Task processing method, device and equipment in integrated product development, and storage medium

Abstract

The application relates to a task processing method, device and equipment in integrated product development and a storage medium. The method comprises the following steps: determining a target function type according to a product to be developed demand document; determining at least one development task according to a pre-constructed mapping relationship between the function type and the development task; determining a target development object according to a required skill type, an estimated working hour, a skill portrait and a work load, and taking at least one of skill matching degree maximization, load balancing degree maximization and task completion time minimization as an optimization target; generating a risk prediction result through a risk prediction model according to task characteristics and object characteristics; in the case that the risk prediction result indicates that there is a preset risk, determining and executing a risk exclusion measure according to historical risk records and the risk prediction result, and pushing the at least one development task to the target development object after the risk is excluded to obtain a development result. The product development efficiency and quality are improved.

Description

Technical Field

[0001] This application relates to the field of integrated product development technology, and in particular to a task processing method, apparatus, device and storage medium in integrated product development. Background Technology

[0002] Using Integrated Product Development (IPD) as a product development management system can accelerate product launch, reduce costs, and optimize resource allocation.

[0003] Traditional task implementation in integrated product development relies heavily on manual management, paper documents, email communication, and offline review meetings. Furthermore, risk identification depends on human experience, leading to low product development efficiency, information asymmetry, and delayed decision-making. As enterprises grow and product complexity increases, traditional methods are no longer adequate to meet rapidly changing market demands. Therefore, there is an urgent need for a task processing method in integrated product development to improve both efficiency and quality. Summary of the Invention

[0004] Therefore, it is necessary to provide a task processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product that integrates product development to address the above-mentioned technical problems, thereby improving product development efficiency and quality.

[0005] Firstly, this application provides a task processing method in integrated product development, the method comprising:

[0006] Obtain the product requirements document and determine the target functional types of the product based on the product requirements document;

[0007] Based on the mapping relationship between function types and development tasks in the pre-built knowledge graph, at least one development task corresponding to the target function type is determined;

[0008] Based on the skill types required for at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development objects corresponding to each of the at least one development task are determined with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0009] For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, a risk prediction result is generated for the current development task through a pre-trained risk prediction model; if the risk prediction result for the current development task indicates the existence of a preset risk, risk mitigation measures for the current development task are determined based on historical risk records and corresponding risk prediction results.

[0010] Implement risk mitigation measures for each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0011] In one embodiment, the target functional types of the product to be developed are determined based on the product requirements document, including:

[0012] Extract the product name field and function name field from the product requirements document to be developed;

[0013] Link the product name field with the product entity in the knowledge graph to determine the target product entity;

[0014] Link the function name field with the function entity of the target product in the knowledge graph, determine the target function entity corresponding to the target product entity, and determine the target function type of the product to be developed based on the target function entity.

[0015] In one embodiment, the product name field and function name field are extracted from the product requirements document to be developed, including:

[0016] The product requirement document to be developed is preprocessed to obtain preprocessed text.

[0017] A pre-trained classification model is used to classify the pre-processed text to obtain classification labels;

[0018] Based on the category labels, the preprocessed text is converted into standardized description text, and the product name field and function name field are extracted from the standardized description text.

[0019] In one embodiment, the skill profile of any development object includes at least its proficiency in the skill types required for at least one development task and its remaining work capacity; based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development object corresponding to each of the at least one development task is determined with at least one of the following optimization objectives: maximizing skill matching, maximizing load balancing, and minimizing task completion time:

[0020] Determine the load balancing among multiple development objects based on their respective workloads.

[0021] For each development task, based on the skill types required for the current development task and the proficiency of each of the multiple development objects in those skill types, determine the skill matching degree between each of the multiple development objects and the current development task; based on the estimated working hours of the current development task and the remaining working capacity of each of the multiple development objects, determine the task completion time for each of the multiple development objects; based on each skill matching degree, each task completion time, and load balancing degree, with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as the optimization objective, determine the target development object corresponding to the current development task.

[0022] In one embodiment, based on skill matching degree, task completion time, and load balancing degree, with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as optimization objectives, the target development object corresponding to the current development task is determined, including:

[0023] Multiple candidate allocation schemes are randomly generated; each candidate allocation scheme includes at least one candidate development object corresponding to each development task.

[0024] For each candidate allocation scheme, the skill matching degree, task completion time and load balancing degree corresponding to the current candidate allocation scheme are weighted and the weighted result is used as the fitness of the current candidate allocation scheme.

[0025] Multiple reference allocation schemes are randomly selected from multiple candidate allocation schemes. The target allocation scheme with the highest fitness is determined from the multiple reference allocation schemes. Multiple target allocation schemes are obtained through multiple rounds of selection.

[0026] Multiple target allocation schemes are cross-recombined to obtain a recombined allocation scheme. Among the recombined allocation schemes, the target recombined allocation scheme with the highest fitness is determined. The target recombined allocation scheme includes at least one target development object corresponding to each development task.

[0027] In one embodiment, the method further includes:

[0028] Obtain the task execution progress of each target development object;

[0029] Based on the task execution progress, a pre-trained progress prediction model is used to determine the progress trend;

[0030] Based on the progress trend and planned schedule, determine the risk of task delay and generate early warning information that matches the risk of task delay.

[0031] Secondly, this application also provides a task processing device for integrated product development, comprising:

[0032] The acquisition module is used to acquire the product requirements document to be developed and, based on the product requirements document, determine the target functional types of the product to be developed.

[0033] The first determining module is used to determine at least one development task corresponding to the target function type based on the mapping relationship between function types and development tasks in the pre-built knowledge graph.

[0034] The second determination module is used to determine the target development object corresponding to each of the at least one development task based on the skill type required for each of the at least one development task, the estimated working hours, the skill profiles of the multiple development objects, and the workload, with the optimization objective being at least one of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0035] The third determination module is used to generate risk prediction results for each development task based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task through a pre-trained risk prediction model; if the risk prediction results for the current development task indicate the existence of a preset risk, the risk mitigation measures for the current development task are determined based on historical risk records and corresponding risk prediction results.

[0036] The execution module is used to execute the risk mitigation measures corresponding to each development task, and after the risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0038] Obtain the product requirements document and determine the target functional types of the product based on the product requirements document;

[0039] Based on the mapping relationship between function types and development tasks in the pre-built knowledge graph, at least one development task corresponding to the target function type is determined;

[0040] Based on the skill types required for at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development objects corresponding to each of the at least one development task are determined with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0041] For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, a risk prediction result is generated for the current development task through a pre-trained risk prediction model; if the risk prediction result for the current development task indicates the existence of a preset risk, risk mitigation measures for the current development task are determined based on historical risk records and corresponding risk prediction results.

[0042] Implement risk mitigation measures for each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0043] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0044] Obtain the product requirements document and determine the target functional types of the product based on the product requirements document;

[0045] Based on the mapping relationship between function types and development tasks in the pre-built knowledge graph, at least one development task corresponding to the target function type is determined;

[0046] Based on the skill types required for at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development objects corresponding to each of the at least one development task are determined with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0047] For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, a risk prediction result is generated for the current development task through a pre-trained risk prediction model; if the risk prediction result for the current development task indicates the existence of a preset risk, risk mitigation measures for the current development task are determined based on historical risk records and corresponding risk prediction results.

[0048] Implement risk mitigation measures for each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0049] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0050] Obtain the product requirements document and determine the target functional types of the product based on the product requirements document;

[0051] Based on the mapping relationship between function types and development tasks in the pre-built knowledge graph, at least one development task corresponding to the target function type is determined;

[0052] Based on the skill types required for at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development objects corresponding to each of the at least one development task are determined with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0053] For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, a risk prediction result is generated for the current development task through a pre-trained risk prediction model; if the risk prediction result for the current development task indicates the existence of a preset risk, risk mitigation measures for the current development task are determined based on historical risk records and corresponding risk prediction results.

[0054] Implement risk mitigation measures for each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0055] The task processing methods, devices, computer equipment, computer-readable storage media, and computer program products in the aforementioned integrated product development process, by acquiring the product requirements document to be developed, determining the target functional type of the product based on the requirements document, and identifying at least one development task corresponding to the target functional type based on the mapping relationship between functional types and development tasks in a pre-built knowledge graph, can quickly identify at least one development task from the product requirements document using the pre-built knowledge graph. Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, and with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time—the target development object corresponding to each of the at least one development task is determined. This multi-objective optimization method can accurately match suitable development objects for each development task, achieving intelligent allocation of development tasks and ensuring that task allocation conforms to the skill characteristics and workload of the development objects. This reduces workload, improves resource utilization and development collaboration efficiency. For each development task, based on the task characteristics and the target development object characteristics, a pre-trained risk prediction model generates a risk prediction result for the current development task. If the risk prediction result indicates the existence of a preset risk, risk mitigation measures are determined based on historical risk records and corresponding risk prediction results. Thus, before each development task is executed, risks are automatically identified by the risk prediction model based on task and object characteristics, and corresponding risk mitigation measures are automatically associated, effectively reducing the probability of risk in development tasks. The risk mitigation measures corresponding to each development task are executed, and after risk mitigation, at least one development task is pushed to the corresponding target development object to obtain the development results of the product to be developed by each target development object. This achieves full-process collaboration from requirements analysis, task decomposition, task allocation, risk intervention to task push, improving product development efficiency and quality. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is an application environment diagram for integrating task processing methods in product development in one embodiment.

[0058] Figure 2This is a flowchart illustrating a task processing method in integrated product development in one embodiment;

[0059] Figure 3 This is a structural block diagram of a task processing device integrated in product development in one embodiment;

[0060] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0063] The task processing method in integrated product development provided in this application embodiment can be applied to, for example... Figure 1The application environment shown is illustrated. Terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server. This embodiment uses the method applied to terminal 102 as an example for illustration. It is understood that this method can also be applied to servers, and can also be applied to systems including terminals and servers, and implemented through the interaction between the terminal and the server. Terminal 102 acquires the product requirement document to be developed. Based on the product requirement document, it determines the target function type of the product to be developed. Based on the mapping relationship between function types and development tasks in the pre-built knowledge graph, it determines at least one development task corresponding to the target function type. Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, it determines the target development object corresponding to each of the at least one development task with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time. For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, it generates a risk prediction result corresponding to the current development task through a pre-trained risk prediction model. If the risk prediction result corresponding to the current development task indicates the existence of a preset risk, it determines the risk mitigation measures corresponding to the current development task based on historical risk records and corresponding risk prediction results. It executes the risk mitigation measures corresponding to each development task and pushes at least one development task to the corresponding target development object after risk mitigation to obtain the development results of the product to be developed for each target development object. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, drones, low-altitude aircraft, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, and projection equipment. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0064] In one exemplary embodiment, such as Figure 2 As shown, a task processing method in integrated product development is provided, which is applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps 201 to 205. Wherein:

[0065] Step 201: Obtain the product requirements document to be developed, and determine the target function type of the product to be developed based on the product requirements document.

[0066] The product requirements document (PQV) is a descriptive document that specifies the functions, performance, and acceptance criteria of the product to be developed. It can include various document formats. The PQV can originate from original requirements documents, including user interview records, market research reports, customer service tickets, competitor analysis reports, complaints, and suggestions.

[0067] Optionally, the original requirement document is converted into structured requirement text and saved to the product requirement document to be developed. Specifically, a pre-trained classification model is used to determine the requirement type corresponding to the original requirement document, key requirement fields are extracted from the original requirement document, and data is converted according to a preset standardized format based on the requirement type and key requirement fields to obtain structured requirement text.

[0068] Optionally, based on a pre-built knowledge graph, the structured requirement texts in the product requirement document to be developed are linked to the corresponding functional entities to obtain the target functional entities. The target functional type is then determined based on the target functional entities. The target functional type indicates the functional type of the product to be developed.

[0069] Step 202: Based on the mapping relationship between function types and development tasks in the pre-built knowledge graph, determine at least one development task corresponding to the target function type.

[0070] The knowledge graph is a pre-built collection of domain knowledge. It includes multiple entities, their individual attributes, and the links between them. For example, the knowledge graph stores links between functional entities and task entities, indicating the mapping between corresponding functional types and development tasks. This mapping determines at least one development task corresponding to the target functional type. For example, development tasks might include music interface design, music data model design, music service development, and player interface development.

[0071] In some embodiments, a structured knowledge graph is constructed based on unstructured documents such as product design documents, technical architecture documents, requirement change records, project risk cases, test reports, maintenance work orders, and historical experience summaries. Integrated product development includes stages such as concept, planning, development, verification, release, and lifecycle. The knowledge graph provides entity relationship query and reasoning capabilities for each stage of integrated product development, enabling cross-stage knowledge reuse.

[0072] Optionally, entities in the knowledge graph include product entities, functional entities, technical component entities, requirement entities, task entities, risk case entities, work object entities, document entities, etc. Relationships between entities include implementation, inclusion, invocation, dependency, allocation, satisfaction, cause, response, association, reference, participation, expertise, collaboration, version, etc.

[0073] Optionally, the knowledge graph construction process includes: extracting information from unstructured documents using parsing tools, including titles, paragraphs, tables, charts, etc.; removing interfering information such as document formatting marks and page breaks, and performing sentence segmentation and word segmentation. An entity recognition model is invoked to identify entities in the text; the identified entities are linked to standard entities in the initial knowledge graph, and new entity nodes are created for new entities that do not match; a relation classification model is invoked to identify the relationship type between entity pairs; low-quality relations are removed based on a confidence threshold (e.g., 0.8) and rule filtering (e.g., circular dependency detection); and the extracted entities and relations are written into a graph database, with entities stored as nodes and relations stored as edges, thus obtaining the knowledge graph.

[0074] Step 203: Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of the multiple development objects, and the workload, determine the target development object corresponding to each of the at least one development task with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0075] Here, "Skill Type" refers to the professional competency tags required to complete the corresponding development task. Examples include skill certificates, job levels, and historical project tags. "Estimated Working Hours" refers to the estimated time required to complete the corresponding development task. "Development Object" refers to the object on which the development task is executed, which can include human objects, electronic objects, and mechanical objects. "Skill Profile" is a data-driven description of the skill types of the corresponding development object, which can include skill tags, proficiency scores, project experience, performance scores, and available workload. "Workload" refers to the total amount of work already undertaken by the corresponding development object.

[0076] Skill matching indicates the degree of match between the skill type of the development object and the development task. Load balancing indicates the overall load level among the development objects. Task completion time refers to the estimated time required for a given development object to complete its corresponding development task.

[0077] By identifying at least one target development object for each development task, and through multi-objective joint optimization, the goal is to at least maximize skill matching, maximize load balancing, and minimize task completion time. This enables intelligent allocation of development tasks, improving resource utilization and development collaboration efficiency.

[0078] Step 204: For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, generate the risk prediction result corresponding to the current development task through a pre-trained risk prediction model; if the risk prediction result corresponding to the current development task indicates the existence of a preset risk, determine the risk elimination measures corresponding to the current development task based on historical risk records and corresponding risk prediction results.

[0079] Before assigning development tasks to each target development object, risk prediction can be performed for each development task, and risk mitigation measures can be determined to reduce the probability of risk in the development task.

[0080] Task characteristics refer to data metrics that describe the inherent attributes of a given development task. Examples include task complexity, number of parallel tasks, external dependencies, and clarity of requirements.

[0081] Object characteristics refer to data metrics that describe the state, skills, etc., of each development object. Examples include team size, skill matching, load balancing, and object stability.

[0082] Risk prediction models are machine learning models trained using historical development project data and corresponding risk labels. For example, risk prediction models employ the XGBoost (eXtreme Gradient Boosting) classifier, which can handle high-dimensional sparse features, automatically select and combine features, support multi-classification tasks, and has strong generalization capabilities.

[0083] Optionally, the training samples for the risk prediction model include multi-dimensional feature vectors, such as task complexity, team size, skill matching, workload balance, dependency chain length, number of parallel tasks, personnel stability, technological novelty, frequency of requirement changes, number of interfaces, external dependencies, test coverage, defect density, schedule deviation rate, budget consumption rate, communication frequency, document completeness, proportion of automated testing, risk history, average experience of team members, critical path length, resource conflict index, technical debt index, requirement clarity, acceptance criterion clarity, supplier dependence, regulatory change risk, and market environment changes.

[0084] The risk prediction result is the output generated by the risk prediction model after the task characteristics and object characteristics are input. It indicates the type of risk present in the corresponding development task. For example, the risk prediction result includes risk type and risk level. Risk types include schedule risk, quality risk, resource risk, technical risk, and demand risk.

[0085] The preset risk is a pre-defined risk type. When the preset risk is found in the risk prediction results, risk mitigation measures are triggered to avoid overreacting to minor risks.

[0086] Historical risk records refer to risk cases that have occurred in the past. If the risk type indicated by the risk prediction results exists in the historical risk records, the risk mitigation measures in the historical risk records will be used as the risk mitigation measures corresponding to the current development task.

[0087] Optionally, the risk type indicated by the risk prediction results is matched with each risk case in the historical risk record to determine the target risk case, and the risk elimination measures corresponding to the target risk case are used as the risk elimination measures corresponding to the current development task.

[0088] Step 205: Implement risk mitigation measures for each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0089] The risk mitigation measures specify the steps to take to eliminate the risk. Optionally, an execution script corresponding to the risk mitigation measures can be generated and executed to achieve risk mitigation.

[0090] After the risk elimination results indicate that the risk has been eliminated, each development task is pushed to the corresponding target development object to instruct each development object to execute the development task and obtain the development result. The development result is the output of the development task, which may include movable code, test reports, etc.

[0091] In the task processing method described above for integrated product development, the following steps are taken: First, the product requirements document is obtained. Based on this document, the target functional type of the product is determined. Then, based on the mapping relationship between functional types and development tasks in a pre-built knowledge graph, at least one development task corresponding to the target functional type is identified. This allows for the rapid identification of at least one development task from the product requirements document using the pre-built knowledge graph. Next, based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development object corresponding to each of the at least one development task is determined with at least one of the following optimization objectives: maximizing skill matching, maximizing load balancing, and minimizing task completion time. This multi-objective optimization method accurately matches suitable development objects to each development task, achieving intelligent task allocation. This ensures that task allocation matches the skill characteristics and workload of the development objects, improving resource utilization and development efficiency. Collaborative efficiency: For each development task, based on the task characteristics and the target development object characteristics, a pre-trained risk prediction model generates a risk prediction result for the current development task. If the risk prediction result indicates the existence of a preset risk, risk mitigation measures are determined based on historical risk records and corresponding risk prediction results. Thus, before each development task is executed, risks are automatically identified by the risk prediction model based on task and object characteristics, and corresponding risk mitigation measures are automatically associated, effectively reducing the probability of risk in development tasks. The risk mitigation measures corresponding to each development task are executed, and after risk mitigation, at least one development task is pushed to the corresponding target development object to obtain the development results of the product to be developed by each target development object. This achieves full-process collaboration from requirements analysis, task decomposition, task allocation, risk intervention to task push, improving product development efficiency and quality.

[0092] In an exemplary embodiment, determining the target function type of the product to be developed based on the product requirements document includes: extracting a product name field and a function name field from the product requirements document; linking the product name field with product entities in a knowledge graph to determine the target product entity; and linking the function name field with the function entities in the knowledge graph that link the target product to determine the target function entity corresponding to the target product entity, so as to determine the target function type of the product to be developed based on the target function entity.

[0093] Specifically, fields related to the product name are extracted from the product requirements document to be developed as product name fields, and fields related to the function name are extracted as function name fields.

[0094] Identify the target product entity corresponding to the product name field in the knowledge graph. For example, calculate the similarity between each product entity in the knowledge graph and the product name field to obtain the first similarity calculation result. The product entity with the highest first similarity calculation result among all product entities is selected as the target product entity.

[0095] The knowledge graph stores the links between product entities and functional entities. Based on these links, candidate functional entities corresponding to the target product entity are determined. Each candidate functional entity is then compared to the function name field to calculate a second similarity score. The functional entity with the highest second similarity score among the candidate functional entities is selected as the target functional entity.

[0096] Each functional entity indicates a corresponding functional type, and the functional type corresponding to the target functional entity is taken as the target functional type of the product to be developed.

[0097] Optionally, the knowledge graph stores the link relationships between product entities, component entities, and functional entities. Based on these link relationships, candidate component entities corresponding to the target product entity are determined. The similarity of each candidate component entity's linked candidate functional entity to the functional name field is calculated to obtain a second similarity calculation result. The functional entity with the highest second similarity calculation result among all candidate functional entities is selected as the target functional entity.

[0098] In this embodiment, by sequentially linking the target product entity and the target function entity in the knowledge graph, the target function type corresponding to the target function entity is determined, thereby enabling the automatic identification of the target function type from the product requirement document to be developed, which is beneficial to improving the efficiency of integrated product development.

[0099] In an exemplary embodiment, extracting the product name field and function name field from the product requirements document to be developed includes: preprocessing the product requirements document to obtain preprocessed text; classifying the preprocessed text using a pre-trained classification model to obtain classification labels; converting the preprocessed text into standardized descriptive text based on the classification labels; and extracting the product name field and function name field from the standardized descriptive text.

[0100] The preprocessing of the product requirements document includes at least one of the following: text cleaning, analysis and part-of-speech tagging, named entity recognition, and syntactic dependency parsing. For example, text cleaning refers to using regular expressions to remove HTML tags, special characters, stop words, etc., converting unstructured text in the product requirements document into a standardized text format. Part-of-speech tagging refers to segmenting Chinese words using a word segmenter to provide a foundation for subsequent semantic analysis. Named entity recognition refers to using a named view recognition model to identify named entities such as product names, functional modules, technical terms, and organizational affiliations in the product requirements document. Syntactic dependency parsing refers to using a parser to perform Chinese syntactic dependency analysis, extracting syntactic structures such as subject-verb-object, attributive-head, and adverbial-head relationships to provide syntactic-level information for semantic understanding of the requirements.

[0101] A classification model is used to classify the preprocessed text into requirements, resulting in classification labels that indicate the type of requirement. The classification results include functional requirement types, non-functional requirement types, and constraint types. Functional requirement types describe the functions the system should perform; non-functional requirement types describe the system's quality attributes (such as performance, security, and availability); and constraint types describe the system's limitations (such as regulatory compliance, cost constraints, and technology selection constraints).

[0102] Optionally, the classification model uses a BERT (Bidirectional Encoder Representations from Transformers) pre-trained model as the encoder, and a fully connected classification layer is connected after its output layer to output a three-dimensional classification vector. A sigmoid activation function is used to support simultaneous multi-label matching. The training process of the classification model includes: each requirement sample is labeled with a three-dimensional label, for example, "The system should support user login" is labeled [1,0,0] (functional requirement type), "The system response time should be less than 2 seconds" is labeled [0,1,0] (non-functional requirement type), and "The system must meet the Level 3 Information Security Protection Requirements" is labeled [0,0,1] (constraint type). The labeled requirement samples are used to perform supervised fine-tuning of the BERT model to obtain the classification model. For example, the training parameters are set as follows: learning rate 2e-5, batch size 32, training epochs 10, using the AdamW optimizer, and the loss function is binary cross-entropy.

[0103] After preprocessing the text input classification model, a three-dimensional classification vector is automatically output as the requirement classification result. The classification label is determined by the requirement classification result and a pre-set classification threshold (such as 0.5). For example, when the output vector is [0.92, 0.73, 0.21], the classification label is determined to include functional requirement type and non-functional requirement type, but exclude constraint type.

[0104] Optionally, based on the demand type indicated by the category label, the preprocessed text is converted into standardized description text, and the product name field and function name field are extracted from the standardized description text.

[0105] Specifically, key requirement fields are extracted from the preprocessed text, including requirement description, requirement source, related products, and high-frequency keywords. For example, high-frequency keywords can be the Top N keywords extracted by the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm. Requirement identifiers can also be generated according to a preset format. For example, following the format REQ-[Product Abbreviation]-[Year]-[Serial Number], the generated requirement identifier would be REQ-SMS-2026-0042.

[0106] Optionally, a large language model can be used to standardize key requirement fields and requirement identifiers, remove colloquial expressions, supplement implicit business logic, and output standardized requirement text. For example, the prompt words of the large language model could be: Please standardize the following requirement description, remove colloquial expressions, supplement implicit business assumptions, and maintain technical accuracy. Original description: [Requirement text including key requirement fields and requirement identifiers].

[0107] Optionally, the standardized requirement text and category labels are assembled into standardized description text according to a preset standardized format. The preset standardized format can be JSON. For example, the standardized description text is as follows:

[0108] {"requirement_id": "REQ-SMS-2024-0042","description": "The system should support two authentication methods: username / password login and SMS verification code login","type": "Functional Requirements - User Authentication","source": "User Interview - Customer A","product": "Member Management System","keywords": ["Login", "Authentication", "Password", "SMS Verification"],"priority": null,"dependencies": [],"status": "To be sorted"}.

[0109] Optionally, each requirement identifier indicates a different requirement type. Multiple requirement types are prioritized to determine the execution order of development tasks, ensuring that higher-priority requirement types receive their corresponding development tasks first.

[0110] For example, determine the business value, urgency, and implementation cost of each type of requirement. Weight these three factors to obtain a weighted score. Use this weighted score to determine the priority of each requirement type. For instance, a requirement type with a weighted score greater than the first score has high priority. A requirement type with a weighted score no greater than the first score but greater than the second score has medium priority. A requirement type with a weighted score no greater than the second score has low priority.

[0111] In this embodiment, by preprocessing and classifying the product requirement document to be developed, and then converting it into standardized description text, the product name field and function name field are extracted from the standardized description text, which is beneficial to the standardization and accuracy of the extracted content.

[0112] In an exemplary embodiment, the skill profile of any development object includes at least its proficiency in the skill types required for at least one development task and its remaining workload. Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of the multiple development objects, and the workload, and with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time—the target development object corresponding to each of the at least one development task is determined. This includes: determining the load balancing among the multiple development objects based on their respective workloads; for each development task, determining the skill matching degree between the multiple development objects and the current development task based on the skill types required for the current development task and the proficiency of the multiple development objects in those skill types; determining the task completion time for the multiple development objects for the current development task based on the estimated working hours of the current development task and the remaining workload of the multiple development objects; and determining the target development object corresponding to the current development task based on each skill matching degree, each task completion time, and the load balancing degree, with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0113] Among them, proficiency indicates the degree to which the corresponding development object has mastered the corresponding development personnel. Remaining work capacity indicates the remaining workload that the corresponding development object can undertake.

[0114] For example, the maximum and minimum workloads are determined in the workloads of at least one development object, and the average workloads of at least one development object are averaged to obtain the average load. The difference between the maximum and minimum workloads is divided by the average load to obtain the relative fluctuation. The load balance is obtained by subtracting the relative fluctuation from 1.

[0115] Optionally, for each development object, the development object's proficiency in each skill type is multiplied by the importance coefficient of that skill type to obtain the weighted score of that development object for that skill type; the weighted scores of the development object for each skill type are summed to obtain the fit score; the importance coefficients of each skill type are summed to obtain the total coefficient; and the quotient between the fit score and the total coefficient is taken as the skill matching degree.

[0116] Optionally, the task completion time can be obtained by quoting the estimated working hours with the remaining working capacity of each development object. The remaining working capacity can be represented by a value between 0 and 1, with a larger value indicating more remaining working capacity.

[0117] The task allocation scheme that maximizes skill matching, maximizes load balancing, or minimizes task completion time can be used as the final task allocation scheme. Alternatively, the task allocation scheme that maximizes any two or three of these factors can be used as the final task allocation scheme.

[0118] For example, the objective function can be expressed as MaxF = w1 × skill matching degree + w2 × load balancing degree - w3 × task completion time. Here, MaxF represents the objective value, and w1, w2, and w3 represent the weights of skill matching degree, load balancing degree, and task completion time, respectively. Multi-objective optimization can be achieved by adjusting these weights.

[0119] Optionally, the multi-objective optimization process must also follow preset constraints. For example, preset constraints may include: each development task is assigned to a development object; the workload of the development object does not exceed the remaining working capacity; the assignment of development tasks satisfies dependencies: if development task i depends on development task h, then the start time of development task i is later than the completion time of development task h.

[0120] In this embodiment, the suitability of the development object to the development task is comprehensively evaluated by skill matching degree, complexity balance degree, and task completion time. A multi-objective optimization method is adopted to match personnel skills to improve work suitability, balance team load to avoid uneven work and rest, shorten task time, and efficiently and rationally allocate development resources.

[0121] In an exemplary embodiment, based on skill matching degree, task completion time, and load balancing degree, and with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as optimization objectives, the target development object corresponding to the current development task is determined, including: randomly generating multiple candidate allocation schemes; each candidate allocation scheme including at least one candidate development object corresponding to each development task; for each candidate allocation scheme, weighting the skill matching degree, task completion time, and load balancing degree corresponding to the current candidate allocation scheme to obtain the weighted result as the fitness of the current candidate allocation scheme; randomly selecting multiple reference allocation schemes from the multiple candidate allocation schemes, determining the target allocation scheme with the highest fitness from the multiple reference allocation schemes, and obtaining multiple target allocation schemes through multiple rounds of selection; cross-recombining the multiple target allocation schemes to obtain recombined allocation schemes, determining the target recombined allocation scheme with the highest fitness from the recombined allocation schemes, the target recombined allocation scheme including at least one target development object corresponding to each development task.

[0122] The candidate allocation schemes are formed through random combination, with each scheme indicating at least one candidate development object corresponding to a specific development task. The fitness of each candidate allocation scheme is obtained through a weighted solution. Fitness indicates the degree to which each development object in the candidate allocation scheme is adapted to the corresponding development task; a higher fitness indicates that the candidate allocation scheme is closer to the optimization objective.

[0123] The reference allocation scheme is an allocation scheme randomly selected from the candidate allocation schemes, and the target allocation scheme is the target allocation scheme with the highest fitness among the reference allocation schemes. Multiple target allocation schemes are obtained through multiple rounds of selection.

[0124] Cross-reorganization refers to the partial exchange of any two target allocation schemes. For example, in target allocation scheme A, development tasks 1-10 are assigned to development object A, and development tasks 11-20 are assigned to development object B, resulting in reorganization allocation scheme B. Reorganization allocation scheme B inherits the allocation scheme of development tasks 1-10 in target allocation scheme A, and development tasks 11-20 are re-randomized.

[0125] Optionally, the reorganization allocation scheme can be mutated with probability Pm, that is, the development object corresponding to any development task can be randomly adjusted.

[0126] Optionally, the E allocation schemes with the highest fitness among multiple target allocation schemes do not participate in crossover and mutation.

[0127] The allocation scheme with the highest fitness among the reorganization allocation scheme and the target allocation scheme is determined and used as the target reorganization allocation scheme.

[0128] In this embodiment, random generation and cross-recombination are used to avoid multi-objective optimization from getting stuck in local optima, thus achieving a reasonable allocation of development tasks.

[0129] In an exemplary embodiment, the method further includes: obtaining the task execution progress of each target development object; determining the progress trend by using a pre-trained progress prediction model based on the task execution progress; determining the task delay risk based on the progress trend and the planned progress, and generating early warning information matching the task delay risk.

[0130] The task execution progress indicates the degree of completion of the corresponding development task. The progress prediction model is a pre-trained machine learning model used to predict the progress trend of development tasks. Optionally, the training process of the progress prediction model includes: using historical task progress data sequences as training samples, including daily task completion percentage, actual working hours consumed, and personnel input time. The model output includes predictions of remaining task working hours, task delay probability, and overall project completion time. The progress prediction model is obtained through multiple rounds of iterative training. Optionally, a two-layer LSTM (Long Short-Term Memory) network can be used, with 128 hidden layer dimensions and a time window length of 14 days.

[0131] A progress trend is a prediction of the direction of a development task's progress. For example, a progress trend indicates the progress at each future time point.

[0132] The planned schedule is the baseline schedule specified before the development task begins, indicating the progress that should be achieved at each point in time. By comparing the progress trend with the planned schedule at each time point, if the progress trend at any time point lags behind the planned progress at that time point, it is determined that the current development task is at risk of delay.

[0133] Task delay risk is used to indicate whether a corresponding development task is at risk of delay, and can also indicate the risk level. For example, if the percentage lag in the progress trend compared to the planned progress is less than a first preset percentage, the delay risk is determined to be low risk. If the percentage lag in the progress trend compared to the planned progress is not less than the first preset percentage and is less than a second preset percentage, the delay risk is determined to be medium risk. If the percentage lag in the progress trend compared to the planned progress is not less than the second preset percentage, the delay risk is determined to be high risk.

[0134] For example, when the risk level of a task delay exceeds a preset level, an early warning message is generated. The early warning message is a structured alert notification, including the risk type, risk level, and suggested intervention actions. Different risk levels correspond to different suggested intervention actions. For instance, a high-risk intervention might suggest notifying the responsible party for the relevant development task, while a low-risk intervention might suggest updating the progress status.

[0135] In this embodiment, the progress trend is determined by a pre-trained progress prediction model. Combined with the planned progress, the risk of task delay is predicted and early warning information is generated, which helps to identify risks in a timely manner and improve the efficiency of development task execution.

[0136] To illustrate the task processing methods and effects in integrated product development in this solution in detail, a specific embodiment is provided below:

[0137] Obtain the product requirements document to be developed. Based on the requirements document, determine the target functional type of the product to be developed. Based on the mapping relationship between functional types and development tasks in the pre-built knowledge graph, determine at least one development task corresponding to the target functional type. Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, determine the target development object corresponding to each of the at least one development task with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time. For each development task, generate a risk prediction result for the current development task based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task through a pre-trained risk prediction model. If the risk prediction result for the current development task indicates the existence of a preset risk, determine the risk mitigation measures for the current development task based on historical risk records and corresponding risk prediction results. Execute the risk mitigation measures corresponding to each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed for each target development object.

[0138] In one embodiment, the target function type of the product to be developed is determined based on the product requirements document, including: extracting the product name field and the function name field from the product requirements document; linking the product name field with the product entity in the knowledge graph to determine the target product entity; and linking the function name field with the function entity in the knowledge graph that links to the target product to determine the target function entity corresponding to the target product entity, so as to determine the target function type of the product to be developed based on the target function entity.

[0139] In one embodiment, extracting the product name field and function name field from the product requirements document to be developed includes: preprocessing the product requirements document to obtain preprocessed text; classifying the preprocessed text using a pre-trained classification model to obtain classification labels; converting the preprocessed text into standardized description text based on the classification labels; and extracting the product name field and function name field from the standardized description text.

[0140] In one embodiment, the skill profile of any development object includes at least its proficiency in the skill types required for at least one development task and its remaining workload. Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of the multiple development objects, and the workload, and with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time—the target development object corresponding to each of the at least one development task is determined. This includes: determining the load balancing among the multiple development objects based on their respective workloads; for each development task, determining the skill matching between the multiple development objects and the current development task based on the skill types required for the current development task and the proficiency of the multiple development objects in those skill types; determining the task completion time for the multiple development objects for the current development task based on the estimated working hours of the current development task and the remaining workload of the multiple development objects; and determining the target development object corresponding to the current development task based on each skill matching, each task completion time, and the load balancing, with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0141] In one embodiment, based on skill matching degree, task completion time, and load balancing degree, and with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as optimization objectives, the target development object corresponding to the current development task is determined, including: randomly generating multiple candidate allocation schemes; each candidate allocation scheme including at least one candidate development object corresponding to each development task; for each candidate allocation scheme, weighting the skill matching degree, task completion time, and load balancing degree corresponding to the current candidate allocation scheme to obtain the weighted result as the fitness of the current candidate allocation scheme; randomly selecting multiple reference allocation schemes from the multiple candidate allocation schemes, determining the target allocation scheme with the highest fitness from the multiple reference allocation schemes, and obtaining multiple target allocation schemes through multiple rounds of selection; cross-recombining the multiple target allocation schemes to obtain recombined allocation schemes, and determining the target recombined allocation scheme with the highest fitness from the recombined allocation schemes, wherein the target recombined allocation scheme includes at least one target development object corresponding to each development task.

[0142] In one embodiment, the method further includes: obtaining the task execution progress of each target development object; determining the progress trend by using a pre-trained progress prediction model based on the task execution progress; determining the task delay risk based on the progress trend and the planned progress, and generating early warning information matching the task delay risk.

[0143] The task processing method described above in integrated product development obtains the product requirements document, determines the target function type of the product based on the requirements document, and identifies at least one development task corresponding to the target function type based on the mapping relationship between function types and development tasks in a pre-built knowledge graph. This allows for the rapid identification of at least one development task from the product requirements document using the pre-built knowledge graph. Based on the skill types required for each of the at least one development task, estimated working hours, skill profiles of multiple development objects, and workload, and with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time—the method identifies the target development object corresponding to each of the at least one development task. This multi-objective optimization method accurately matches suitable development objects to each development task, achieving intelligent task allocation, ensuring that task allocation matches the skill characteristics and workload of the development objects, and improving resource utilization and development efficiency. Collaborative efficiency: For each development task, based on the task characteristics and the target development object characteristics, a pre-trained risk prediction model generates a risk prediction result for the current development task. If the risk prediction result indicates the existence of a preset risk, risk mitigation measures are determined based on historical risk records and corresponding risk prediction results. Thus, before each development task is executed, risks are automatically identified by the risk prediction model based on task and object characteristics, and corresponding risk mitigation measures are automatically associated, effectively reducing the probability of risk in development tasks. The risk mitigation measures corresponding to each development task are executed, and after risk mitigation, at least one development task is pushed to the corresponding target development object to obtain the development results of the product to be developed by each target development object. This achieves full-process collaboration from requirements analysis, task decomposition, task allocation, risk intervention to task push, improving product development efficiency and quality.

[0144] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0145] Based on the same inventive concept, this application also provides a task processing apparatus for implementing the task processing method in integrated product development as described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method. Therefore, the specific limitations in one or more embodiments of the task processing apparatus for integrated product development provided below can be found in the limitations of the task processing method for integrated product development described above, and will not be repeated here.

[0146] In one exemplary embodiment, such as Figure 3 As shown, a task processing device 300 for integrated product development is provided, comprising: an acquisition module 301, a first determination module 302, a second determination module 303, a third determination module 304, and an execution module 305, wherein:

[0147] The acquisition module 301 is used to acquire the product requirements document to be developed and to determine the target function type of the product to be developed based on the product requirements document.

[0148] The first determining module 302 is used to determine at least one development task corresponding to the target function type based on the mapping relationship between function types and development tasks in the pre-built knowledge graph.

[0149] The second determining module 303 is used to determine the target development object corresponding to each of at least one development task based on the skill type required for each of at least one development task, the estimated working hours, the skill profiles of each of the multiple development objects, and the workload, with the optimization objective being at least one of maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0150] The third determining module 304 is used to generate a risk prediction result for each development task based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task through a pre-trained risk prediction model; and if the risk prediction result for the current development task indicates the existence of a preset risk, determine the risk elimination measures for the current development task based on historical risk records and corresponding risk prediction results.

[0151] The execution module 305 is used to execute the risk elimination measures corresponding to each development task, and after the risk elimination, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

[0152] The task processing device in the aforementioned integrated product development process acquires the product requirements document, determines the target function type of the product based on the requirements document, and identifies at least one development task corresponding to the target function type based on the mapping relationship between function types and development tasks in a pre-built knowledge graph. This allows for the rapid identification of at least one development task from the product requirements document using the pre-built knowledge graph. Based on the skill types required for each of the at least one development task, estimated working hours, skill profiles of multiple development objects, and workload, and with at least one of the following optimization objectives—maximizing skill matching, maximizing load balancing, and minimizing task completion time—the device identifies the target development object corresponding to each of the at least one development task. This multi-objective optimization method accurately matches suitable development objects to each development task, achieving intelligent task allocation, ensuring that task allocation matches the skill characteristics and workload of the development objects, and improving resource utilization and development efficiency. Collaborative efficiency: For each development task, based on the task characteristics and the target development object characteristics, a pre-trained risk prediction model generates a risk prediction result for the current development task. If the risk prediction result indicates the existence of a preset risk, risk mitigation measures are determined based on historical risk records and corresponding risk prediction results. Thus, before each development task is executed, risks are automatically identified by the risk prediction model based on task and object characteristics, and corresponding risk mitigation measures are automatically associated, effectively reducing the probability of risk in development tasks. The risk mitigation measures corresponding to each development task are executed, and after risk mitigation, at least one development task is pushed to the corresponding target development object to obtain the development results of the product to be developed by each target development object. This achieves full-process collaboration from requirements analysis, task decomposition, task allocation, risk intervention to task push, improving product development efficiency and quality.

[0153] In one embodiment, based on the product requirements document, the target function type of the product to be developed is determined. The acquisition module 301 is further configured to: extract the product name field and the function name field from the product requirements document; link the product name field with the product entity in the knowledge graph to determine the target product entity; link the function name field with the function entity in the knowledge graph that links the target product to determine the target function entity corresponding to the target product entity, so as to determine the target function type of the product to be developed based on the target function entity.

[0154] In one embodiment, the product name field and function name field are extracted from the product requirement document to be developed. The acquisition module 301 is further configured to: preprocess the product requirement document to be developed to obtain preprocessed text; classify the preprocessed text using a pre-trained classification model to obtain classification labels; convert the preprocessed text into standardized description text according to the classification labels, and extract the product name field and function name field from the standardized description text.

[0155] In one embodiment, the skill profile of any development object includes at least its proficiency in the skill types required for at least one development task and its remaining work capacity. Based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of the multiple development objects, and the workload, the target development object corresponding to each of the at least one development task is determined with at least one of the following optimization objectives: maximizing skill matching, maximizing load balancing, and minimizing task completion time. The second determining module 303 is further configured to: determine the load balancing among the multiple development objects based on their respective workloads; for each development task, determine the skill matching between the multiple development objects and the current development task based on the skill types required for the current development task and the proficiency of the multiple development objects in the skill types; determine the task completion time for the multiple development objects for the current development task based on the estimated working hours of the current development task and the remaining work capacity of the multiple development objects; and determine the target development object corresponding to the current development task based on each skill matching, each task completion time, and the load balancing, with at least one of the following optimization objectives: maximizing skill matching, maximizing load balancing, and minimizing task completion time.

[0156] In one embodiment, based on skill matching degree, task completion time, and load balancing degree, with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as optimization objectives, the target development object corresponding to the current development task is determined. The second determining module 303 is further configured to: randomly generate multiple candidate allocation schemes; each candidate allocation scheme includes at least one candidate development object corresponding to each development task; for each candidate allocation scheme, the skill matching degree, task completion time, and load balancing degree corresponding to the current candidate allocation scheme are weighted to obtain a weighted result as the fitness of the current candidate allocation scheme; multiple reference allocation schemes are randomly selected from the multiple candidate allocation schemes, and the target allocation scheme with the highest fitness is determined from the multiple reference allocation schemes, and multiple target allocation schemes are obtained through multiple rounds of selection; the multiple target allocation schemes are cross-recombined to obtain recombined allocation schemes, and the target recombined allocation scheme with the highest fitness is determined from the recombined allocation schemes, wherein the target recombined allocation scheme includes at least one target development object corresponding to each development task.

[0157] In one embodiment, the execution module 305 is further configured to: obtain the task execution progress of each target development object; determine the progress trend by using a pre-trained progress prediction model based on the task execution progress; determine the task delay risk based on the progress trend and the planned progress, and generate early warning information matching the task delay risk.

[0158] The modules in the task processing device of the aforementioned integrated product development can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0159] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a task processing method in integrated product development. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0160] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0161] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0162] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0163] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0164] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0165] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0166] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0167] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A task processing method in integrated product development, characterized in that, The method includes: Obtain the product requirements document to be developed, and determine the target functional types of the product to be developed based on the product requirements document; Based on the mapping relationship between function types and development tasks in the pre-constructed knowledge graph, at least one development task corresponding to the target function type is determined; Based on the skill types required for at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, the target development objects corresponding to each of the at least one development task are determined with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time. For each development task, based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task, a risk prediction result corresponding to the current development task is generated through a pre-trained risk prediction model; if the risk prediction result corresponding to the current development task indicates the existence of a preset risk, risk mitigation measures corresponding to the current development task are determined based on historical risk records and corresponding risk prediction results. Implement risk mitigation measures for each development task, and after risk mitigation, push at least one development task to the corresponding target development object to obtain the development results of each target development object for the product to be developed.

2. The method according to claim 1, characterized in that, The step of determining the target functional type of the product to be developed based on the product requirements document includes: Extract the product name field and function name field from the product requirements document to be developed; Link the product name field with the product entity in the knowledge graph to determine the target product entity; The function name field is linked to the function entity that links the target product in the knowledge graph to determine the target function entity corresponding to the target product entity, so as to determine the target function type of the product to be developed based on the target function entity.

3. The method according to claim 2, characterized in that, The step of extracting the product name field and function name field from the product requirements document to be developed includes: The product requirement document to be developed is preprocessed to obtain preprocessed text; The pre-trained classification model is used to classify the pre-processed text to obtain classification labels; Based on the classification labels, the preprocessed text is converted into standardized description text, and the product name field and function name field are extracted from the standardized description text.

4. The method according to claim 1, characterized in that, The skill profile of any development object includes at least its proficiency in the skill types required for at least one development task and its remaining work capacity; the determination of the target development object corresponding to each of the at least one development task, based on the skill types required for each of the at least one development task, the estimated working hours, the skill profiles of multiple development objects, and the workload, with at least one of the following optimization objectives: maximizing skill matching, maximizing load balancing, and minimizing task completion time, includes: Determine the load balancing among multiple development objects based on their respective workloads. For each development task, based on the skill types required for the current development task and the proficiency of each of the multiple development objects in those skill types, the skill matching degree between each of the multiple development objects and the current development task is determined; based on the estimated working hours of the current development task and the remaining working capacity of each of the multiple development objects, the task completion time for each of the multiple development objects is determined; based on each skill matching degree, each task completion time, and the load balancing degree, with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as the optimization objective, the target development object corresponding to the current development task is determined.

5. The method according to claim 4, characterized in that, The step of determining the target development object corresponding to the current development task based on the skill matching degree, the task completion time, and the load balancing degree, with at least one of maximizing skill matching degree, maximizing load balancing degree, and minimizing task completion time as the optimization objective, includes: Multiple candidate allocation schemes are randomly generated; each candidate allocation scheme includes at least one candidate development object corresponding to each development task. For each candidate allocation scheme, the skill matching degree, task completion time and load balancing degree corresponding to the current candidate allocation scheme are weighted, and the weighted result is used as the fitness of the current candidate allocation scheme. Multiple reference allocation schemes are randomly selected from multiple candidate allocation schemes. The target allocation scheme with the highest fitness is determined from the multiple reference allocation schemes. Multiple target allocation schemes are obtained through multiple rounds of selection. Multiple target allocation schemes are cross-recombined to obtain a recombined allocation scheme. Among the recombined allocation schemes, the target recombined allocation scheme with the highest fitness is determined. The target recombined allocation scheme includes at least one target development object corresponding to each development task.

6. The method according to claim 1, characterized in that, The method further includes: Obtain the task execution progress of each target development object; Based on the task execution progress, a pre-trained progress prediction model is used to determine the progress trend; Based on the progress trend and planned schedule, the risk of task delay is determined, and early warning information matching the risk of task delay is generated.

7. A task processing device for integrated product development, characterized in that, The device includes: The acquisition module is used to acquire the product requirements document to be developed and determine the target function type of the product to be developed based on the product requirements document. The first determining module is used to determine at least one development task corresponding to the target function type based on the mapping relationship between function types and development tasks in a pre-built knowledge graph. The second determination module is used to determine the target development object corresponding to each of the at least one development task based on the skill type required for each of the at least one development task, the estimated working hours, the skill profiles of the multiple development objects, and the workload, with the optimization objective of maximizing skill matching, maximizing load balancing, and minimizing task completion time. The third determination module is used to generate a risk prediction result for each development task based on the task characteristics of the current development task and the object characteristics of the target development object corresponding to the current development task through a pre-trained risk prediction model; and when the risk prediction result for the current development task indicates the existence of a preset risk, determine the risk elimination measures for the current development task based on historical risk records and corresponding risk prediction results. The execution module is used to execute the risk elimination measures corresponding to each development task, and after the risk elimination, push at least one development task to the corresponding target development object to obtain the development results of the product to be developed by each target development object.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images