A training resource intelligent recommendation and scheduling method and system

Abstract

The present invention provides a method and system for intelligent recommendation and scheduling of training resources, relating to the field of data processing technology. The method comprises: obtaining a position data set and a resource data set, determining subject association relationships and position mapping relationships based thereon, extracting subject constraint information and position dependency information, performing time matching between the training requirement duration of each subject and the available resource duration, obtaining a set of candidate scheduling pairs, performing resource conflict detection, removing conflicting pairs, and sorting them according to position dependency information, extracting a set of candidate resources corresponding to each subject, calculating a score for each candidate resource, obtaining a resource score set, selecting the resource identifier with the largest score in each subject as a recommended resource, and obtaining scheduling plan data. The present invention can make a comprehensive judgment on resource scheduling when resources are limited.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for intelligent recommendation and scheduling of training resources. Background Art

[0002] In existing basic training systems, training resource scheduling typically relies on manual training plans and course assignments. After importing current-level training content based on the superior plan, implementation plans must be manually developed, with subjects matched to positions and exercises configured one by one. Subject assignments primarily adhere to the training syllabus, combined with position training requirements, and a training resource system is established across multiple dimensions, including test papers, samples, and lesson plans. Although some systems have introduced subject priority management mechanisms, resource allocation is still primarily based on static tables, lacking dynamic adaptability and intelligent recommendation capabilities. This can make accurate and rapid scheduling responses difficult, especially when training resources are diverse and positions are complex.

[0003] For example, in the automatic generation of course plans, existing solutions are often based on simplified priority algorithms, using factors such as course start and end times and daily class hours as the basic basis for scheduling, and employing rule-based logic to advance training schedules day by day. However, these approaches generally overlook resource conflict analysis across multiple positions and the interdependence of training tasks, which can easily lead to problems such as course conflicts, uneven training intensity, and duplicate resource allocation. In cases of overlapping positions, the same resource may be contested by multiple courses, and the system may lack a comprehensive judgment and dynamic adjustment mechanism for resource priorities. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for intelligent recommendation and scheduling of training resources, aiming to solve the problems mentioned in the background technology.

[0005] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0006] In a first aspect, a method for intelligently recommending and scheduling training resources is provided, the method comprising:

[0007] Obtain position data sets and resource data sets, determine subject association relationships and position mapping relationships based on them, extract subject constraint information and position dependency information, and obtain constraint data sets;

[0008] According to the constraint data set, the training demand duration and resource availability duration of each subject are matched to obtain a set of candidate scheduling pairs;

[0009] Perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain the conflict filter set;

[0010] According to the conflict filter set, the conflicting pairs are removed, and the remaining candidate scheduling pairs are used as the valid scheduling set. They are sorted according to the position dependency information to obtain a sorted data sequence.

[0011] According to the sorted data sequence, the candidate resource set corresponding to each subject is extracted, and the score value of each candidate resource is calculated to obtain the resource score set;

[0012] According to the resource scoring set, the resource identifier with the largest scoring value in each subject is selected as the recommended resource, and its time period label and position number in the valid scheduling set are extracted to obtain the scheduling plan data.

[0013] Furthermore, the position data set and resource data set are obtained, and based on them, the subject association relationship and position mapping relationship are determined, and the subject constraint information and position dependency information are extracted to obtain the constraint data set, including:

[0014] Obtain the position numbers of all positions, the subject sets corresponding to each position, and the training required duration for each subject to obtain the position data set; obtain the resource numbers of all resources, the resource availability duration, the subject types supported by the resources, and the position identifiers to which the resources belong to obtain the resource data set;

[0015] Based on the position data set, determine the occurrence of each subject in different positions, extract the continuous training relationship and non-parallel constraints between subjects, and obtain subject association relationship data;

[0016] According to the position data set, determine the matching structure between the position identifier and the subject set, and obtain the position mapping relationship data;

[0017] According to the subject association relationship data, the order relationship, mutually exclusive execution flag and training stage overlap relationship between each subject and other subjects are extracted to obtain the subject constraint information data;

[0018] Based on the position mapping relationship data and the position identifiers of the resources in the resource data, the sharing and preemption of resources between different positions are analyzed to obtain position dependency information data;

[0019] The subject constraint information data and the position dependency information data are merged to obtain a rule information set, and the subject identification, position number, time condition and execution relationship label are extracted to obtain a constraint data set.

[0020] Furthermore, based on the constraint data set, the training requirement duration of each subject and the available resource duration are time-matched to obtain a set of candidate scheduling pairs, including:

[0021] Obtain the subject ID, total training duration, daily training duration requirement, and whether segmented training is allowed to obtain subject duration requirement data; obtain the resource ID, continuous available time period, and resource support subject ID set to obtain resource time segment data;

[0022] According to the subject constraint information in the constraint data set, other subjects that have a mutually exclusive relationship with the target subject during the arrangement period are screened out to obtain a set of matching subjects;

[0023] According to the resource time segment data, all resource identifiers required for each subject identification are extracted, and combined with the continuous available time period of each resource, the resource schedulable window set is obtained;

[0024] For each subject in the subject duration demand data, combined with the flag of whether segmented training is allowed, a set of time periods that can cover the total training duration are selected from the resource schedulable window set to obtain a time matching item;

[0025] For each time matching item, determine the subject ID, matching resource ID, planned start time, planned end time, and daily training duration value, generate a scheduling pair candidate record, and summarize all scheduling pair candidate records to obtain a candidate scheduling pair set.

[0026] Furthermore, resource conflict detection is performed on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and a conflict filter set is obtained, including:

[0027] Group all candidate scheduling pair records in the candidate scheduling pair set by resources, cluster all scheduling pairs according to resource identifiers, and obtain a resource pair set;

[0028] Analyze any two scheduling pairs in the resource pair set to determine whether their corresponding start and end times have overlapping intervals. If the result is yes, mark them as time-overlapping pairs.

[0029] Based on the time overlapping pairs, further determine whether their corresponding subject identifiers are different. If the result is yes, mark the pairs as conflicting pairs and obtain a conflicting identifier list;

[0030] Each scheduling pair in the conflict identification list is removed from the candidate scheduling pair set, and the remaining part of the candidate scheduling pair set is used as the resource conflict-free pair set;

[0031] The conflict identification list and the resource non-conflict pair group set are merged and marked as conflict scheduling subset and non-conflicting scheduling subset respectively to obtain the conflict filter set.

[0032] Furthermore, based on the conflict filter set, the conflicting pairs are removed, and the remaining candidate scheduling pairs are used as the valid scheduling set. They are sorted according to the position dependency information to obtain a sorted data sequence, including:

[0033] According to the conflict filter set, the scheduling pairs marked as conflicting scheduling subsets are removed from the candidate scheduling pair set, and the remaining scheduling pairs are retained to obtain the valid scheduling set;

[0034] According to the valid scheduling set, the subject identification and position number information in each scheduling pair is extracted to construct the subject position comparison data;

[0035] According to the position dependency information, determine the priority relationship set between different positions and form a position dependency graph;

[0036] Map each position number in the subject position comparison data to the position dependency graph, establish the transitive dependency relationship between subjects, and obtain the subject scheduling directed graph;

[0037] Sort the subject scheduling directed graph, determine the execution order list of subjects according to the direction of the dependency edges in the graph, and obtain the sorted path sequence;

[0038] Map the subject identifiers in the sorting path sequence back to the corresponding scheduling pairs in the valid scheduling set, arrange them in order, and obtain a sorted data sequence.

[0039] Furthermore, based on the sorted data sequence, the candidate resource sets corresponding to each subject are extracted, and the scoring values ​​of each candidate resource are calculated to obtain a resource scoring set, including:

[0040] According to each subject in the sorted data sequence, all resources associated with it are extracted from the effective scheduling set to form a course candidate resource set and obtain subject resource mapping data;

[0041] Based on the subject resource mapping data, for each course candidate resource, count the number of times it appears in the course candidate resource set to obtain the resource appearance count, calculate the proportion of resources currently selected by other subjects, extract the proportion of time the current subject uses the resource, and calculate the resource usage frequency item based on the interaction of the three;

[0042] For each resource, we count the set of position numbers it can serve, extract the subject coverage ratio for each position, and normalize the position priority weight based on the graph depth of the position node in the position dependency graph. We then calculate the position coverage frequency item based on the subject coverage ratio and position priority weight.

[0043] For each subject in the sorted data sequence, record its ranking in the sorting, calculate its priority coefficient based on the total number of subjects, and extract its completed training time and planned total training time to calculate the subject completion item;

[0044] The resource usage frequency item, position coverage frequency item and subject completion item are weighted and summed to calculate the score value of different resources for each subject. Each score value and its corresponding subject identifier and resource identifier are combined into a score record to obtain a resource score set.

[0045] Furthermore, based on the resource scoring set, the resource identifier with the largest scoring value in each subject is selected as the recommended resource, and its time period label and position number in the valid scheduling set are extracted to obtain the scheduling plan data, including:

[0046] For each subject ID, select the resource ID with the highest score from the resource score set and determine it as the recommended resource for the subject;

[0047] Find the scheduling pair corresponding to the subject and its corresponding recommended resource in the valid scheduling set, and extract the time period label in the scheduling pair as the training start and end time;

[0048] Extract the corresponding position number from the scheduling pair group as the associated position identifier of the subject;

[0049] According to the subject identification, recommended resource identification, training start and end time and position number, the resource allocation detail records are obtained, and all resource allocation detail records are summarized to obtain the scheduling plan data.

[0050] In a second aspect, a training resource intelligent recommendation and scheduling system is provided, the system comprising:

[0051] The constraint module is used to obtain the position data set and resource data set, and determine the subject association relationship and position mapping relationship based on them, extract the subject constraint information and position dependency information, and obtain the constraint data set;

[0052] The matching module is used to match the training requirement duration of each subject with the available resource duration based on the constraint data set to obtain a set of candidate scheduling pairs;

[0053] A detection module is used to perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain a conflict filter set;

[0054] The sorting module is used to remove conflicting pairs from the conflict filter set, take the remaining candidate scheduling pairs as the valid scheduling set, and sort them according to the job dependency information to obtain a sorted data sequence;

[0055] The scoring module is used to extract the candidate resource set corresponding to each subject based on the sorted data sequence, and calculate the score value of each candidate resource to obtain a resource scoring set;

[0056] The scheduling module is used to select the resource identifier with the largest score value in each subject as the recommended resource based on the resource scoring set, and extract its time period label and position number in the valid scheduling set to obtain the scheduling plan data.

[0057] The above solution of the present invention includes at least the following beneficial effects:

[0058] The present invention obtains a position data set and a resource data set, and the system establishes a mapping relationship between subjects and positions, and extracts information such as the training duration of the subject, the time when resources are available, and forms a complete constraint data set. In the scheduling stage, the system accurately matches the duration of the subject's demand for resources with the time window in which the resources are continuously available, ensuring that each training task can be completed within the required time, and avoiding the situation where resources do not fully meet the training requirements and are forcibly allocated. This matching mechanism supports the flexible scheduling of training tasks under the prerequisites, and at the same time, combines whether the subject supports the setting of segmented training to achieve configurable scheduling granularity. In situations where resource allocation is limited or training is intensive, this method can significantly improve the system's scheduling success rate and the completion quality of training tasks, and enhance the overall responsiveness of the training system.

[0059] The present invention sets up a special resource conflict detection mechanism to judge the resource overlap of candidate scheduling pairs. In particular, when it is detected that two or more subjects request the same resource and there are overlapping intervals in time, the scheduling scheme of this group is identified as a conflicting pair and eliminated. Unlike the existing technology, this mechanism not only relies on static rule judgment, but also combines the time tags and resource identifiers generated during the scheduling process for real-time analysis. As a result of this processing, the system can dynamically screen out unreasonable scheduling options and ensure that each resource is only occupied by a unique subject in the same time period, thereby avoiding the common resource "preemption" problem in traditional training scheduling and improving the stability of the system and controllability during actual execution. This operation is particularly critical for training tasks with tight job resources and high subject parallelism, ensuring that the scheduling plan is operational during implementation.

[0060] The present invention constructs a position dependency graph through position dependency information, and converts it into a scheduling directed graph of subjects, and then performs topological sorting to obtain a subject sorting path. This sorting logic not only takes into account the order of training within a single position, but also integrates the priorities and dependencies across positions. For example, some subjects need to be executed after the completion of specific basic subjects. This mechanism effectively solves the problem of chaotic training order in traditional scheduling systems due to the lack of task dependency judgment, and ensures that the scheduling results have a rigorous logical closed loop and training rationality. By driving the scheduling execution through sorted data, the system can guide the subjects to be activated at the right time, ensuring the continuity of the training process and the integrity of the connection between position tasks, thereby achieving a more scientific training organization method and improving the synergy and docking of the overall training system.

[0061] The present invention introduces a resource scoring mechanism to conduct a multi-dimensional comprehensive scoring of each resource, including resource usage frequency, position coverage frequency and subject completion. Resource usage frequency reflects the current utilization rate of resources in scheduling, position coverage frequency considers the adaptability of resources in multiple positions and their priority weights in the position dependency graph, and subject completion evaluates the proportion of remaining training time of the subject. The system generates a score value by weighted summation of these three indicators to select the most suitable resource for the subject. Compared with static weight or priority decision logic, this scoring mechanism fully reflects the principle of resource scheduling optimization in a dynamic environment, achieving a balance between reasonable allocation of resources, maximization of utilization and guarantee of training goals. The final recommendation result not only has a high degree of matching, but also has stronger interpretability and goal orientation, providing a solid decision-making basis for the intelligent scheduling system.

[0062] The present invention can support the segmented execution of training subjects. During the scheduling stage, it can identify whether the subject is allowed to be split into multiple training segments for execution, and match the continuous available time periods of resources accordingly. This design greatly improves the flexibility of resource scheduling, allowing the system to complete the training task configuration to the maximum extent under conditions of resource fragmentation or local conflicts, and alleviate the impact of resource shortages on the overall scheduling success rate. In the case of uneven time distribution of training subjects or high degree of position overlap, the system can flexibly splice low-frequency idle resources through segmented training, improve resource utilization, and reduce training window time. At the same time, for subjects that do not support segmented training, the complete time window can be forced to match to ensure training continuity. This differentiated processing strategy fully adapts to a variety of complex training models and management requirements, and improves the system's adaptability to diverse scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flowchart of a method for intelligent recommendation and scheduling of training resources provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0064] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0065] like Figure 1 As shown, an embodiment of the present invention provides a method for intelligent recommendation and scheduling of training resources, the method comprising:

[0066] Obtain position data sets and resource data sets, determine subject association relationships and position mapping relationships based on them, extract subject constraint information and position dependency information, and obtain constraint data sets;

[0067] According to the constraint data set, the training demand duration and resource availability duration of each subject are matched to obtain a set of candidate scheduling pairs;

[0068] Perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain the conflict filter set;

[0069] According to the conflict filter set, the conflicting pairs are removed, and the remaining candidate scheduling pairs are used as the valid scheduling set. They are sorted according to the position dependency information to obtain a sorted data sequence.

[0070] According to the sorted data sequence, the candidate resource set corresponding to each subject is extracted, and the score value of each candidate resource is calculated to obtain the resource score set;

[0071] According to the resource scoring set, the resource identifier with the largest scoring value in each subject is selected as the recommended resource, and its time period label and position number in the valid scheduling set are extracted to obtain the scheduling plan data.

[0072] In this embodiment of the present invention, a position dataset and a resource dataset are obtained, and based on them, subject association relationships and position mapping relationships are determined, subject constraint information and position dependency information are extracted, and a constraint dataset is obtained. By systematically extracting structured data, a comprehensive dependency and constraint logic between subjects and positions is formed, effectively supporting the subsequent scheduling strategy formulation;

[0073] Based on the constraint data set, the training demand duration of each subject and the available resource duration are matched to obtain a set of candidate scheduling pairs. This ensures that each training task can be reasonably scheduled according to resource time, reducing scheduling gaps and fragmentation and improving resource utilization.

[0074] Perform resource conflict detection on candidate scheduling pairs to determine whether different subjects correspond to the same resource identifier and have overlapping time periods. This generates a conflict filter set, which pre-checks resource overlap before scheduling decisions are made to avoid duplicate resource use in the same time period. This prevents scheduling failures or interruptions caused by resource conflicts during actual training execution, and improves the reliability of scheduling plan execution.

[0075] Based on the conflict filter set, conflicting pairs are removed, and the remaining candidate scheduling pairs are used as the valid scheduling set. They are then sorted according to position dependency information to obtain a sorted data sequence. This ensures that the execution of training courses meets the dependency order requirements of the organizational management structure, avoids overlapping and confusing training links, improves the systematicity, scientific nature, and dependency controllability of the training process, and enhances the internal consistency of the scheduling logic.

[0076] Based on the sorted data sequence, we extract the candidate resource set corresponding to each subject and calculate the score value of each candidate resource to obtain the resource score set. This makes the resource recommendation process more scientific and explainable, avoids "blind selection" of resources or over-reliance on static priorities, and makes the best choice based on the score value in the scenario where multiple subjects compete for resources, thereby improving the task fit of the scheduling results and the overall resource utilization efficiency of the system.

[0077] Based on the resource scoring set, the resource identifier with the largest score in each subject is selected as the recommended resource, and its time period label and position number in the effective scheduling set are extracted to obtain the scheduling plan data, completing the closed-loop output of the entire scheduling chain, providing a data foundation for the informatization construction of the training management system, and ensuring that the output scheduling results have high adaptability, high execution feasibility and traceability.

[0078] In a preferred embodiment of the present invention, a position dataset and a resource dataset are obtained, and subject association relationships and position mapping relationships are determined based on them, subject constraint information and position dependency information are extracted, and a constraint dataset is obtained, including:

[0079] Obtain the position numbers of all positions, the subject sets corresponding to each position, and the training required duration for each subject to obtain the position data set; obtain the resource numbers of all resources, the resource availability duration, the subject types supported by the resources, and the position identifiers to which the resources belong to obtain the resource data set;

[0080] Based on the position data set, determine the occurrence of each subject in different positions, extract the continuous training relationship and non-parallel constraints between subjects, and obtain subject association relationship data;

[0081] According to the position data set, determine the matching structure between the position identifier and the subject set, and obtain the position mapping relationship data;

[0082] According to the subject association relationship data, the order relationship, mutually exclusive execution flag and training stage overlap relationship between each subject and other subjects are extracted to obtain the subject constraint information data;

[0083] Based on the position mapping relationship data and the position identifiers of the resources in the resource data, the sharing and preemption of resources between different positions are analyzed to obtain position dependency information data;

[0084] The subject constraint information data and the position dependency information data are merged to obtain a rule information set, and the subject identification, position number, time condition and execution relationship label are extracted to obtain a constraint data set.

[0085] In an embodiment of the present invention, the position numbers of all positions, the subject sets corresponding to each position, and the training requirement duration of each subject are obtained to obtain a position data set, which provides a direct mapping relationship between positions and subjects, helps to quickly locate position training requirements, and thus supports the generation of scheduling targets; the resource numbers, resource availability duration, resource support subject types, and position identifiers of all resources are obtained to obtain a resource data set, which establishes training support constraints between subjects and resources to avoid resource mismatches or unavailable training resources; based on the position data set, the occurrence of each subject in different positions is determined, and the continuous training relationship and non-parallel constraints between subjects are extracted to obtain subject association relationship data. By clearly defining the execution constraints of the subjects, invalid or conflicting scheduling plans can be avoided in the subsequent time matching and sorting stages, the priority and dependency consistency of task execution can be guaranteed, and the controllability of the scheduling process can be improved; based on the position data set, the matching structure between the position identifier and the subject set is determined to obtain position mapping relationship data, which reveals the cross-position distribution of subjects, can determine whether training resources have reuse potential in multiple positions, and provide support for subsequent resource conflict detection and sorting scheduling.

[0086] Based on the subject association relationship data, the sequential relationship, mutually exclusive execution flags and overlapping relationship of training stages between each subject and other subjects are extracted to obtain subject constraint information data, and all abstract rules are explicitly modeled to enhance the decision-making accuracy of the scheduling algorithm and avoid scheduling failures or failure to achieve training goals due to hidden constraints; based on the position mapping relationship data and the position identification of the resources in the resource data, the sharing and preemption of resources between different positions in the resource dimension are analyzed to obtain position dependency information data, which can understand the resource relationship between different positions, give priority to meeting the needs of positions with high dependency intensity, reduce scheduling conflicts between positions, improve resource allocation fairness and the overall scheduling quality of the system; merge the subject constraint information data and the position dependency information data to obtain a rule information set, and extract the subject identification, position number, time condition and execution relationship label to obtain a constraint data set. The analysis results of multiple dimensions are integrated and abstracted into a unified scheduling control input, which improves the organization and automation of the entire scheduling system and supports resource scheduling in concurrent, multi-objective and high-complexity scenarios.

[0087] Among them, based on the position data set, the occurrence of each subject in different positions is determined, the continuous training relationship and non-parallel constraints between subjects are extracted, and the subject association relationship data is obtained, which specifically includes:

[0088] First, by traversing each position record in the position dataset, the frequency of each subject appearing in each position is counted, and a reverse mapping relationship between subject and position is established to identify whether multiple positions have the same set of subjects. The system then analyzes the order in which multiple subjects are combined within the same position. If subject A and subject B are found to follow a fixed order in most positions (e.g., subject A always comes before subject B), a "continuous training relationship" is recorded. Furthermore, if certain subjects are explicitly defined in the training rules as incompatible (e.g., due to resource conflicts or repeated training phases), or if they have never overlapped in historical tasks, the system identifies them as "non-parallelizable subjects" and labels them as mutually exclusive pairs. Furthermore, to improve the accuracy of the model, a priori or expert rules can be introduced. For example, based on the training syllabus, pairs of subjects with basic dependencies can be defined as "sequential execution groups," while subjects using the same key resources can be labeled "non-parallelizable groups." Finally, the system outputs all subject pairs with continuity or mutual exclusion characteristics as structured subject association data, which serves as the basis for subsequent subject scheduling order determination and mutual exclusion screening.

[0089] Among them, based on the subject association relationship data, the order relationship, mutually exclusive execution flag and training stage overlap relationship between each subject and other subjects are extracted to obtain the subject constraint information data, which specifically includes:

[0090] Based on the previously constructed subject association data, a subject dependency graph is established and the graph structure is semantically categorized and labeled, further refined into three core constraint information types: sequential relationships, mutually exclusive execution, and phase overlap relationships. For sequential relationships, the system analyzes the subject dependency path for directed "predecessor→successor" structures, automatically marking subject A as required to execute before subject B, and recording its priority level. For mutually exclusive execution flags, the system examines subject pairs with mutually exclusive flags. If the resources used by the two subjects physically overlap, there is a conflict between the training tasks in the same period, or the subject's definition restricts them from being executed simultaneously, the subject is explicitly marked as "non-simultaneously schedulable." For phase overlap relationships, the system learns from previous training records or uses a rule engine to identify subjects that, while not conflicting in terms of training content, may have overlapping time periods. For example, one subject only has theoretical lectures in the first two hours, while another subject has practical exercises in the last two hours. The two subjects do not overlap at the resource level and are therefore marked as "phase overlappable." The system outputs these three types of information in a unified form as structured constraint entries. Each entry includes subject ID, associated subject ID, constraint type, constraint direction (such as whether it is a one-way dependency) and time reference information, which ultimately constitute subject constraint information data, providing key logical conditions for subsequent steps such as resource time matching and scheduling sequence sorting.

[0091] Among them, according to the position mapping relationship data and the position identifiers of the resources in the resource data, the sharing and preemption of resources between different positions are analyzed to obtain the position dependency information data, which specifically includes:

[0092] First, based on the position mapping data, the system determines which positions have training needs for each subject. Then, combined with the resource dataset, the system extracts the position identifiers and supported subject sets of each resource, thereby determining the resource-position relationships. By comparing the supported subjects of a resource with its position identifiers, as well as the distribution of subjects across positions, the system identifies which positions have overlapping needs for a resource. If the system discovers that two or more positions are scheduling the same resource, and there is a potential risk of overlapping scheduling periods, the system determines that there is a resource preemption issue between positions and establishes resource contention edges between these positions. Resources serving only a single position or serving non-overlapping subjects are marked as isolated resources. Furthermore, the system constructs position priority indicators based on position training level, position functional priority, or training task urgency. These indicators are mapped as weights to the edges of the resource contention graph, generating a position dependency graph. In this dependency graph, nodes represent positions, edges represent resource sharing relationships, edge directionality represents resource allocation priority, and edge weights serve as ranking indicators. Ultimately, the system structures and outputs the dependency paths formed between all positions due to resource intersections as position dependency information data, which includes fields such as position pairs, resource intersections, contention probabilities, and priority labels. This information data will be used in the scheduling and sorting stage to determine the order of tasks and the urgency of scheduling.

[0093] In a preferred embodiment of the present invention, based on the constraint data set, the training requirement duration of each subject and the resource available duration are time-matched to obtain a candidate scheduling pair set, including:

[0094] Obtain the subject ID, total training duration, daily training duration requirement, and whether segmented training is allowed to obtain subject duration requirement data; obtain the resource ID, continuous available time period, and resource support subject ID set to obtain resource time segment data;

[0095] According to the subject constraint information in the constraint data set, other subjects that have a mutually exclusive relationship with the target subject during the arrangement period are screened out to obtain a set of matching subjects;

[0096] According to the resource time segment data, all resource identifiers required for each subject identification are extracted, and combined with the continuous available time period of each resource, the resource schedulable window set is obtained;

[0097] For each subject in the subject duration demand data, combined with the flag of whether segmented training is allowed, a set of time periods that can cover the total training duration are selected from the resource schedulable window set to obtain a time matching item;

[0098] For each time matching item, determine the subject ID, matching resource ID, planned start time, planned end time, and daily training duration value, generate a scheduling pair candidate record, and summarize all scheduling pair candidate records to obtain a candidate scheduling pair set.

[0099] In an embodiment of the present invention, the subject identification, total training duration, daily training duration requirement, and whether segmented training is allowed are obtained to obtain subject duration requirement data, thereby achieving structured and standardized processing of subject training requirements and providing an input basis for subsequent time matching; resource identification, continuous available time period, and resource support subject identification set are obtained to obtain resource time segment data, and the time dimension information of resource use is expressed in a standardized manner, laying a data foundation for subsequent time window-based matching processing; according to the subject constraint information in the constraint data set, other subjects that have a mutually exclusive relationship with the target subject during the target subject arrangement are screened out to obtain a matchable subject set, and by eliminating subjects that do not meet the scheduling conditions based on the execution constraints, the correctness of the scheduling logic is ensured, and a legal input set is provided for subsequent candidate resource screening and time window calculation, thereby avoiding logical conflicts between subjects from the source and improving scheduling feasibility; according to the resource time segment data, all resource identifications required for each subject identification are extracted, and combined with the continuous available time period of each resource, a resource schedulable window set is obtained, and the resources are matched. The source is transformed from an abstract entity into a measurable time slice, realizing the transformation of the scheduling problem from "resource allocation" to "time block filling", providing an operational search space for subsequent specific matching, and enhancing scheduling flexibility and resource utilization efficiency; for each subject in the subject duration demand data, combined with the flag of whether segmented training is allowed, a set of continuous time periods that can cover the total training duration are selected from the resource schedulable window in turn to obtain time matching items, achieving a precise match between subject requirements and resource capabilities, and providing multiple scheduling options while ensuring logical legitimacy, leaving sufficient optimization space for the system's subsequent scoring and conflict judgment stages; for each time matching item, the subject identifier, matching resource identifier, proposed start time, proposed end time, and daily training duration value are determined, and a scheduling pair candidate record is generated. All scheduling pair candidate records are summarized to obtain a candidate scheduling pair set. By centrally managing all scheduling candidate plans, the system can provide a high-quality input data source for subsequent operations, improving the consistency of the overall scheduling logic and the modularity of system execution.

[0100] Among them, according to the subject constraint information in the constraint data set, other subjects that have a mutually exclusive relationship with the target subject during the arrangement period are screened out to obtain a set of matching subjects, specifically including:

[0101] First, an estimated scheduling time range is established for each target subject. Based on the total training duration and daily training duration requirements for the subject, the minimum required time span is estimated. This is combined with the windows in resource time segments that may support the subject to form an initial set of candidate scheduling time ranges. The system then extracts constraint information related to the subject from the constraint dataset, including mutually exclusive execution flags, sequential dependency labels, and phase overlap indicators. The system parses these labels to identify other sets of subjects that cannot be scheduled concurrently with the subject. For example, if two subjects have mutually exclusive flags, or their training phases have non-overlapping properties, or are explicitly marked as requiring sequential execution (e.g., subject B must follow subject A), the subject is temporarily excluded from the candidate space for parallel scheduling. The system then analyzes and eliminates all subjects with such constraints one by one, forming a set of subjects that can theoretically coexist within the pre-scheduled time range of the target subject Ti, called the matchable subject set. This set serves as input for the next step of resource scheduling matching to ensure that logical conflicts between subjects are avoided in subsequent steps. During the entire process, each elimination has a traceable constraint basis, and the final collection has logical closure and scheduling security.

[0102] For each subject in the subject duration requirement data, combined with the flag of whether segmented training is allowed, a set of continuous time periods that can cover the total training duration are selected from the resource schedulable window to obtain time matching items, including:

[0103] The system traverses each subject T in the subject duration requirement data, extracting its total training duration, minimum daily training duration, and whether segmented training is permitted. It then centrally selects the set of resources R associated with it within the resource schedulability window. For each resource R, the system retrieves a list of its time segments, each with a start and end time. The system first determines whether the length of the time segment is greater than or equal to the minimum daily training duration to ensure that the basic daily training requirements are met. It then makes a policy decision based on whether segmented training is permitted.

[0104] If the flag for whether segmented training is allowed is no (i.e., segmentation is not allowed), the system only retains those time segments whose single time period length is greater than or equal to the minimum daily training time as available candidates; for these time periods, the system directly generates scheduling candidates, which include subjects, resources, start time and end time, etc., indicating that the training goal of the subject can be completed in one go by the resource within the time period.

[0105] If the flag for "allow segmented training" is yes (i.e., segmentation is allowed), the system combines all available time slots for the resource. Using a greedy strategy or dynamic programming, it selects time slots starting with a forward traversal, accumulating the training duration until the accumulated time reaches or exceeds the total training duration. During the combination process, it is necessary to ensure that the length of each selected time slot is at least equal to the total training duration to avoid excessive granularity. After the combination is complete, T is bound to the combined time slot sequence and R to form a set of scheduling matches. Each time slot combination is encapsulated as a scheduling pair candidate record, which includes fields such as resource identifier, subject identifier, start and end time sequence, and daily training schedule. This creates multiple scheduling matching paths for subsequent scoring and screening.

[0106] In a preferred embodiment of the present invention, resource conflict detection is performed on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and a conflict filter set is obtained, including:

[0107] Group all candidate scheduling pair records in the candidate scheduling pair set by resources, cluster all scheduling pairs according to resource identifiers, and obtain a resource pair set;

[0108] Analyze any two scheduling pairs in the resource pair set to determine whether their corresponding start and end times have overlapping intervals. If the result is yes, mark them as time-overlapping pairs.

[0109] Based on the time overlapping pairs, further determine whether their corresponding subject identifiers are different. If the result is yes, mark the pairs as conflicting pairs and obtain a conflicting identifier list;

[0110] Each scheduling pair in the conflict identification list is removed from the candidate scheduling pair set, and the remaining part of the candidate scheduling pair set is used as the resource conflict-free pair set;

[0111] The conflict identification list and the resource non-conflict pair group set are merged and marked as conflict scheduling subset and non-conflicting scheduling subset respectively to obtain the conflict filter set.

[0112] In an embodiment of the present invention, all candidate records of scheduling pair groups in the candidate scheduling pair group set are grouped by resources, and all scheduling pair groups are clustered according to resource identifiers to obtain a resource pair group set. Clustering is performed through resource identifiers to localize potential time conflict problems in advance, thereby reducing the computational burden of irrelevant data. Any two scheduling pair groups in the resource pair group set are analyzed to determine whether there is an overlapping interval between their corresponding start time and end time. If the result is yes, they are marked as time overlapping pairs, which can accurately identify conflicting situations in which the same resource is concurrently requested in multiple subjects, thereby avoiding the common problem of duplicate resource allocation in traditional training systems. Based on the time overlapping pairs, it is further determined whether the corresponding subject identifiers are different. If the result is yes, the pair is marked as a conflicting pair group, and a conflict identifier list is obtained to confirm. The system only marks the scheduling pairs that truly constitute resource competition, which enhances the intelligence and accuracy of scheduling judgment and avoids misjudgment that leads to reduced resource utilization; each scheduling pair in the conflict identification list is removed from the candidate scheduling pair set, and the remaining part of the candidate scheduling pair set is used as the resource conflict-free pair set to ensure that the scheduling results are executable and consistent in actual deployment. Eliminating conflict pairs can effectively reduce the probability of tasks being interrupted or reallocated in the subsequent execution stage, ensuring the complete implementation of training; the conflict identification list and the resource conflict-free pair set are merged and marked as conflict scheduling subsets and non-conflicting scheduling subsets respectively to obtain a conflict filter set, and a structured conflict filter set is constructed. The system realizes data isolation between the conflict detection stage and the sorting and scoring stage to avoid erroneous or invalid data interfering with the subsequent scheduling logic.

[0113] Among them, all the candidate scheduling pair group records in the candidate scheduling pair group set are grouped by resources, and all scheduling pairs are clustered according to resource identifiers to obtain a resource pair group set, which specifically includes:

[0114] First, the system receives a set of candidate scheduling pairs generated in the previous stage. Each candidate scheduling pair record contains data fields such as the subject ID, matching resource ID, planned start time, planned end time, and daily training duration. To facilitate conflict detection, the system performs structured classification on all scheduling pair records in this set, grouping them using resource ID as the clustering dimension.

[0115] The system traverses each scheduling pair record from the candidate scheduling pair set in turn and reads the resource identification field it contains. To ensure processing efficiency, the system establishes a hash mapping table with the resource identifier as the primary key during the initialization phase. Whenever a new scheduling pair is read, it determines whether its resource identifier already exists in the mapping table as the primary key. If the resource identifier does not yet appear, the system will create a corresponding primary key in the mapping table and use an empty list as the initialization value; if the resource identifier already exists, the current scheduling pair record will be directly appended to the scheduling pair list corresponding to the resource identifier. In this way, the system can complete the resource clustering process of all scheduling pairs in a single linear scan, thereby efficiently constructing a structural mapping relationship of "resource identifier-scheduling pair list".

[0116] After traversing and categorizing all scheduling pair records, the system extracts the scheduling pair list corresponding to each resource identifier in the hash map, forming a resource pair set. Each element in this set is a resource pair, containing all the time segments in which the resource was scheduled within the candidate scheduling pair set. This resource pair set enables the system to perform subsequent resource-based concurrency conflict detection and time overlap detection.

[0117] In a preferred embodiment of the present invention, conflicting pairs are removed from the conflict filter set, and the remaining candidate scheduling pairs are used as the valid scheduling set. The remaining candidate scheduling pairs are sorted according to the position dependency information to obtain a sorted data sequence, including:

[0118] According to the conflict filter set, the scheduling pairs marked as conflicting scheduling subsets are removed from the candidate scheduling pair set, and the remaining scheduling pairs are retained to obtain the valid scheduling set;

[0119] According to the valid scheduling set, the subject identification and position number information in each scheduling pair is extracted to construct the subject position comparison data;

[0120] According to the position dependency information, determine the priority relationship set between different positions and form a position dependency graph;

[0121] Map each position number in the subject position comparison data to the position dependency graph, establish the transitive dependency relationship between subjects, and obtain the subject scheduling directed graph;

[0122] Sort the subject scheduling directed graph, determine the execution order list of subjects according to the direction of the dependency edges in the graph, and obtain the sorted path sequence;

[0123] Map the subject identifiers in the sorting path sequence back to the corresponding scheduling pairs in the valid scheduling set, arrange them in order, and obtain a sorted data sequence.

[0124] In an embodiment of the present invention, according to the conflict filter set, the scheduling pairs marked as conflict scheduling subsets are removed from the candidate scheduling pair set, and the remaining scheduling pairs are retained to obtain a valid scheduling set, ensuring that each resource in the generated valid scheduling set has a clear time ownership and will not be occupied by multiple subjects in parallel, avoiding the "resource preemption" and "subject mutual exclusion violation" problems that are prone to occur in traditional static scheduling; according to the valid scheduling set, the subject identification and position number information in each scheduling pair is extracted, and the subject position comparison data is constructed, and a mapping path between the training task and the position is established, so that the logical relationship between the positions can be transmitted to the subject execution order level, providing a logical basis for sorting; according to the position dependency information, the priority relationship set between different positions is determined to form a position dependency graph. This operation improves the structural integrity and logical consistency of the training scheduling, and avoids disordered subject execution or broken training process due to the failure to transmit position dependency; the subject Each position number in the position comparison data is mapped to the position dependency graph, and the transitive dependency relationship between subjects is established to obtain a subject scheduling directed graph, which intuitively reflects the scheduling dependency path between subjects and can capture indirect dependency relationships, effectively preventing "circular dependency" or "logical deadlock" in the training process, and enhancing the system's adaptability under complex training plans; the subject scheduling directed graph is sorted, and the execution order list of subjects is determined according to the direction of the dependency edge in the graph to obtain a sorted path sequence, so that the subject scheduling results strictly follow the position dependency order, significantly improving the clarity and execution correctness of the training scheduling logic, and avoiding training effect deviations caused by advance or delay of subjects; the subject identifiers in the sorted path sequence are mapped back to the corresponding scheduling pairs in the effective scheduling set, and arranged in sequence to obtain a sorted data sequence, combining the subject execution order at the logical level with the scheduling records at the data level, realizing a closed-loop mapping from position dependency logic to resource allocation execution.

[0125] Among them, the position numbers in the subject position comparison data are mapped to the position dependency graph, and the transitive dependency relationship between subjects is established to obtain the subject scheduling directed graph, which specifically includes:

[0126] First, the system extracts the subject identifier and position number information contained in each schedule pair from the valid schedule set, constructing a subject-position mapping data table. Each record in this table represents the position number P corresponding to a subject M. The system then combines this data with a pre-generated position dependency graph, a directed graph structure in which each edge represents the execution order between predecessor and successor positions. For example, if position P1 must complete its training task before position P2, then there is a directed edge from P1 to P2 in the graph. The system traverses the subject-position mapping data and analyzes each pair of subjects (M1, M2). If M1 belongs to position P1 and M2 to position P2, and there is a directed edge from P1 to P2 in the position dependency graph, a directed edge from M1 to M2 is added at the subject level. The system repeats this process until all possible subject pairs have been analyzed, ultimately forming a directed subject scheduling graph. Each node in the graph structure corresponds to a subject identifier, and each edge indicates that the subject must be executed before another subject, forming a subject scheduling dependency network based on position relationships.

[0127] Among them, the subject scheduling directed graph is sorted, and the execution order list of the subjects is determined according to the direction of the dependency edges in the graph to obtain the sorting path sequence, which specifically includes:

[0128] After the subject scheduling directed graph is constructed, the system performs a topological sorting operation on it to determine the execution order of the subjects. Topological sorting starts with nodes with "in-degree 0", that is, subjects that currently have no predecessor dependencies. The system first scans all subject nodes in the graph, identifies the set of subjects with in-degree 0, and adds them to the initialized execution sequence list. Then, these subject nodes and all their output edges are removed from the graph, and the in-degree of the remaining nodes is recalculated. In the new graph, subject nodes with in-degree 0 are searched again, and the above process is repeated until all subject nodes are output to the execution sequence and the sorting is completed. If no node is found to have an in-degree of 0 during the sorting process, it means that there is a circular dependency relationship in the graph. The system will trigger the exception handling mechanism, prompting a dependency conflict and terminating the scheduling. After successfully completing the topological sorting, the sorting path sequence output by the system is the global execution order of the subjects, ensuring that all dependencies are met.

[0129] The subject identifiers in the sorting path sequence are mapped back to the corresponding scheduling pairs in the valid scheduling set, and arranged in sequence to obtain a sorted data sequence, which specifically includes:

[0130] After the system obtains the subject sorting path sequence, it reads each subject identifier in turn and performs a matching search in the valid scheduling set based on the identifier. Each subject corresponds to one or more scheduling pair candidate records in the valid scheduling set. The system gives priority to scheduling pairs that meet the following conditions based on the sorting order: its resource time period arrangement satisfies the sorting dependency, that is, the time arrangement is later than the scheduling time of all predecessor subjects. To this end, the system uses the training end time of the sorted subject as the time boundary to filter and match the scheduling time of subsequent subjects to ensure that there is no reverse order conflict in time. When there are multiple scheduling pairs that meet the conditions, the system gives priority to the one with the highest score based on the score value in the resource scoring set. After the matching is completed, the system adds the scheduling pair record to the sorted data sequence. This process continues until all subject identifiers in the sorting path sequence are mapped to their best scheduling pairs in the valid scheduling set.

[0131] In a preferred embodiment of the present invention, a candidate resource set corresponding to each subject is extracted based on the sorted data sequence, and a score value of each candidate resource is calculated to obtain a resource score set, including:

[0132] According to each subject in the sorted data sequence, all resources associated with it are extracted from the effective scheduling set to form a course candidate resource set and obtain subject resource mapping data;

[0133] Based on the subject resource mapping data, for each course candidate resource, the number of times it appears in the course candidate resource set is counted to obtain the resource appearance count, the proportion of its current selection by other subjects is calculated, the proportion of the current subject's usage time of the resource is extracted, and the resource usage frequency item is calculated based on the interaction of the three; for each resource, the set of position numbers it can serve is counted, the subject coverage ratio under each position is extracted, and based on the graph depth of the position node in the position dependency graph, the position priority weight is normalized to obtain the position coverage frequency item. Based on the subject coverage ratio and the position priority weight, the position coverage frequency item is calculated;

[0134] For each subject in the sorted data sequence, record its ranking in the sorting, calculate its priority coefficient based on the total number of subjects, and extract its completed training time and planned total training time to calculate the subject completion item;

[0135] The resource usage frequency item, position coverage frequency item and subject completion item are weighted and summed to calculate the score value of different resources for each subject. Each score value and its corresponding subject identifier and resource identifier are combined into a score record to obtain a resource score set.

[0136] In an embodiment of the present invention, according to each subject in the sorted data sequence, all resources associated with it are extracted from the effective scheduling set to form a course candidate resource set, and subject resource mapping data is obtained. All resource sets that may meet the current subject training needs in the scheduling are archived to provide an analysis basis for subsequent scoring; according to the subject resource mapping data, for each course candidate resource, the number of times it appears in the course candidate resource set is counted to obtain the number of resource appearances, the proportion of its current selection by other subjects is calculated, the proportion of the current subject's use time of the resource is extracted, and the resource usage frequency item is calculated based on the interaction of the three. By integrating the resource appearance frequency, usage overlap and adaptability to subject requirements, the system can evaluate whether the resource is scarce or exclusive, and then determine its scheduling priority, solving the resource problem. It solves the conflicts and duplications in resource allocation and improves the fairness and rationality of resource utilization; for each resource, counts the set of position numbers that it can serve, extracts the subject coverage ratio under each position, and based on the graph depth of the position node in the position dependency graph, normalizes the position priority weight, and calculates the position coverage frequency item according to the subject coverage ratio and position priority weight, which can measure the adaptability of resources to key positions in the system, thereby achieving the continuity of the position scheduling priority strategy and improving the overall resource allocation efficiency of the system; for each subject in the sorted data sequence, record its position in the sort, calculate its priority coefficient based on the total number of subjects, and extract its completed training time and planned total training time, calculate the subject completion item, and respond to the progress of training tasks in a timely manner. For subjects that have not been completed and are ranked high, higher weights will be given for resource scheduling to ensure that key training tasks can be prioritized in the scheduling plan, thereby improving the system's satisfaction and timeliness with training node goals; the resource usage frequency item, position coverage frequency item and subject completion item are weighted and summed, and the score value of each subject for different resources is calculated. Each score value and its corresponding subject identifier and resource identifier are combined into a score record to obtain a resource score set, realizing a multi-dimensional quantitative evaluation of the matching degree between resources and subjects, and providing a quantitative decision-making basis for subsequent recommended scheduling.

[0137] The calculation formula of the score value is:

[0138] ,

[0139] in, For resources Relative to subjects The rating value of is the index of the resource, is the index of the subject, For resources The number of times it appears in the candidate resource set, For resources The proportion of subjects assigned, For subjects Using Resources The duration ratio, is the total number of positions, is the index of the position, For resources Available for positions The proportion of For positions The number of subjects required, For positions The priority weight, For subjects Hours of training completed, For subjects The total duration of the training program, For subjects The priority coefficient, is the coefficient.

[0140] in, is the weight coefficient, the sum of which is 1. The specific weight value setting is dynamically adjustable and can be optimized based on the management strategy of the training system, current task requirements, and resource supply and demand conditions to achieve the best matching of scheduling results and maximize the execution effect.

[0141] In the case of extremely limited training resources (such as a sudden increase in the number of trainees, limited to 1-2 sets of hardware resources, etc.), the frequency of resource use and the proportion of preemption should be given priority, and resources should be allocated to subjects with less competition and high exclusivity as much as possible. In this case, the proportion of the first item (resource frequency) in the formula should be increased, that is, the resource frequency should be increased. , to avoid the scheduling results being concentrated on high-conflict resources, position versatility and subject priority still retain reference significance, but the effect is weak, so Adjust downward accordingly. The values ​​are 0.6, 0.2, and 0.2 respectively; in the case where a training center supports simultaneous training for multiple different positions (such as management, R&D, operation and maintenance, etc.), the system needs to give priority to allocating resources to resources with high coverage of positions and suitable for key positions (large graph depth and high priority). Therefore, the weight item To significantly improve and ensure the rationality of inter-position scheduling. At the same time, reduce the impact of resource conflicts (i.e. reduce ), allowing a certain degree of resource duplication scheduling to improve concurrency capabilities, at this time The values ​​are 0.2, 0.6, and 0.2 respectively. When some subjects are in the final stage of the training cycle or key assessment tasks are about to come, the system needs to prioritize resources to subjects with low training completion but high priority (such as management skills refresher training). At this time, the third item of the formula (subject urgency) should be considered to improve The weight of the training task is to ensure that the scheduling results give priority to the allocation of resources for unfinished key subjects. Position coverage and resource conflicts are temporarily weakened in exchange for the improvement of the overall quality of training task completion. The values ​​are 0.2, 0.2, and 0.6 respectively.

[0142] in, For positions The priority weight is based on the depth of the graph structure of each position node in the position dependency graph. In the previous step, the system has constructed a position dependency directed graph based on the resource dependency and subject continuity between positions. Each node in the graph represents a position, indicating that a position depends on the resources or task execution order of another position. For each position node , the system calculates its graph depth through depth-first traversal , which is the maximum path length of the position node from the starting node in the dependency path. The greater the graph depth, the later the position is in the execution process, and it often needs to complete the training tasks of the predecessor position first. In order to standardize the depth of the position graph as a weight item in the scoring, the system calculates the maximum depth of all positions. Perform normalization to get the position The priority weight calculation formula is: This priority weight reflects the relative execution stage of the position in the global training process. A larger value indicates that the position is at a later stage and is usually more dependent on pre-training content. It is suitable as an indicator for resource priority scheduling.

[0143] in, For subjects The priority coefficient is used to describe the current subject The urgency index in the scheduling sorting is derived from the ranking of the subject in the sorting data sequence. The sorting data sequence is obtained after the scheduling directed graph of subjects has been constructed through the position dependency relationship in the previous claim, and a sorted list of all subjects is given in the order of the dependency path in the graph. Assuming that the total number of subjects is , current subject The ranking is (The ranking starts at 0, and the lower the number, the higher the priority). The system calculates its priority coefficient based on its position in the overall ranking using the following formula: The result of this formula is a real number between [0,1]. The closer it is to 1, the higher the current subject is and the higher its scheduling priority. Together with the subject's training completion rate, it constitutes the training task urgency item in the scoring formula. That is, unfinished subjects that are ranked higher will receive higher scores and be given priority in scheduling recommended resources.

[0144] In a preferred embodiment of the present invention, based on the resource scoring set, the resource identifier with the largest scoring value in each subject is selected as the recommended resource, and its time period label and position number in the valid scheduling set are extracted to obtain the scheduling plan data, including:

[0145] For each subject ID, select the resource ID with the highest score from the resource score set and determine it as the recommended resource for the subject;

[0146] Find the scheduling pair corresponding to the subject and its corresponding recommended resource in the valid scheduling set, and extract the time period label in the scheduling pair as the training start and end time;

[0147] Extract the corresponding position number from the scheduling pair group as the associated position identifier of the subject;

[0148] According to the subject identification, recommended resource identification, training start and end time and position number, the resource allocation detail records are obtained, and all resource allocation detail records are summarized to obtain the scheduling plan data.

[0149] In an embodiment of the present invention, for each subject identifier, the resource identifier with the highest score is screened from the resource scoring set and determined as the recommended resource for the subject, thereby achieving optimal decision-making for resource recommendation, avoiding blind selection, and improving the matching degree between the recommendation result and the actual training needs; the scheduling pair group corresponding to the subject and its corresponding recommended resource is searched in the effective scheduling set, and the time period label in the scheduling pair group is extracted as the training start and end time, ensuring the executability of the resource allocation time in the scheduling plan and directly linking it to the validity verification of the system's early conflict detection and time matching stage; the corresponding position number is extracted from the scheduling pair group as the associated position identifier of the subject, clarifying the responsible party and operating position of the subject training, and making the scheduling plan position-oriented; based on the subject identifier, recommended resource identifier, training start and end time, and position number, a resource allocation detail record is obtained, and all resource allocation detail records are summarized to obtain scheduling plan data, which is not only clearly readable, but also supports seamless connection between the system and other training platforms, has high consistency, accuracy and traceability, and can meet the needs of various usage scenarios such as multi-level resource scheduling, task management and data evaluation.

[0150] An embodiment of the present invention further provides a training resource intelligent recommendation and scheduling system, the system comprising:

[0151] The constraint module is used to obtain the position data set and resource data set, and determine the subject association relationship and position mapping relationship based on them, extract the subject constraint information and position dependency information, and obtain the constraint data set;

[0152] The matching module is used to match the training requirement duration of each subject with the available resource duration based on the constraint data set to obtain a set of candidate scheduling pairs;

[0153] A detection module is used to perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain a conflict filter set;

[0154] The sorting module is used to remove conflicting pairs from the conflict filter set, take the remaining candidate scheduling pairs as the valid scheduling set, and sort them according to the job dependency information to obtain a sorted data sequence;

[0155] The scoring module is used to extract the candidate resource set corresponding to each subject based on the sorted data sequence, and calculate the score value of each candidate resource to obtain a resource scoring set;

[0156] The scheduling module is used to select the resource identifier with the largest score value in each subject as the recommended resource based on the resource scoring set, and extract its time period label and position number in the valid scheduling set to obtain the scheduling plan data.

[0157] It should be noted that this system is a system corresponding to the above method, and all implementation methods in the above method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0158] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the above-described method. All implementations in the above-described method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0159] The embodiment of the present invention further provides a computer-readable storage medium storing instructions, which, when executed on a computer, causes the computer to execute the above-described method. All implementations in the above-described method embodiment are applicable to this embodiment and can achieve the same technical effects.

[0160] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A training resource intelligent recommendation and scheduling method, characterized in that: The method comprises: Obtain position data sets and resource data sets, determine subject association relationships and position mapping relationships based on them, extract subject constraint information and position dependency information, and obtain constraint data sets; According to the constraint data set, the training demand duration and resource availability duration of each subject are matched to obtain a set of candidate scheduling pairs; Perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain the conflict filter set; According to the conflict filter set, the conflicting pairs are removed, and the remaining candidate scheduling pairs are used as the valid scheduling set. They are sorted according to the job dependency information to obtain a sorted data sequence, including: According to the position dependency information, determine the priority relationship set between different positions and form a position dependency graph; According to the sorted data sequence, the candidate resource set corresponding to each subject is extracted, and the score value of each candidate resource is calculated to obtain the resource score set; According to the resource scoring set, the resource identifier with the largest score in each subject is selected as the recommended resource, and its time period label and position number in the valid scheduling set are extracted to obtain the scheduling plan data; According to the sorted data sequence, the candidate resource set corresponding to each subject is extracted, and the score value of each candidate resource is calculated to obtain the resource score set, including: According to each subject in the sorted data sequence, all resources associated with it are extracted from the effective scheduling set to form a course candidate resource set, and the subject resource mapping data is obtained; Based on the subject resource mapping data, for each course candidate resource, count the number of times it appears in the course candidate resource set to obtain the resource appearance count, calculate the proportion of resources currently selected by other subjects, extract the proportion of time the current subject uses the resource, and calculate the resource usage frequency item based on the interaction of the three; For each resource, we count the set of position numbers it can serve, extract the subject coverage ratio for each position, and normalize the position priority weight based on the graph depth of the position node in the position dependency graph. We then calculate the position coverage frequency item based on the subject coverage ratio and position priority weight. For each subject in the sorted data sequence, record its ranking in the sorting, calculate its priority coefficient based on the total number of subjects, and extract its completed training time and planned total training time to calculate the subject completion item; The resource usage frequency item, position coverage frequency item and subject completion item are weighted and summed to calculate the score value of different resources for each subject. Each score value and its corresponding subject identifier and resource identifier are combined into a score record to obtain a resource score set.

2. The method for intelligent recommendation and scheduling of training resources according to claim 1, characterized in that: Obtain the position data set and resource data set, and determine the subject association relationship and position mapping relationship based on them, extract the subject constraint information and position dependency information, and obtain the constraint data set, including: Obtain the position numbers of all positions, the subject sets corresponding to each position, and the training required duration for each subject to obtain the position data set; obtain the resource numbers of all resources, the resource availability duration, the subject types supported by the resources, and the position identifiers to which the resources belong to obtain the resource data set; Based on the position data set, determine the occurrence of each subject in different positions, extract the continuous training relationship and non-parallel constraints between subjects, and obtain the subject association relationship data; According to the position data set, determine the matching structure between the position identifier and the subject set, and obtain the position mapping relationship data; According to the subject association relationship data, the order relationship, mutually exclusive execution flag and training stage overlap relationship between each subject and other subjects are extracted to obtain the subject constraint information data; Based on the position mapping relationship data and the position identifiers of the resources in the resource data, the sharing and preemption of resources between different positions are analyzed to obtain position dependency information data; The subject constraint information data and the position dependency information data are merged to obtain a rule information set, and the subject identification, position number, time condition and execution relationship label are extracted to obtain a constraint data set.

3. The method for intelligent recommendation and scheduling of training resources according to claim 2, characterized in that: Based on the constraint data set, the training requirement duration and resource availability duration of each subject are matched to obtain a set of candidate scheduling pairs, including: Obtain the subject ID, total training duration, daily training duration requirement, and whether segmented training is allowed to obtain subject duration requirement data; obtain the resource ID, continuous available time period, and resource support subject ID set to obtain resource time segment data; According to the subject constraint information in the constraint data set, other subjects that have a mutually exclusive relationship with the target subject during the arrangement period are screened out to obtain a set of matching subjects; According to the resource time segment data, all resource identifiers required for each subject identification are extracted, and combined with the continuous available time period of each resource, the resource schedulable window set is obtained; For each subject in the subject duration demand data, combined with the flag of whether segmented training is allowed, a set of time periods that can cover the total training duration are selected from the resource schedulable window set to obtain a time matching item; For each time matching item, determine the subject ID, matching resource ID, planned start time, planned end time, and daily training duration value, generate a scheduling pair candidate record, and summarize all scheduling pair candidate records to obtain a candidate scheduling pair set.

4. The method for intelligent recommendation and scheduling of training resources according to claim 3, characterized in that: Perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain the conflict filter set, including: Group all candidate scheduling pair records in the candidate scheduling pair set by resources, cluster all scheduling pairs according to resource identifiers, and obtain a resource pair set; Analyze any two scheduling pairs in the resource pair set to determine whether their corresponding start and end times have overlapping intervals. If the result is yes, mark them as time-overlapping pairs. Based on the time overlapping pairs, further determine whether their corresponding subject identifiers are different. If the result is yes, mark the pairs as conflicting pairs and obtain a conflicting identifier list; Each scheduling pair in the conflict identification list is removed from the candidate scheduling pair set, and the remaining part of the candidate scheduling pair set is used as the resource conflict-free pair set; The conflict identification list and the resource non-conflict pair group set are merged and marked as conflict scheduling subset and non-conflicting scheduling subset respectively to obtain the conflict filter set.

5. The method for intelligent recommendation and scheduling of training resources according to claim 4, characterized in that: According to the conflict filter set, the conflicting pairs are removed, and the remaining candidate scheduling pairs are used as the valid scheduling set. They are sorted according to the job dependency information to obtain a sorted data sequence, including: According to the conflict filter set, the scheduling pairs marked as conflicting scheduling subsets are removed from the candidate scheduling pair set, and the remaining scheduling pairs are retained to obtain the valid scheduling set; According to the valid scheduling set, the subject identification and position number information in each scheduling pair is extracted to construct the subject position comparison data; Map each position number in the subject position comparison data to the position dependency graph, establish the transitive dependency relationship between subjects, and obtain the subject scheduling directed graph; Sort the subject scheduling directed graph, determine the execution order list of subjects according to the direction of the dependency edges in the graph, and obtain the sorted path sequence; Map the subject identifiers in the sorting path sequence back to the corresponding scheduling pairs in the valid scheduling set, arrange them in order, and obtain a sorted data sequence.

6. The method for intelligent recommendation and scheduling of training resources according to claim 5, characterized in that: Based on the resource scoring set, the resource identifier with the largest score in each subject is selected as the recommended resource, and its time period label and position number in the valid scheduling set are extracted to obtain the scheduling plan data, including: For each subject ID, select the resource ID with the highest score from the resource score set and determine it as the recommended resource for the subject; Find the scheduling pair corresponding to the subject and its corresponding recommended resource in the valid scheduling set, and extract the time period label in the scheduling pair as the training start and end time; Extract the corresponding position number from the scheduling pair group as the associated position identifier of the subject; According to the subject identification, recommended resource identification, training start and end time and position number, the resource allocation detail records are obtained, and all resource allocation detail records are summarized to obtain the scheduling plan data.

7. A training resource intelligent recommendation and scheduling system, characterized in that: The system is used to perform the method according to any one of claims 1 to 6, and the system comprises: The constraint module is used to obtain the position data set and resource data set, and determine the subject association relationship and position mapping relationship based on them, extract the subject constraint information and position dependency information, and obtain the constraint data set; The matching module is used to match the training requirement duration of each subject with the available resource duration based on the constraint data set to obtain a set of candidate scheduling pairs; A detection module is used to perform resource conflict detection on the candidate scheduling pair set to determine whether different subjects correspond to the same resource identifier and have overlapping time, and obtain a conflict filter set; The sorting module is used to remove conflicting pairs from the conflict filter set, take the remaining candidate scheduling pairs as the valid scheduling set, and sort them according to the job dependency information to obtain a sorted data sequence; The scoring module is used to extract the candidate resource set corresponding to each subject based on the sorted data sequence, and calculate the score value of each candidate resource to obtain a resource scoring set; The scheduling module is used to select the resource identifier with the largest score value in each subject as the recommended resource based on the resource scoring set, and extract its time period label and position number in the valid scheduling set to obtain the scheduling plan data.

8. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images