Support guarantee scheme automatic generation method and device, equipment and medium

Abstract

The application provides a support guarantee scheme automatic generation method, device, equipment and medium, and relates to the technical field of artificial intelligence, and the method comprises the following steps: obtaining a historical data set under a to-be-supported guarantee scene, wherein the historical data set comprises respective task demand contents corresponding to different time points in continuous time; determining a prediction result set according to the historical data set, wherein the prediction result set comprises respective prediction demand change values corresponding to a plurality of task demand contents; and generating an optimal support guarantee scheme according to a plurality of prediction demand change values, in combination with a resource constraint set and a business rule set. The method aims to quickly generate and analyze the optimal support guarantee scheme through an automatic and intelligent method, improve the efficiency and accuracy of the generation and analysis of the optimal support guarantee scheme, provide clear action guidelines for commanders and support guarantee personnel, ensure the feasibility and efficiency of the task, and respond to rapid response and effective decision-making in emergency situations.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of artificial intelligence, and in particular to a support guarantee scheme automatic generation method, device, equipment and medium. BACKGROUND

[0002] In the prior art, the generation and analysis of support guarantee schemes usually rely on human experience and judgment. By consulting historical data, real-time information and relevant literature, combined with personal professional knowledge and intuition, the optimal support guarantee scheme is formulated and evaluated. This method can solve the problem to some extent, but it has obvious shortcomings: first, the speed and accuracy of manual data processing are limited, and it is difficult to respond quickly in emergency situations; second, human factors may lead to subjectivity and inconsistency in decision-making, affecting the feasibility and effectiveness of the optimal support guarantee scheme; finally, the traditional scheme generation method lacks deep mining and intelligent analysis of a large amount of data, making it difficult to find potential risks and opportunities.

[0003] In this way, if an emergency occurs in the support guarantee scenario, the entire process cannot fully meet the needs of rapid response and effective decision-making. SUMMARY

[0004] The present application provides a support guarantee scheme automatic generation method, device, equipment and medium to solve the defect that if an emergency occurs in the support guarantee scenario, the entire process cannot fully meet the needs of rapid response and effective decision-making in the prior art. The purpose is to quickly generate and analyze the optimal support guarantee scheme through automated and intelligent methods, improve the efficiency and accuracy of the generation and analysis of the optimal support guarantee scheme, provide clear action guidelines for commanders and support personnel, ensure the feasibility and efficiency of the task, and respond to rapid response and effective decision-making in emergency situations.

[0005] The present application provides a support guarantee scheme automatic generation method, comprising:

[0006] Obtain a historical data set under a support guarantee scenario, the historical data set comprising respective task demand contents at different times within a continuous time;

[0007] According to the historical data set, determine a prediction result set, the prediction result set comprising a plurality of respective predicted demand change values of task demand contents;

[0008] According to a plurality of predicted demand change values, combined with a resource constraint set and a business rule set, generate an optimal support guarantee scheme.

[0009] The application provides a support guarantee scheme automatic generation method, which comprises the following steps: inputting historical data sets into a long short-term memory network model to obtain an initial prediction result set output by the long short-term memory network model; determining a loss function corresponding to the long short-term memory network model according to the historical data sets and the initial prediction result set; updating model parameters of the long short-term memory network model according to the loss function to obtain a trained long short-term memory network model; and inputting the historical data sets into the trained long short-term memory network model to obtain the prediction result set output by the trained long short-term memory network model.

[0010] The application provides a support guarantee scheme automatic generation method, which comprises the following steps: obtaining a task set and an initial time table under a to-be-supported scenario, wherein tasks in the task set correspond to start execution times in the initial time table one by one; determining a target task combination according to the task set, wherein tasks in the target task combination belong to the task set; generating a target time table according to the task set, the initial time table and a resource constraint set; and generating an optimal support guarantee scheme according to a plurality of predicted demand change values, the target task combination, the target time table and a business rule set.

[0011] The application provides a support guarantee scheme automatic generation method, which comprises the following steps: determining a target task combination according to a task set, wherein the target task combination is obtained by rearranging tasks in the task set; determining respective weights and estimated costs of all tasks in each task combination; determining a task execution cost corresponding to the task combination according to all weights and all estimated costs; and determining the target task combination corresponding to the minimum task execution cost in a plurality of task execution costs.

[0012] The application provides a support guarantee scheme automatic generation method, which comprises the following steps: determining an initial cost of each task in a task set when the task is executed at a corresponding start execution time; determining respective use constraints of all resource constraints in a resource constraint set according to the initial time table and the resource constraint set; and determining a target time table with a minimum total cost according to all initial costs and all use constraints.

[0013] According to the support guarantee scheme automatic generation method provided by the application, the method further comprises: for each task in the task set, performing the following operations: in the process of multiple simulations of the task, determining the occurrence probability and the influence degree of the target risk on the task at each simulation; according to the resource constraint set, the occurrence probability and the influence degree, determining the task execution result corresponding to the task under the target risk; according to the distribution of multiple task execution results, determining the influence situation of the target risk on the task execution.

[0014] According to the support guarantee scheme automatic generation method provided by the application, the method further comprises: saving the current version corresponding to the optimal support guarantee scheme; if it is detected that the optimal support guarantee scheme receives new data, and / or the variation degree of the key parameters corresponding to the optimal support guarantee scheme is greater than a preset threshold, updating the current version.

[0015] The application further provides a support guarantee scheme automatic generation device, comprising:

[0016] The acquisition module is used for acquiring a historical data set under a to-be-supported guarantee scene, wherein the historical data set comprises task demand contents corresponding to different time points in continuous time;

[0017] The processing module is used for determining a prediction result set according to the historical data set, wherein the prediction result set comprises prediction demand change values corresponding to multiple task demand contents respectively; and generating an optimal support guarantee scheme according to multiple prediction demand change values, in combination with a resource constraint set and a business rule set.

[0018] The application further provides an electronic device comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor implements the support guarantee scheme automatic generation method according to any one of the above when executing the program.

[0019] The application further provides a non-transient computer readable storage medium, wherein the non-transient computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the support guarantee scheme automatic generation method according to any one of the above.

[0020] The application further provides a computer program product comprising a computer program, wherein the computer program is executed by a processor to implement the support guarantee scheme automatic generation method according to any one of the above.

[0021] The application provides a support guarantee scheme automatic generation method, device, equipment and medium, which comprises the following steps: obtaining a historical data set in a to-be-supported guarantee scene, wherein the historical data set comprises task demand contents corresponding to different time points in continuous time; determining a prediction result set according to the historical data set, wherein the prediction result set comprises prediction demand change values corresponding to a plurality of task demand contents; and generating an optimal support guarantee scheme according to a plurality of prediction demand change values, in combination with a resource constraint set and a business rule set. The method aims to quickly generate and analyze the optimal support guarantee scheme by means of automatic and intelligent methods, improve the efficiency and accuracy of the generation and analysis of the optimal support guarantee scheme, provide clear action guidelines for commanders and support guarantee personnel, ensure the feasibility and efficiency of the task, and respond quickly and make effective decisions in emergency situations. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0023] Figure 1 FIG. 1 is a flowchart of the support guarantee scheme automatic generation method provided by the application;

[0024] Figure 2a FIG. 2 is a scene diagram of the support guarantee scheme automatic generation method provided by the application;

[0025] Figure 2b FIG. 3 is a structure diagram of the support guarantee scheme automatic generation system provided by the application;

[0026] Figure 2c FIG. 4 is a function diagram of the support guarantee scheme automatic generation system provided by the application;

[0027] Figure 3 FIG. 5 is a structure diagram of the support guarantee scheme automatic generation device provided by the application;

[0028] Figure 4 FIG. 6 is a structure diagram of the electronic device provided by the application. DETAILED DESCRIPTION

[0029] In order to make the objects, technical solutions and advantages of the present application clearer, the technical solutions will be clearly and completely described below in conjunction with the accompanying drawings in the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work belong to the scope of protection of the present application.

[0030] It should be noted that the execution subject involved in the embodiments of the present application can be a support guarantee scheme automatic generation device, or an electronic device. Optionally, the electronic device can include a computer, a mobile terminal, a wearable device, etc.

[0031] The embodiments of the present application will be further described below taking an electronic device as an example.

[0032] As shown in Figure 1 , it is a flowchart of the support guarantee scheme automatic generation method provided by the present application, which can include:

[0033] 101, obtaining a historical data set under a to-be-supported guarantee scene, the historical data set including respective task demand contents at different time points in continuous time.

[0034] The to-be-supported guarantee scene generally refers to a specific situation or environment that needs external support or reinforcement to maintain or restore the normal operation of a certain function, service, activity or system.

[0035] Optionally, the to-be-supported guarantee scene can include a military operation scene, an emergency rescue scene, a public service scene, an industrial production scene and a production operation scene, etc.

[0036] The historical data set can be represented by H={X1,X2,...,X T}, T≥2. Wherein, X t represents the corresponding task demand content at the t time point, X t ={x1,x2,...,x n1}, n1≥2, x i represents the i th data item in the n1 data items, and in addition, the n1 data items can be used to construct a marshalling database, a support force database and a basic resource database, etc. Temporary data.

[0037] Optionally, all data items in the historical data set come from a data source set.

[0038] The data source set can be represented by D={d1,d2,...,d n2}, n2≥2, d irepresents the i-th data source in the n2 data sources, such as intelligence, weather, communication, personnel grouping, and log files.

[0039] Optionally, each data source can include: basic data, log data, user data, and dispatch grouping data.

[0040] The task requirement content refers to specific requirements or descriptions for activities in the to-be-supported support scenario.

[0041] Optionally, the task requirement content can include: task type and quantity. Optionally, the task type can include: material transportation, medical aid, and support personnel; and the quantity can include: the quantity of materials to be transported and the number of support personnel.

[0042] Optionally, the electronic device obtaining the historical data set in the to-be-supported support scenario can include: the electronic device obtaining an original data set in the to-be-supported support scenario; and the electronic device preprocessing the original data set to obtain the historical data set.

[0043] The original data set can be represented by Y = {y1, y2,..., y n3}n3≥n1≥2, y i represents the i-th original data in the n3 original data.

[0044] After obtaining the original data set in the to-be-supported support scenario, the electronic device can preprocess the original data set to obtain a historical data set with higher accuracy, because the original data in the original data set can have redundant data, duplicate data, and missing values, resulting in low accuracy of the original data set.

[0045] Optionally, the electronic device preprocessing the original data set to obtain the historical data set can include at least one of the following implementation manners:

[0046] Implementation manner 1: in the case of preprocessing as a deduplication process, the electronic device performs deduplication processing on the original data set to obtain the historical data set.

[0047] The electronic device can perform deduplication processing on the original data set to obtain a data set without duplicate values, U = {u1, u2,..., uk}, k≥2. k}n3≥n1≥2, y and satisfy [1, 2,..., k] inside, ν The entire process can ensure that there is no duplicate original data in the data set U, thereby improving the accuracy and reliability of the data set U.

[0048] It should be noted that in the case where the preprocessing only includes the de-duplication processing, the data set U is the historical data set H described above.

[0049] Implementation 2: In the case where the preprocessing is missing value supplement, the electronic device supplements the missing values of the original data set to obtain the historical data set.

[0050] The electronic device can use the filling parameter to process the missing values to implement the missing value supplement to obtain the historical data set.

[0051] The filling parameter can include a mean filling parameter, a median filling parameter, and a mode filling parameter, etc.

[0052] Specifically, assuming that the non-missing values of a feature of the original data set are {y1, y2,..., y n3}, in the case where the filling parameter is the mean filling parameter, the filling value is fill_value=mean({x In the case where the filling parameter is the median filling parameter, the filling value is fill_value=median({x i |x i non-missing}), and median(·) represents a median solving function; in the case where the filling parameter is the mode filling parameter, the filling value is fill_value=mode({x i |x i non-missing}), and mode(·) represents a mode solving function. The entire process can ensure the integrity of the original data set, so that each original data has a corresponding value.

[0053] Implementation 3: In the case where the preprocessing is outlier detection, the electronic device detects outliers of the original data set to obtain the historical data set.

[0054] The electronic device can use the Inter-Quartile Range (IQR) method to determine that the quartiles of a feature of the original data set are Q1 and Q3, at this time, the outlier threshold is outlier_threshold=Q1-1.5*IQR or Q3+1.5*IQR, where IQR=Q3-Q1. In the original data set, the data values greater than the outlier threshold are outliers, and the remaining data can construct the historical data set. The entire process evaluates the distribution of all original data through the quartiles, and is relatively stable for shape and distribution changes of the original data.

[0055] Implementation 4: In the case where the preprocessing is data clustering, the electronic device clusters the data of the original data set to obtain the historical data set.

[0056] The electronic device can use K-means clustering analysis, specifically using a first objective function, to perform clustering analysis on all original data in the original data set to minimize the sum of distances of all original data to the cluster center to which the original data belongs, and then determine the historical data set. The first objective function is J1 represents the sum of distances of all original data to the cluster center to which the original data belongs; K represents the total number of clusters, K≥2; c α represents the center of cluster a; w iα represents the original data y i For the weight of cluster a, if the original data y i belongs to cluster a, then w iα =1, otherwise, w iα =0; ||·||2 2 represents the 2-norm of a vector. The entire process can eliminate redundant or duplicate data to improve the accuracy of the historical data set.

[0057] It should be noted that in the above implementation modes 1-4, in terms of data integration, a large amount of historical data can be deeply and quickly integrated and analyzed, and in-depth mining and pattern recognition of a large amount of historical data can be realized to effectively improve the accuracy of the historical data set and provide strong data support for subsequent optimal support guarantee schemes.

[0058] 102. According to the historical data set, determine a prediction result set, the prediction result set including a plurality of task demand content respectively corresponding to a prediction demand change value.

[0059] The prediction demand change value refers to the change amount or change trend of the task demand content at a future time point or time period relative to a current or historical reference point.

[0060] In some embodiments, the electronic device determines a prediction result set according to the historical data set, which can include: the electronic device inputs the historical data set into a long short-term memory network model (LSTM) to obtain an initial prediction result set output by the long short-term memory network model; the electronic device determines a loss function corresponding to the long short-term memory network model according to the historical data set and the initial prediction result set; the electronic device updates the model parameters of the long short-term memory network model according to the loss function to obtain a trained long short-term memory network model; and the electronic device inputs the historical data set into the trained long short-term memory network model to obtain a prediction result set output by the trained long short-term memory network model.

[0061] The long short-term memory network model is a machine learning model, which can be represented as M, and the parameter set corresponding to the long short-term memory network model is Θ.

[0062] After acquiring a historical dataset H, the electronic device can input all historical data in H one by one into a Long Short-Term Memory (LSTM) network model M to obtain an initial prediction result set output by the LTM network model M. All initial prediction results in this set correspond one-to-one with all historical data, and this initial prediction result set can be represented by M(H; Θ). Because the LTM network model M is particularly adept at handling data with time-series dependencies, and because the temporal information in historical data is crucial for predicting future changes in demand in task demand prediction, the LTM network model M can capture these dependencies, thus making more accurate predictions, and the resulting initial prediction results are also relatively accurate.

[0063] Then, the electronic device determines the loss function based on the historical dataset H and the initial prediction result set M(H;Θ), which can be the mean squared error (MSE) loss function. This indicates that L(H,M(H;Θ)) represents the loss value, and H... t This represents the historical demand data up to time t, corresponding to the task requirements. Next, the electronic device minimizes the loss function to obtain the optimization parameter Θ*, which is the parameter value corresponding to the minimum loss value L(H,M(H;Θ)), specifically using the formula Θ*=argminL(H,M(H;Θ)), where argmin(·) represents the optimization parameter determination function. Then, based on this optimization parameter Θ*, the model parameters of the Long Short-Term Memory (LSTM) network model M are updated to obtain the trained LTM network model. Finally, the electronic device inputs the historical dataset H into the trained LTM network model to obtain the prediction result set output by the trained LTM network model. This prediction result set can be represented by P={X ^ T+1 ,X ^ T+2 ,...,X ^ T+h} represents the prediction step size, where h represents the prediction step size. Optional, for X ^ T+h Specifically, formula X is used. ^ T+h =M(H;Θ*) is calculated. The entire process involves iterative training and parameter optimization of the long short-term memory network model M, which continuously improves the accuracy of the long short-term memory network model M, and thus continuously improves the accuracy of the prediction result set.

[0064] It should be noted that in terms of data prediction, step 102 uses pattern recognition and trend prediction technologies to more accurately predict future support and security needs, thereby enabling the rational planning and allocation of resources and providing solid data support for the optimal support and security plan in the future.

[0065] 103. Based on multiple predicted demand change values, combined with resource constraint sets and business rule sets, generate the optimal support and assurance plan.

[0066] The resource constraint set contains restrictions on resources, such as the quantity, type, and availability time of resources, to ensure that the actual availability of resources is not exceeded during the planning process. This resource constraint set can be represented as R = {r1, r2, ..., r...} m} indicates that r m ≥2, r i This represents the i-th resource constraint among m types of resource constraints. The resources involved in each resource constraint, such as technical support personnel and equipment in fields like meteorology, target, defense, camouflage, and communications, can only be used by one task at a time.

[0067] The business rule set contains business rules related to task scheduling and resource allocation, such as task priority, task execution order, resource priority rules, resource allocation rules, and cost-benefit trade-off rules, to ensure that the subsequently determined optimal support and assurance solution meets business needs and rule requirements. This business rule set can be represented as B = {b1, b2, ..., b...} n4} indicates that n4≥2, b i This represents the i-th business rule among n4 types of business rules.

[0068] The optimal support and assurance plan provides clear operational guidelines for commanders and support personnel. Urgent and critical tasks are prioritized within this plan, which considers currently available support resources and their capabilities to ensure task allocation matches resource capacity. The plan is designed within the limits of resources and time to ensure the feasibility and efficiency of the mission.

[0069] Optionally, the electronic device generates an optimal support and assurance scheme based on multiple predicted demand change values, combined with a set of resource constraints and a set of business rules. This may include: the electronic device uses a linear programming optimization algorithm to generate an optimal support and assurance scheme based on multiple predicted demand change values, a set of resource constraints, and a set of business rules.

[0070] Specifically, the optimal support and guarantee scheme for electronic devices can be determined using the formula Plan = O(P,R,B). Here, Plan represents the optimal resource allocation and task scheduling scheme, i.e., the optimal support and guarantee scheme; O(·) represents a linear programming optimization algorithm.

[0071] The entire process employs a linear programming optimization algorithm, which can find the globally optimal solution under given constraints. This means that electronic devices can comprehensively consider the prediction result set P, the resource constraint set R, and the business rule set B, improving decision-making efficiency and response speed, thereby ensuring that the generated support and assurance solution is globally optimal.

[0072] In some embodiments, the electronic device generates an optimal support and assurance plan based on multiple predicted demand change values, combined with a resource constraint set and a business rule set. This may include: the electronic device acquiring a task set and an initial schedule for the scenario to be supported and assured, wherein the tasks in the task set correspond one-to-one with the start execution time in the initial schedule; the electronic device determining a target task combination based on the task set, wherein the tasks in the target task combination belong to the task set; the electronic device generating a target schedule based on the task set, the initial schedule, and the resource constraint set; and the electronic device generating an optimal support and assurance plan based on multiple predicted demand change values, combined with the target task combination, the target schedule, and the business rule set.

[0073] The task set includes a series of tasks that need to be executed. This task set can be represented as T = {t1, t2, ..., t...} n5} indicates that n5≥2, t i This represents the i-th task out of n5 tasks. It should be noted that the scope of this task includes dispatching support personnel to provide technical assistance in fields such as meteorology, target management, defense, camouflage, and communications.

[0074] The initial timetable includes the start time for each of the above series of tasks. This initial timetable can be represented as S = {s1, s2, ..., s...} n5} indicates that s i This represents the start execution time for the i-th task. It should be noted that the goal of generating the initial schedule is to assign a start execution time to each task, ensuring that all tasks can be executed while satisfying resource constraints and other constraints (such as time windows).

[0075] In the process of generating the optimal support guarantee scheme, after obtaining the task set and the initial schedule, the electronic device can rearrange all or part of the tasks in the task set to obtain a target task combination corresponding to a minimum task execution cost. The tasks in the target task combination are more detailed and have a small structured granularity, which facilitates subsequent execution and supervision of the tasks. Meanwhile, the electronic device determines a target schedule that meets the resource constraints based on the task set and the initial schedule. In this way, the electronic device can quickly process a large amount of task data, determine the target task combination and generate the target schedule in an automated and intelligent manner, and greatly improve work efficiency. Finally, the electronic device generates the optimal support guarantee scheme based on the target task combination, the target schedule, and the business rule set according to the plurality of predicted demand change values. The entire process considers the task set, the initial schedule, the resource constraint set, and the predicted demand change values, which can ensure reasonable allocation and optimized use of resources, avoid resource waste and bottlenecks, and also help reduce costs and improve resource use efficiency, thereby enhancing the overall benefits of support guarantee work.

[0076] In some embodiments, the electronic device determines the target task combination according to the task set, which can include: the electronic device rearranges the tasks in the task set to obtain a plurality of task combinations; for each group of task combinations, the electronic device determines a weight and an estimated cost corresponding to each task in the task combination; according to all weights and all estimated costs, the electronic device determines a task execution cost corresponding to the task combination; and the electronic device determines the target task combination corresponding to the minimum task execution cost among the plurality of task execution costs.

[0077] Specifically, the electronic device can use the formula to determine the target task combination. Wherein, π represents the minimum task execution cost; Π(T) represents a set of all possible arrangements of the task set T, i.e., a set of a plurality of task combinations; π' represents any task combination in the plurality of task combinations; wπ'(i) is the weight of the i th task π'(i) in the task combination π' in the task combination π'; cπ'(i) is the estimated cost corresponding to the i th task π'(i) in the task combination.

[0078] Wherein, wπ'(i) can be determined according to a target factor corresponding to the i th task π'(i) in the task combination. Optionally, the target factor can include at least one of the following: priority, importance, urgency, dependency relationship, and resource limitation, etc.

[0079] In this way, in terms of task scheduling, the entire process not only considers the priority of the task, but also fully considers the importance, urgency, dependency relationship and resource limitation of the task, so that the tasks in the above target task combination have the best execution order, and ensures that the key and urgent tasks are executed preferentially in the case of limited resources.

[0080] In some embodiments, the electronic device generates a target schedule according to the task set, the initial schedule and the resource constraint set, which can include: the electronic device determines the initial cost of each task in the task set when executed at the corresponding start execution time; the electronic device determines the corresponding use constraint of each resource constraint in the resource constraint set according to the initial schedule and the resource constraint set; the electronic device determines the target schedule with the minimum total cost according to all initial costs and all use constraints.

[0081] Specifically, the electronic device can use the formula s.t.g j (S,R)≤0, to generate the target schedule. Wherein, c i (s i ) represents the initial cost of the i-th task in the task set when executed at the corresponding start execution time; g j (S,R) represents the use constraint function of the j-th resource, that is, when the j-th resource is used by multiple tasks at the same time, the function g j (S,R) is greater than 0.

[0082] It should be noted that the optimization objective of the above formula is to find the target schedule with the minimum total cost, while satisfying all resource constraints and business rule constraints.

[0083] In this way, in terms of time management, the entire process combines time management and resource allocation algorithms to generate a compact and feasible target schedule, fully utilizes limited resources, ensures that each task can be effectively executed within the specified time, while maximizing the use of resources, and thus improves the efficiency of task execution.

[0084] Optionally, the electronic device generates an optimal support guarantee scheme according to the plurality of predicted demand change values, in combination with the resource constraint set and the business rule set, which can include: the electronic device generates an initial support guarantee scheme according to the plurality of predicted demand change values, in combination with the resource constraint set and the business rule set; the electronic device dynamically plans the initial support guarantee scheme to obtain the optimal support guarantee scheme.

[0085] In the dynamic planning process, it is assumed that P γ represents the state at the γ stage, which can include multiple dimensions, such as the configuration of support personnel, task progress, resource consumption, etc.; and the second objective function is Wherein, J2 represents the support guarantee scheme when the cost or benefit is optimal, that is, the optimal support guarantee scheme; N represents the total number of stages, N≥2; C γ (·) represents the cost function or benefit function of the γth stage. The optimization problem is to solve min(u1, u2,..., u N )J2 or max(u1, u2,..., u N )J2, that is, to find a set of action sequences u k , so that the second objective function J2 is optimal. The output form of the optimization scheme using the dynamic programming algorithm ensures that the generated optimal support guarantee scheme is clear in content, easy to understand, and can be flexibly adjusted according to actual conditions, improving the feasibility and adaptability of the scheme.

[0086] In some embodiments, the method can further include at least one of the following implementations:

[0087] Implementation 1: For each task in the task set, the following operations are performed: during multiple times of simulating the execution of the task, the electronic device determines the occurrence probability and the impact degree of the target risk on the task at each simulation; the electronic device determines the task execution result corresponding to the task under the target risk according to the resource constraint set, the occurrence probability and the impact degree; and the electronic device determines the impact of the target risk on the execution of the task according to the distribution of the multiple task execution results.

[0088] Specifically, the electronic device can use the formula S α = g(R, P α (r β ), I α (r β )) to determine the impact of the βth risk, which is the target risk, on the execution of the task.

[0089] Wherein, S α represents the task execution result corresponding to the execution of the task in the αth simulation in M simulations, M≥2; g(·) represents a simulation function; P α (r β ) represents the occurrence probability of the βth risk on the task in the αth simulation of the task; and I α (r β ) represents the impact degree of the βth risk on the task in the αth simulation of the task.

[0090] For the βth risk, the electronic device obtains as many task execution results as the number of simulations, and then determines the distribution of the task execution results to evaluate the impact of the βth risk on the execution of the task. Based on this, the electronic device can evaluate the impact of different risks on the execution of the task.

[0091] The above decision-making process based on data and simulation no longer relies on artificial experience and judgment, is more scientific and objective, improves objectivity and accuracy, helps to reduce delays and failures in task execution, and improves efficiency and success rate of task execution.

[0092] Implementation 2: The electronic device pre-constructs a risk matrix, where the rows represent risk types, the columns represent risk levels, and the risk matrix can be represented by F; at the same time, the electronic device constructs a strategy library, where one risk response measure corresponds to one risk type and one risk level; and the electronic device determines the target risk response measure for the potential risk based on the risk matrix and the strategy library when detecting the potential risk.

[0093] Wherein, the risk type can be represented by r = {r1, r2, …, r n6 n6≥2, r β represents the βth risk type (referred to as the βth risk) among the n6 risk types.

[0094] Optionally, the risk level includes low risk, medium risk, high risk, etc.

[0095] The strategy library includes a series of executable risk response measures. The strategy library can be represented by Q = {q1, q2, …, q n6 n6≥2, q i represents the i th risk response measure among the n6 risk response measures.

[0096] After the electronic device detects the potential risk, it can quantify and analyze the potential risk. Specifically, based on the risk matrix, the risk type and risk level corresponding to the potential risk are determined, and the corresponding target risk response measure is queried. Specifically, the formula q β = F(r β , risk level) can be used. In this way, the electronic device can automatically recommend the corresponding effective risk response measure to reduce the impact and uncertainty of the potential risk on task execution, which greatly reduces the uncertainty in the task execution process and enhances the robustness and reliability of the target risk response measure.

[0097] Implementation 3: The electronic device saves the current version corresponding to the optimal support and guarantee scheme; if it is detected that the optimal support and guarantee scheme receives new data, and / or the degree of change of the key parameters corresponding to the optimal support and guarantee scheme is greater than a preset threshold, the current version is updated.

[0098] Wherein, the current version corresponding to the optimal support and guarantee scheme can be represented by P v , and v represents the version number.

[0099] The electronic device can save the current version corresponding to the optimal support and guarantee scheme after obtaining the optimal support and guarantee scheme, so that consistent and accurate data can be accessed at any time. Then, if it is detected that the optimal support and guarantee scheme receives new data, the current version is updated. And / or, the degree of change of the key parameters corresponding to the optimal support and guarantee scheme is obtained. If the absolute value of the degree of change of the key parameters is greater than the preset threshold, the current version is updated. The entire process enables the electronic device to immediately identify and trigger the update mechanism in the case of receiving new data or the degree of change of the key parameters exceeding the preset threshold. This timely response enables the optimal support and guarantee scheme to quickly adapt to new environments and changes in demand, maintaining the effectiveness and accuracy of the optimal support and guarantee scheme, and enabling the optimal support and guarantee to quickly respond when adjustment is needed.

[0100] It should be noted that if the current version is updated, the version number after the update can be represented by v+1. At this time, in the case of updating the optimal support and guarantee scheme from version v to version v+1, a change set ΔP v→v+1 The change set contains all the modified, added or deleted contents. Based on this, by linking a series of change sets, a complete version history corresponding to the optimal support and guarantee scheme can be formed, that is,

[0101] Implementation 4: The electronic device outputs the optimal support and guarantee scheme individually.

[0102] The electronic device can display the optimal support and guarantee scheme in text and / or broadcast the optimal support and guarantee scheme in voice to realize individualized output of the optimal support and guarantee scheme, so that the user can intuitively and easily understand the optimal support and guarantee scheme.

[0103] In the embodiment of the application, a historical data set under a to-be-supported scenario is obtained, the historical data set including respective task demand contents at different time points in continuous time; a prediction result set is determined according to the historical data set, the prediction result set including respective predicted demand change values of the multiple task demand contents; and an optimal support and guarantee scheme is generated according to the multiple predicted demand change values, in combination with a resource constraint set and a business rule set. The method aims to quickly generate and analyze the optimal support and guarantee scheme through an automated and intelligent method, improve the efficiency and accuracy of generation and analysis of the optimal support and guarantee scheme, provide clear action guidelines for commanders and support and guarantee personnel, and ensure the feasibility and efficiency of the task to cope with rapid response and effective decision-making in emergency situations.

[0104] In order to better understand the embodiments of the application, the automatic generation method of the support and guarantee scheme is further described as follows:

[0105] For example, Figure 2aThe image shown is a schematic diagram illustrating a scenario of the automatic generation method for support and protection schemes provided by this invention. From... Figure 2a It can be seen from this:

[0106] (1) Regarding the integration, analysis, and planning results: The electronic equipment performed data collection, data preprocessing, data mining, and demand forecasting. Specifically, the electronic equipment collected data based on basic data, log data, user data, and dispatch group data to obtain the raw dataset; then, it preprocessed the raw dataset to obtain the historical dataset, and then performed demand forecasting on the historical dataset to obtain the forecast result set.

[0107] (2) In terms of describing the task content in detail: the electronic device uses the prediction result set as a basis, combined with business data (such as task set) and planning scheme data, to realize the preliminary planning of the scheme, and then sorts the tasks to obtain the target task combination.

[0108] (3) Regarding the clear execution order and schedule: Electronic devices generate target schedules based on business data.

[0109] (4) In terms of risk assessment and response measures: After generating the optimal support and protection plan based on the above-mentioned target task combination and target timetable, the electronic equipment can simulate the risks based on the basic data and determine the risk response measures for potential risks.

[0110] (5) In terms of scheme output and update: Based on the plan scheme data, the electronic equipment performs dynamic planning and scheme version control of the optimal support guarantee scheme, and then outputs the final optimal support guarantee scheme.

[0111] like Figure 2b The diagram shown is a structural schematic of the automatic generation system for support and assurance solutions provided by this invention. Figure 2b It can be seen from this: Figure 2a The integrated analysis and planning results, detailed descriptions of task content, clear execution sequence and timelines, and solution output and updates are all implemented at the application layer of the support and assurance solution automatic generation system. This support and assurance solution automatic generation system also includes a service provision layer, a software development layer, a resource element layer, and a basic support layer, providing technical support to the application layer.

[0112] like Figure 2c The diagram shown is a functional schematic of the automatic generation system for support and assurance solutions provided by this invention; from Figure 2c As can be seen from this, the functions of the automatic support and guarantee plan generation system can include: integrating analysis and planning results, resource allocation and optimization, planning maneuver routes, grouping and task allocation, and generating support and guarantee plans.

[0113] The support guarantee scheme automatic generation device provided by the application is described below, and the support guarantee scheme automatic generation device described below can be correspondingly referred to the support guarantee scheme automatic generation method described above.

[0114] As shown in Figure 3 , it is a structural schematic diagram of the support guarantee scheme automatic generation device provided by the application, which can include:

[0115] The acquisition module 301 is configured to acquire a historical data set under a to-be-supported guarantee scene, the historical data set including respective task demand contents at different time points in continuous time;

[0116] The processing module 302 is configured to determine a prediction result set according to the historical data set, the prediction result set including respective predicted demand change values of multiple task demand contents; and generate an optimal support guarantee scheme according to the multiple predicted demand change values, in combination with a resource constraint set and a business rule set.

[0117] Optionally, the processing module 302 is specifically configured to input the historical data set into a long short-term memory network model to obtain an initial prediction result set output by the long short-term memory network model; determine a loss function corresponding to the long short-term memory network model according to the historical data set and the initial prediction result set; update model parameters of the long short-term memory network model according to the loss function to obtain a trained long short-term memory network model; and input the historical data set into the trained long short-term memory network model to obtain the prediction result set output by the trained long short-term memory network model.

[0118] Optionally, the processing module 302 is specifically configured to acquire a task set and an initial schedule under the to-be-supported guarantee scene, a task in the task set corresponding to a start execution time in the initial schedule one by one; determine a target task combination according to the task set, the target task combination including tasks in the task set; generate a target schedule according to the task set, the initial schedule and the resource constraint set; and generate the optimal support guarantee scheme according to the multiple predicted demand change values, in combination with the target task combination, the target schedule and the business rule set.

[0119] Optionally, the processing module 302 is specifically configured to rearrange the tasks in the task set to obtain multiple task combinations; determine respective weights and estimated costs of all tasks in each task combination; determine a task execution cost corresponding to the task combination according to all the weights and all the estimated costs; and determine the target task combination corresponding to the minimum task execution cost in the multiple task execution costs.

[0120] Optionally, the processing module 302 is specifically configured to determine initial costs of all tasks in the task set when the tasks are executed at corresponding start execution times respectively; determine corresponding usage constraints of all resource constraints in the resource constraint set according to the initial schedule and the resource constraint set; and determine the target schedule with the minimum total cost according to all initial costs and all usage constraints.

[0121] Optionally, the processing module 302 is further configured to perform the following operations for each task in the task set: determine an occurrence probability and an influence degree of the target risk on the task at each simulation time during multiple times of simulating execution of the task; determine a task execution result corresponding to the task under the target risk according to the resource constraint set, the occurrence probability and the influence degree; and determine an influence situation of the target risk on execution of the task according to a distribution of multiple task execution results.

[0122] Optionally, the processing module 302 is further configured to save a current version corresponding to the optimal support and guarantee scheme; and update the current version if it is detected that the optimal support and guarantee scheme receives new data and / or a key parameter corresponding to the optimal support and guarantee scheme changes by more than a preset threshold.

[0123] As shown in Figure 4 FIG. 1 is a structural schematic diagram of an electronic device provided by the present application, which can include a processor 410, a communications interface 420, a memory 430 and a communications bus 440, wherein the processor 410, the communications interface 420 and the memory 430 complete mutual communication through the communications bus 440. The processor 410 can invoke a logical instruction in the memory 430 to execute a support and guarantee scheme automatic generation method, which includes: acquiring a historical data set under a to-be-supported and guaranteed scene, the historical data set including respective corresponding task demand contents at different times within a continuous time; determining a prediction result set according to the historical data set, the prediction result set including respective corresponding prediction demand change values of multiple task demand contents; and generating an optimal support and guarantee scheme according to the multiple prediction demand change values, in combination with a resource constraint set and a business rule set.

[0124] In addition, the logic instructions in the memory 430 described above can be implemented in the form of software functional units and sold or used as independent products, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the part of the technical solutions that make essential contributions to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0125] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the support and guarantee scheme automatic generation method provided by the above-mentioned methods. The method comprises: obtaining a historical data set under a to-be-supported scenario, the historical data set comprising respective task demand contents at different time points in continuous time; determining a prediction result set according to the historical data set, the prediction result set comprising respective prediction demand change values of a plurality of task demand contents; and generating an optimal support and guarantee scheme according to a plurality of prediction demand change values, in combination with a resource constraint set and a business rule set.

[0126] In another aspect, the present application also provides a computer program product, which comprises a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program can be executed by a processor to enable a computer to execute the support and guarantee scheme automatic generation method provided by the above-mentioned methods. The method comprises: obtaining a historical data set under a to-be-supported scenario, the historical data set comprising respective task demand contents at different time points in continuous time; determining a prediction result set according to the historical data set, the prediction result set comprising respective prediction demand change values of a plurality of task demand contents; and generating an optimal support and guarantee scheme according to a plurality of prediction demand change values, in combination with a resource constraint set and a business rule set.

[0127] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed to multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0128] Through the description of the above embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course can also be realized by hardware. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in each embodiment or some parts of the embodiments.

[0129] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for automatically generating a support assurance program, characterized by, The method comprises the following steps: acquiring a historical data set under a to-be-supported support scenario, the historical data set comprising respective task demand contents at different time points in continuous time; determining a prediction result set according to the historical data set, the prediction result set comprising respective prediction demand change values corresponding to a plurality of task demand contents; acquiring a task set and an initial schedule under the to-be-supported support scenario, the tasks in the task set corresponding to starting execution times in the initial schedule one by one; determining a target task combination according to the task set, the tasks in the target task combination belonging to the task set; generating a target schedule according to the task set, the initial schedule and a resource constraint set; generating an optimal support and support scheme according to the plurality of prediction demand change values, the target task combination, the target schedule and a business rule set; the step of determining a prediction result set according to the historical data set comprises the following steps: inputting the historical data set into a long short-term memory network model to obtain an initial prediction result set output by the long short-term memory network model; determining a loss function corresponding to the long short-term memory network model according to the historical data set and the initial prediction result set; updating model parameters of the long short-term memory network model according to the loss function to obtain a trained long short-term memory network model; inputting the historical data set into the trained long short-term memory network model to obtain the prediction result set output by the trained long short-term memory network model.

2. The method of claim 1, wherein, the step of determining a target task combination according to the task set comprises the following steps: rearranging the tasks in the task set to obtain a plurality of task combinations; determining respective weights and estimated costs of all tasks in each group of task combinations according to each group of task combinations; determining a task execution cost corresponding to each group of task combinations according to all weights and all estimated costs; determining the target task combination corresponding to the minimum task execution cost in the plurality of task execution costs.

3. The method of claim 1, wherein, the step of generating a target schedule according to the task set, the initial schedule and the resource constraint set comprises the following steps: determining initial costs of all tasks in the task set when the tasks are executed at corresponding starting execution times; determining respective use constraints of all resource constraints in the resource constraint set according to the initial schedule and the resource constraint set; determining the target schedule with the minimum total cost according to all initial costs and all use constraints.

4. The method according to any one of claims 1 to 3, characterized in that, The method further comprises the following steps: for each task in the task set, the following operations are performed: determining a probability of occurrence and an influence degree of a target risk on the task in each simulation in a plurality of times of simulating execution of the task; determining a task execution result corresponding to the task under the target risk according to the resource constraint set, the probability of occurrence and the influence degree; determining an influence situation of the target risk on execution of the task according to a distribution of a plurality of task execution results.

5. The method according to any one of claims 1 to 3, characterized in that, The method further comprises the following steps: saving a current version corresponding to the optimal support and support scheme; If it is detected that the optimal support guarantee scheme receives new data, and / or the degree of change of the key parameters corresponding to the optimal support guarantee scheme is greater than a preset threshold, the current version is updated.

6. An apparatus for implementing the method of automatically generating a support assurance scheme according to any one of claims 1 to 5, characterized in that, The method comprises the following steps: An acquisition module is configured to acquire a historical data set in a to-be-supported scenario, the historical data set comprising respective task demand contents at different time points in continuous time; A processing module is configured to determine a prediction result set according to the historical data set, the prediction result set comprising respective predicted demand change values of a plurality of task demand contents; acquire a task set and an initial schedule in the to-be-supported scenario, a task in the task set corresponding to a start execution time in the initial schedule in a one-to-one manner; determine a target task combination according to the task set, the target task combination comprising tasks belonging to the task set; generate a target schedule according to the task set, the initial schedule and a resource constraint set; and generate an optimal support guarantee scheme according to the plurality of predicted demand change values, in combination with the target task combination, the target schedule and a business rule set.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the support guarantee scheme automatic generation method according to any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the support guarantee scheme automatic generation method according to any one of claims 1 to 5.

Images