepresents the work order task of the vth operation instruction, u represents the number of work order tasks for process approval, and v represents the number of work order tasks for the operation instruction.

[0021] As a preferred solution of the progress evaluation method based on work order tasks described in the present invention, the current progress of the work order tasks is evaluated, including, if the current work order task is an operation instruction, updating the completed part and its time consumption, estimating the time consumption of the unfinished part, and generating the progress evaluation result of the current task;

[0022] The time consumption for the unfinished part is estimated as follows:

[0023] Update the proportion P0 of the completed part of the current task in the task progress, and record the time h0 of the completed part of the current work order task;

[0024] Obtain the execution record of the current operation instruction in the historical record, and calculate the average time h1 of the task progress in the record when it is P0, and the average time h2 of the remaining task progress (1-P0);

[0025] Evaluate the time taken by the current work order task for the unfinished part:

[0026]

[0027] The progress evaluation result of the current task is: H+h0, and the progress evaluation result of the previous task is updated in real time according to the synchronous update ratio P0;

[0028] If the current work order task is process approval, obtain the position number of the approver waiting for approval, so as to evaluate the approval time of the current work order task:

[0029] The input is the job number N currently waiting for approval. The mean square error is used as the loss function to predict the time required for the approval of the current work order task through a multi-layer perceptron.

[0030] T pred =W3·ReLU(W2·ReLU(W1·N+b1)+b2)+b3

[0031] Where N represents the number of input tasks; W1, W2, W3 represent the weight matrices of different layers; b1, b2, b3 represent bias terms; T pred It takes time for the forecast to be approved.

[0032] As a preferred solution of the work order task-based progress evaluation method of the present invention, the predicted other tasks that have not yet been executed include: in the process approval sequence SP = {SP1, SP2, ..., SP u} and the sequence of operation instructions CZ={CZ1,CZ2,...,CZ v}, filter the completed tasks and count the time TY; filter the current tasks and count the time H+h0; filter the remaining elements in the two sequences as other tasks that have not been executed;

[0033] Make predictions for other tasks that have not been executed in SP and other tasks that have not been executed in CZ respectively;

[0034] Assume that there are e elements in the sequence of operation instructions that are other tasks that have not been executed, then the task sequence of the generated operation instructions CZ'={CZ' u-v-1 , CZ' u-v -2, ..., CZ' v};

[0035] Calculate the estimated time for each element in the sequence CZ'. For the a-th element CZ' a , count the maximum and average time consumption of the same operation instruction in the historical records, and then calculate the difference δ;

[0036] The ath element CZ' a The time loss prediction is expressed as:

[0037]

[0038] in, Indicates the same as CZ' in the history a The average time consumption of the same operation instruction; θ represents the equipment failure rate under the current operation instruction; γ represents the historical records with CZ' a The average failure rate of the same operation instruction; ∈ represents random disturbance, the value is a random number between 0 and δ; HCZ' a Indicates the ath element CZ' a The time loss prediction results;

[0039] The estimated time consumption of each element in the sequence CZ' is accumulated to obtain the total time consumption prediction result HCZ of the operation instructions in other tasks that have not been executed.

[0040] As a preferred solution of the progress evaluation method based on work order tasks described in the present invention, the prediction of scheduling of other tasks that have not been executed also includes: assuming that there are d elements in the sequence of process approval that are other tasks that have not been executed, then the task sequence SP'={SP' u-d-1 , SP' u-d -2, ..., SP' u};

[0041] Use a multi-layer perceptron to predict the approval time of each element in turn, and calculate the bth element SP' b When the time loss is , obtain the post number N' waiting for approval, input N' into the multi-layer perceptron to obtain the bth element SP' b After the time loss prediction result is completed, the prediction of the time loss of the next element of the sequence SP' is performed;

[0042] The estimated time consumption of each element in the sequence SP' is accumulated to obtain the total time consumption prediction result HSP of the process approval in other tasks that have not been executed.

[0043] As a preferred solution of the progress evaluation method based on work order tasks described in the present invention, wherein: the prediction of the completed work order includes statistically analyzing all the prediction results and the actual time consumption to obtain the total time consumption prediction of the work order task: HZ=HSP+HCZ+TY+H+h0;

[0044] During the task progress, the total time consumption forecast of the work order task is updated in real time according to the actual completion progress; and the reliability of the forecast result is evaluated through the evaluation function to generate a reliability level;

[0045] Calculate the first-order rate of change ΔT of the total time consumed for each prediction k , used to capture immediate changing trends;

[0046] ΔT k =T k -T k-1

[0047] Among them, T k Indicates the total time consumed for the kth prediction, where k represents the number of predictions;

[0048] Calculate the second derivative Δ for each prediction 2 T k , used to measure the acceleration or deceleration trend of the rate of change;

[0049] Δ 2 T k =ΔT k -ΔT k-1

[0050] Calculate the standard deviation of the current rate of change:

[0051]

[0052] in, is the average value of all first-order change rates, and g represents the index of the first-order change rate;

[0053] By weighting volatility to focus on more recent forecasts, current volatility is expressed as:

[0054]

[0055] Among them, V x represents the standard deviation of the rate of change at the xth update, w x Represents the weight of the rate of change at the xth update;

[0056] Current Trending Rating:

[0057]

[0058] Among them, ρ and σ are weights used to adjust the first-order rate of change and the second-order derivative; τ is a parameter that adjusts the impact of volatility on the score; sign represents a return function, which returns 1 when the value is greater than 0; returns 0 when the value is equal to 0; and returns -1 when the value is less than 0.

[0059] Map the ratings to a reliability range:

[0060]

[0061] Among them, μ is a parameter that adjusts the sensitivity of the score;

[0062] By preset threshold and Ω, judge the level, if If The credibility is moderate; if Ω≤R(T k ), then it is judged to be highly reliable;

[0063] When the number of updates meets the calculation requirements, the credibility level of each prediction result is output;

[0064] The latest prediction results including the credibility level are presented on the front end first. Through human-computer interaction, you can choose to view the prediction results of the last two updates, but no credibility level is generated.

[0065] A progress evaluation system based on work order tasks using the method of the present invention is characterized by:

[0066] The task generation unit obtains each work order task of the complete work order and the status of the work order task;

[0067] A prediction unit, which evaluates the current progress of the work order task and predicts the schedule of other tasks that have not yet been executed;

[0068] The output unit combines the evaluated current task progress with the other task schedules to generate a prediction for the completion of the work order.

[0069] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.

[0070] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.

[0071] Beneficial effects of the present invention: The progress evaluation method based on work order tasks provided by the present invention realizes accurate evaluation and dynamic prediction of the progress of work order tasks by comprehensively analyzing the status of work order tasks, historical data, equipment failure rate, and volatility of prediction results. Compared with traditional methods, the present invention captures prediction trends by calculating the first-order rate of change and second-order derivatives while updating the prediction results in real time, and performs a multi-dimensional evaluation of the reliability of the prediction results in combination with volatility analysis, thereby generating a credibility level. This method effectively improves the accuracy and reliability of work order management, helps users quickly judge the credibility of prediction results in complex task environments, optimizes decision-making processes, reduces management risks caused by prediction errors, and improves the efficiency of work order execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0073] Figure 1 An overall flow chart of a progress evaluation method based on work order tasks provided for the first embodiment of the present invention. DETAILED DESCRIPTION

[0074] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0075] Example 1

[0076] Reference Figure 1 , is an embodiment of the present invention, and provides a progress evaluation method based on a work order task, comprising:

[0077] S1: Get each work order task of the complete work order, and get the status of the work order task.

[0078] The work order task includes: outputting the content of the work order through the front end, identifying all the work order tasks in the content, obtaining the work order task of the operation instruction and the work order task of the process approval; when the work order task is the process approval, uploading the content of the operation instruction to be executed to the approval personnel; when the work order task is the operation instruction, performing the operation according to the content of the operation instruction; the status of the work order task in the review process includes: among all work order tasks, the status of the process approval includes: approved, not approved, waiting for approval; the status of the operation instruction includes: pending execution, pending approval, executed.

[0079] It is worth mentioning that the same operation instruction in a work order task may include the operation execution of multiple tasks. For example, after the approval is completed, the three tasks ABC are executed at the same time. After all or part of them are completed, the process of the next work order task is carried out.

[0080] Among them, waiting for approval means that after completing the previous process, it reaches the work order task for process approval, and at the same time, the next operation instruction generates a status of pending approval.

[0081] Furthermore, the status of the work order task in the review process also includes, according to the complete work order, arranging each work order task in order according to the task order, and obtaining a sequence R={R q1 , R q2 , ..., R qn}; q represents the type of work order task, process approval is 0, and operation instruction is 1; n represents the number of work order tasks; R qn Indicates the nth work order task of work order task type q.

[0082] Assume R qi and R qj , are two adjacent work order tasks, and i<j; when work order i is a process approval, if it is approved, it will enter the next work order task; if it is not approved, the process will be terminated; when work order i is an operation instruction, the operation completion progress of work order i is updated in real time. If the completion progress of work order i reaches the preset progress, it will enter the next work order task of the task flow.

[0083] Split the sequence R according to the type of work order task to obtain the process approval sequence SP = {SP1, SP2, ..., SP u} and the sequence of operation instructions CZ={CZ1,CZ2,...,CZ v}.

[0084] Among them, SP u Indicates the work order task for the u-th process approval, CZv represents the work order task of the vth operation instruction, u represents the number of work order tasks for process approval, and v represents the number of work order tasks for the operation instruction.

[0085] S2: Evaluate the current progress of the work order task and predict the schedule of other tasks that have not yet been executed.

[0086] Furthermore, the current progress of the work order task is evaluated, including, if the current work order task is an operation instruction, updating the completed part and its time consumption, estimating the time consumption of the unfinished part, and generating a progress evaluation result of the current task.

[0087] The time consumption for the unfinished part is estimated as follows:

[0088] Update the proportion P0 of the completed part of the current task in the task progress, and record the time h0 of the completed part of the current work order task.

[0089] Get the execution record of the current operation instruction in the historical record, and calculate the average time h1 of the task progress in the record when it is P0, and the average time h2 of the remaining task progress (1-P0).

[0090] Evaluate the time taken by the current work order task for the unfinished part:

[0091]

[0092] The progress evaluation result of the current task is: H+h0, and the progress evaluation result of the previous task is updated in real time according to the synchronous update ratio P0.

[0093] If the current work order task is process approval, obtain the position number of the approver waiting for approval, so as to evaluate the approval time of the current work order task:

[0094] The input is the position number N currently waiting for approval. The mean square error is used as the loss function, and the multi-layer perceptron is used to predict the time required for the approval of the current work order task.

[0095] T pred =W3·ReLU(W2·ReLU(W1·N+b1)+b2)+b3

[0096] Where N represents the number of input tasks; W1, W2, W3 represent the weight matrices of different layers; b1, b2, b3 represent bias terms; T pred It takes time for the forecast to be approved.

[0097] By updating the progress ratio and time consumption of the completed part and using historical data to estimate the time consumption of the unfinished part, the system can generate the progress evaluation results of the operation instruction task in real time. This evaluation method not only takes into account the completed part, but also provides a reasonable time estimate for the unfinished part through historical record data, thereby ensuring the accuracy of the progress evaluation.

[0098] For process approval tasks, the designed multi-layer perceptron model (MLP) predicts the time consumption of the current approval task based on historical data by inputting the job number of the approver waiting for approval. In this way, the system can effectively predict the processing time of complex approval tasks and provide more accurate time estimates for the management of the entire work order process. And by updating the progress and time consumption of completed tasks in real time, the overall progress evaluation results of the work order tasks are dynamically adjusted. This dynamic adjustment ensures that the system can reflect the latest status of the task in a timely manner and provide a basis for the scheduling of unfinished tasks.

[0099] Furthermore, the predicted other tasks that have not been executed include the sequence SP of process approval = {SP1, SP2, ..., SP u} and the sequence of operation instructions CZ={CZ1,CZ2,...,CZ v}, filter the completed tasks and count the time TY; filter the current tasks and count the time H+h0; filter the remaining elements in the two sequences as other tasks that have not been executed. Predict the other tasks that have not been executed in SP and other tasks that have not been executed in CZ respectively.

[0100] Assume that there are e elements in the sequence of operation instructions that are other tasks that have not been executed, then the task sequence of the generated operation instructions CZ'={CZ' u-v-1 , CZ' u-v -2, ..., CZ' v}.

[0101] Calculate the estimated time for each element in the sequence CZ'. For the a-th element CZ' a , count the maximum and average time consumed by the same operation instruction in the historical records, and then calculate the difference δ.

[0102] The ath element CZ' a The time loss prediction is expressed as:

[0103]

[0104] in, Indicates the same as CZ' in the history a The average time consumption of the same operation instruction; θ represents the equipment failure rate under the current operation instruction; γ represents the historical records with CZ' aThe average failure rate of the same operation instruction; ∈ represents random disturbance, the value is a random number between 0 and δ; HCZ' a Indicates the ath element CZ' a The time loss prediction result. Normally, the selection of this random interference factor is designed according to the probability distribution, but for the scheduling of work order tasks, it is necessary to reserve the operable time, so the disturbance factor is randomly selected, so that it can be large enough in probability, thereby reserving more time for fault tolerance. The average of this random number can be achieved by accumulating multiple elements, so that the overall time loss prediction is at a high but not very outrageous level. Then, by continuously updating the task work order, the tasks of the operation instructions are gradually reduced, indicating that the probability of large-scale paralysis of equipment is greatly reduced or cannot affect normal production, and the cumulative time consumption of the number of random interferences becomes very small, while still reserving some fault tolerance time. Such a design can guarantee the reliability estimate in normal work.

[0105] The estimated time consumption of each element in the sequence CZ' is accumulated to obtain the total time consumption prediction result HCZ of the operation instructions in other tasks that have not been executed.

[0106] Assume that there are d elements in the process approval sequence that are other tasks that have not been executed, then the task sequence SP' = {SP' u-d-1 , SP' u-d -2, ..., SP' u}.

[0107] Use a multi-layer perceptron to predict the approval time of each element in turn, and calculate the bth element SP' b When the time loss is , obtain the post number N' waiting for approval, input N' into the multi-layer perceptron to obtain the bth element SP' b After the time loss prediction result is completed, the prediction of the time loss of the next element of the sequence SP' is performed.

[0108] The estimated time consumption of each element in the sequence SP' is accumulated to obtain the total time consumption prediction result HSP of the process approval in other tasks that have not been executed.

[0109] S3: Integrate the evaluated current task progress and the other task schedules to generate a prediction for the completion of the work order.

[0110] All prediction results and actual time consumption are counted to obtain the total time consumption prediction of the work order task: HZ=HSP+HCZ+TY+H+h0.

[0111] During the task progress, the total time forecast of the work order task is updated in real time according to the actual completion progress; and the reliability of the forecast result is evaluated through the evaluation function to generate a reliability level.

[0112] Calculate the first-order rate of change ΔT of the total time consumed for each prediction k , used to capture immediate changing trends.

[0113] ΔT k =T k -T k-1

[0114] Among them, T k It represents the total time of the k-th prediction, where k represents the number of predictions.

[0115] Calculate the second derivative Δ for each prediction 2 T k , which is used to measure the acceleration or deceleration trend of the rate of change.

[0116] Δ 2 T k =ΔT k -ΔT k-1

[0117] Calculate the standard deviation of the current rate of change:

[0118]

[0119] in, is the average of all first-order change rates, and g represents the index of the first-order change rate.

[0120] By weighting volatility to focus on more recent forecasts, current volatility is expressed as:

[0121]

[0122] Among them, V x represents the standard deviation of the rate of change at the xth update, w x Indicates the weight of the rate of change at the xth update. Current trend score:

[0123]

[0124] Among them, ρ and σ are weights used to adjust the first-order rate of change and the second-order derivative; τ is a parameter that adjusts the impact of volatility on the score; sign represents a return function, which returns 1 when the value is greater than 0; returns 0 when the value is equal to 0; and returns -1 when the value is less than 0. Map the score to the reliability range:

[0125]

[0126] Among them, μ is a parameter for adjusting the sensitivity of the score. and Ω, judge the level, if If The credibility is moderate; if Ω≤R(T k ), it is judged to be highly reliable.

[0127] When the number of updates meets the calculation requirements, the credibility level for each prediction result is output. The latest prediction result including the credibility level is presented first on the front end. Through human-computer interaction, you can choose to view the prediction results of the last two updates, but no credibility level is generated.

[0128] It should be noted that during the task process, the total time forecast of the work order task will be affected by the actual progress. By updating the forecast results in real time, the system can adjust the total time forecast value in time according to the latest progress information to ensure that the forecast is highly consistent with the actual situation. By calculating the first-order rate of change of the total time of each forecast, the system can instantly capture the changes in the forecast trend, analyze the progress speed and direction of the task, and identify whether there are abnormal fluctuations or delays. By calculating the second-order derivative, the system further measures the acceleration or deceleration trend of the rate of change and identifies whether the change of the forecast result is stable. This helps to judge the potential risks or bottlenecks in the progress of the task. Introducing volatility analysis, by weighting the standard deviation of the rate of change, the system can focus on the recent forecast fluctuations. This volatility assessment can help identify the instability of the forecast results, thereby providing a basis for reliability assessment. Through comprehensive trend scoring, the system generates a reliability level for each forecast result. Through preset thresholds, the system can divide the forecast results into three levels: low, medium, and high, providing users with clear credibility guidance to help users better judge the reliability of the forecast results.

[0129] It is also important to know that this evaluation function is generated based on the changing trend of the prediction results, reflecting the changes in the prediction results. According to the design of the disturbance factor, the prediction time will gradually shorten under normal circumstances. If this happens, it proves that everything is working well, and the output is high credibility. On the contrary, if the prediction time becomes longer, it means that there is a certain possibility of abnormality or miscalculation or error, so a lower credibility is given at this time. If the credibility continues to be low, it proves that the abnormality persists. At this time, it can be determined that a fault has occurred during the task, and it should be repaired in time.

[0130] On the other hand, this embodiment also provides a progress evaluation system based on work order tasks, which includes:

[0131] The task generation unit obtains each work order task of the complete work order and the status of the work order task.

[0132] The prediction unit evaluates the current progress of the work order task and predicts the schedule of other tasks that have not yet been executed.

[0133] The output unit combines the evaluated current task progress with the other task schedules to generate a prediction for the completion of the work order.

[0134] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0135] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0136] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0137] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0138] Example 2 is an embodiment of the present invention, which provides a progress evaluation method based on work order tasks. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0139] The experimental object is a work order management system with different task types and complexities, mainly including process approval tasks and operation instruction tasks. The progress of each task in the system needs to be evaluated in real time, and the time consumption of the unfinished part is predicted based on historical data and the current status of the task. At the same time, in order to ensure the reliability of the prediction results, the multi-layer perceptron (MLP) model is used in combination with volatility analysis and trend analysis of the first-order rate of change and second-order derivative to evaluate the reliability of each prediction result.

[0140] Experimental objects and parameter settings: The work order tasks of the experimental objects are divided into three main parts: process approval tasks, operation instruction tasks, and mixed task sequences. The progress and time consumption of each task are evaluated based on the data collected during the task execution. The experimental data comes from the historical records of different tasks in the system, covering multiple actual work order task executions.

[0141] Experimental implementation process:

[0142] Work order task data collection: First, we extracted the history records of multiple completed work orders from the work order management system, covering the time consumption, completion progress and other data of different types of tasks. The data sample includes 10 process approval tasks and 10 operation instruction tasks to ensure the diversity and representativeness of the data.

[0143] Task classification and sorting: Classify tasks according to the type of work order task (process approval or operation instruction), and generate task sequences according to the order in which the tasks are executed. Mark the status of each task as "approved", "unapproved", "awaiting approval" or "to be executed", "executed", and include them in different task sequences according to their types.

[0144] Real-time progress evaluation and prediction: Evaluate the progress of the current task, especially the operation instruction task, and update the progress ratio and time consumption of the completed part in real time. At the same time, through the task data with similar progress to the current task in the historical records, calculate the average time consumption of the unfinished part and make a prediction to generate the progress evaluation result of the current task.

[0145] Application of Multi-layer Perceptron Model: For process approval tasks, a multi-layer perceptron (MLP) model is used to predict the time consumption. The input features include the number of the waiting approval post for the current task and the time consumption data of historical approval tasks. The model output is the predicted time consumption of the current task.

[0146] Reliability assessment: By calculating the first-order rate of change, second-order derivative and volatility, the instantaneous change trend and stability of the forecast results are comprehensively analyzed. Based on these analysis results, a reliability score for the current forecast result is generated and mapped to a reliability level (low, medium, high).

[0147] Result recording and analysis: Compare the total time prediction result of each test with the actual time, calculate the error, and record the reliability level of the prediction result. Repeat the experiment many times to ensure the representativeness of the data and the stability of the results.

[0148] Table 1 Data record table

[0149]

[0150]

[0151] By analyzing the data in the table, we can clearly see the effectiveness and advantages of the present invention in the work order task progress assessment and time consumption prediction. First, from the comparison between the "total time consumption prediction" and the "actual total time consumption", it can be seen that the prediction error using the method of the present invention is small, especially in task sequence 2, task sequence 4, and task sequence 6, the difference between the prediction result and the actual time consumption is controlled within 5 minutes, showing extremely high prediction accuracy. This is due to the historical data analysis and multi-layer perceptron model adopted by the present invention, which can make reasonable time consumption estimates based on the specific characteristics and historical performance of the task.

[0152] Secondly, by combining the data analysis of "first-order rate of change" and "second-order derivative", the present invention can capture the changing trend in task progress. For task sequence 1, task sequence 3, and task sequence 5, although the first-order rate of change shows a decreasing trend in the total time consumption of the task, the second-order derivative shows that the change in task progress tends to be flat or slightly increased, which indicates that the stability of the task execution process has improved, and the reliability of the prediction results has been verified.

[0153] The data from the volatility analysis further reinforces the advantages of the present invention. In task sequence 2, task sequence 4, and task sequence 6, the standard deviation of volatility is small (all below 0.5), which indicates that the execution process of these tasks is stable, the volatility of the prediction results is small, and therefore has a high reliability level. This is consistent with the actual reliability level evaluation results, showing the significant superiority of the present invention in complex task environments.

[0154] Compared with the prior art, the present invention constructs a multi-dimensional work order task progress evaluation system by combining a multi-layer perceptron model, historical data analysis, first-order rate of change, second-order derivative and volatility analysis. The prior art often relies on a single prediction model or simple linear regression, which cannot effectively cope with dynamic changes in task execution, nor can it provide real-time reliability evaluation. The present invention not only improves the accuracy of the prediction results, but also provides users with clear reliability level guidance, allowing users to more effectively manage work order tasks, optimize resource allocation and decision-making processes, thereby greatly improving the overall efficiency and stability of the work order management system.

[0155] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A progress evaluation method based on work order tasks, characterized in that: include: Get each work order task of the complete work order, and get the status of the work order task; Evaluate the current progress of the work order task and predict the schedule of other tasks that have not yet been executed; The evaluated current task progress and the other task schedules are combined to generate a prediction for the completion of the work order.

2. The progress evaluation method based on work order tasks according to claim 1, characterized in that: The work order task includes outputting the content of the work order through the front end, identifying all the work order tasks in the content, and obtaining the work order tasks of the operation instructions and the work order tasks of the process approval; When the work order task is process approval, upload the content of the operation instruction to be executed to the approver; when the work order task is an operation instruction, perform the operation according to the content of the operation instruction; The status of the work order task in the review process includes: among all work order tasks, the status of process approval includes: approved, unapproved, and waiting for approval; the status of operation instructions includes: pending execution, pending approval, and executed; Among them, waiting for approval means that after completing the previous process, it reaches the work order task for process approval, and at the same time, the next operation instruction generates a status of pending approval.

3. The progress evaluation method based on work order tasks as claimed in claim 2, characterized in that: The state of the work order task in the review process also includes arranging each work order task in order according to the complete work order to obtain a sequence R={R q1 , R q2 , ..., R qn }; Among them, q represents the type of work order task, process approval is 0, and operation instruction is 1; n represents the number of work order tasks; R qn Indicates the nth work order task with work order task type q; Assume R qi and R qj , are two adjacent work order tasks, and i<j; when work order i is a process approval, if it is approved, it will enter the next work order task; if it is not approved, the process will be terminated; when work order i is an operation instruction, the operation completion progress of work order i is updated in real time. If the completion progress of work order i reaches the preset progress, it will enter the next work order task of the task process; Split the sequence R according to the type of work order task to obtain the process approval sequence SP = {SP1, SP2, ..., SP u } and the sequence of operation instructions CZ={CZ1,CZ2,...,CZ v }; Among them, SP u Indicates the work order task for the u-th process approval, CZ v represents the work order task of the vth operation instruction, u represents the number of work order tasks for process approval, and v represents the number of work order tasks for the operation instruction.

4. The progress evaluation method based on work order tasks as claimed in claim 3, characterized in that: The current progress of the work order task is evaluated, including, if the current work order task is an operation instruction, updating the completed part and its time consumption, estimating the time consumption of the unfinished part, and generating a progress evaluation result of the current task; The time consumption for the unfinished part is estimated as follows: Update the proportion P0 of the completed part of the current task in the task progress, and record the time h0 of the completed part of the current work order task; Obtain the execution record of the current operation instruction in the historical record, and calculate the average time h1 of the task progress in the record when it is P0, and the average time h2 of the remaining task progress (1-P0); Evaluate the time taken by the current work order task for the unfinished part: The progress evaluation result of the current task is: H+h0, and the progress evaluation result of the previous task is updated in real time according to the synchronous update ratio P0; If the current work order task is process approval, obtain the position number of the approver waiting for approval, so as to evaluate the approval time of the current work order task: The input is the job number N currently waiting for approval. The mean square error is used as the loss function to predict the time required for the approval of the current work order task through a multi-layer perceptron. <h2 style=";text-align:left;direction:ltr">T<h2 style=";text-align:left;direction:ltr"> pred <h2 style=";text-align:left;direction:ltr"> =W3·ReLU(W2·ReLU(W1·N+b1)+b2)+b3 Where N represents the number of input tasks; W1, W2, W3 represent the weight matrices of different layers; b1, b2, b3 represent bias terms; T pred It takes time for the forecast to be approved.

5. The progress evaluation method based on work order tasks as claimed in claim 4, characterized in that: The predicted other tasks that have not been executed include: u } and the sequence of operation instructions CZ={CZ1,CZ2,...,CZ v }, filter the completed tasks and count the time TY; filter the current tasks and count the time H+h0; filter the remaining elements in the two sequences as other tasks that have not been executed; Make predictions for other tasks that have not been executed in SP and other tasks that have not been executed in CZ respectively; Assume that there are e elements in the sequence of operation instructions that are other tasks that have not been executed, then the task sequence of the generated operation instructions CZ'={CZ' u-v-1 , CZ' u-v -2, ..., CZ' v }; Calculate the estimated time for each element in the sequence CZ'. For the a-th element CZ' a , count the maximum and average time consumption of the same operation instruction in the historical records, and then calculate the difference δ; The ath element CZ' a The time loss prediction is expressed as: in, Indicates the same as CZ' in the history a The average time consumption of the same operation instruction; θ represents the equipment failure rate under the current operation instruction; γ represents the historical records with CZ' a The average failure rate of the same operation instruction; ∈ represents random disturbance, the value is a random number between 0 and δ; HCZ' a Indicates the ath element CZ' a The time loss prediction results; The estimated time consumption of each element in the sequence CZ' is accumulated to obtain the total time consumption prediction result HCZ of the operation instructions in other tasks that have not been executed.

6. The progress evaluation method based on work order tasks as claimed in claim 5, characterized in that: The prediction of the scheduling of other tasks that have not been executed also includes: assuming that there are d elements in the process approval sequence that are other tasks that have not been executed, then the process approval task sequence SP'={SP' u-d-1 , SP' u-d-2 , ..., SP' u }; Use a multi-layer perceptron to predict the approval time of each element in turn, and calculate the bth element SP' b When the time loss is , obtain the post number N' waiting for approval, input N' into the multi-layer perceptron to obtain the bth element SP' b After the time loss prediction result is completed, the prediction of the time loss of the next element of the sequence SP' is performed; The estimated time consumption of each element in the sequence SP' is accumulated to obtain the total time consumption prediction result HSP of the process approval in other tasks that have not been executed.

7. The progress evaluation method based on work order tasks according to claim 6, characterized in that: The prediction of completing the work order includes counting all the prediction results and the actual time consumption to obtain the total time consumption prediction of the work order task: HZ=HSP+HCZ+TY+H+h0; During the task process, the total time forecast of the work order task is updated in real time according to the actual completion progress; And through the evaluation function, the reliability of the prediction results is evaluated to generate a reliability level; Calculate the first-order rate of change ΔT of the total time consumed for each prediction k , used to capture immediate changing trends; Δ T k =T k -T k-1 Among them, T k Indicates the total time consumed for the kth prediction, where k represents the number of predictions; Calculate the second derivative Δ for each prediction 2 T k , used to measure the acceleration or deceleration trend of the rate of change; D 2 T k =ΔT k -ΔT k-1 Calculate the standard deviation of the current rate of change: in, is the average value of all first-order change rates, and g represents the index of the first-order change rate; By weighting volatility to focus on more recent forecasts, current volatility is expressed as: Among them, V x represents the standard deviation of the rate of change at the xth update, w x Represents the weight of the rate of change at the xth update; Current Trending Rating: Among them, ρ and σ are weights used to adjust the first-order rate of change and the second-order derivative; τ is a parameter that adjusts the impact of volatility on the score; sign represents a return function, which returns 1 when the value is greater than 0; returns 0 when the value is equal to 0; and returns -1 when the value is less than 0. Map the ratings to a reliability range: Among them, μ is a parameter that adjusts the sensitivity of the score; By preset threshold and Ω, judge the level, if If The credibility is moderate; if Ω≤R(T k ), then it is judged to be highly reliable; When the number of updates meets the calculation requirements, the credibility level of each prediction result is output; The latest prediction results including the credibility level are presented on the front end first. Through human-computer interaction, you can choose to view the prediction results of the last two updates, but no credibility level is generated.

8. A progress evaluation system based on work order tasks using the method according to any one of claims 1 to 7, characterized in that: The task generation unit obtains each work order task of the complete work order and the status of the work order task; A prediction unit, which evaluates the current progress of the work order task and predicts the schedule of other tasks that have not yet been executed; The output unit combines the evaluated current task progress with the other task schedules to generate a prediction for the completion of the work order.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the progress evaluation method based on the work order task are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a progress evaluation method based on a work order task are implemented.

Images