Software development life cycle whole-process performance evaluation system and method

Abstract

The invention relates to the technical field of software development management, in particular to a software development life cycle whole-process effectiveness evaluation system and method.The method includes the steps that multiple pieces of effectiveness basic data of a target project are collected, each piece of effectiveness basic data is preprocessed to obtain preprocessed data, and then an evaluation model is established; the method comprises the steps of preprocessing data of a target project, inputting the preprocessed data into an evaluation model, calculating a demand level of the target project through the evaluation model in combination with the preprocessed data, continuously learning a corresponding relation between the demand level and a project period to obtain a trained evaluation model, and finally collecting real-time efficiency basic data of the real-time target project. The project data of the real-time target project is adjusted through the trained evaluation model in combination with the real-time efficiency basic data, and the demand level of the target project is calculated by combining the customer data and the demand priority of the sub-demands, so that the core demand of software development is accurately controlled, and the precision and efficiency of demand analysis of software development are improved.

Description

Technical Field

[0001] The present application relates to the technical field of software development management, and in particular to a system and method for evaluating the effectiveness of the entire software development life cycle. Background Art

[0002] Software development is the process of building a software system or the software components of a system according to user requirements. Software development is a systematic project that includes requirements capture, analysis, design, implementation, and testing. Software is generally implemented using a programming language. The Software Development Life Cycle (SDLC) covers the entire process from requirements analysis to software abandonment. Its core phases include requirements analysis, design, coding, testing, deployment, and maintenance. Its core goal is to ensure software quality and development efficiency, primarily following the principle of a structured, phased approach.

[0003] Chinese patent publication number CN114691090A discloses a method and system for evaluating the efficiency of demand flow based on the software development process. By constructing a demand flow scoring system, a demand flow efficiency evaluation system is created. Through this system, the overall statistical information of the demand flow process can be observed, and it can be gradually progressively observed from the macro to the details. Through this scoring system, the efficiency of the demand flow process can be measured, providing guidance for team demand changes and the next direction of action. However, the existing technology does not take into account the impact of customer data on the demand analysis of software development, which makes it difficult to accurately adjust the number of software development personnel and project cycles, and easily causes disharmony of development resources. Summary of the Invention

[0004] Based on this, it is necessary to propose a full-process performance evaluation system and method for the software development life cycle to address the defect that traditional IoT multi-node data is difficult to cope with peak data.

[0005] On the one hand, the present application provides a software development life cycle performance evaluation system, including a collection component and an evaluation component. The collection component collects basic performance data, and the evaluation component is communicated with the collection component. The evaluation component combines the basic performance data to perform performance evaluation on the software development life cycle.

[0006] Preferably, the performance basic data includes demand data, developer data, customer data and project data.

[0007] On the other hand, this application also provides a method for evaluating the effectiveness of the entire software development life cycle, including: Collect multiple basic performance data of the target project, pre-process each basic performance data to obtain pre-processed data; Create an evaluation model; The pre-processed data is input into the evaluation model. The evaluation model combines the pre-processed data to calculate the demand level of the target project, and continuously learns the corresponding relationship between the demand level and the project cycle to obtain a trained evaluation model. Collect the real-time performance basic data of the real-time target project, and adjust the project data of the real-time target project through the trained evaluation model combined with the real-time performance basic data.

[0008] Preferably, a plurality of basic performance data of the target project is collected, and each basic performance data is preprocessed to obtain preprocessed data, including: Create performance data tables; Collect multiple performance basic data of the target project and put all the collected performance basic data into the performance data table; Randomly select a performance basic data from the performance data table, and determine whether the selected performance basic data contains duplicate data; If duplicate data exists, delete any of the duplicate data to obtain deduplicated data; Determine whether there is missing data in the deduplicated data; If there are missing data, the mean is used to fill in the missing data to obtain preprocessed data; Return and randomly select a basic performance data from the performance data table until every basic performance technology in the performance data table is selected, thereby obtaining a plurality of pre-processed data.

[0009] Preferably, the preprocessed data is input into the evaluation model, and the evaluation model is combined with the preprocessed data to calculate the demand level of the target project, and the corresponding relationship between the demand level and the project cycle is continuously learned to obtain a trained evaluation model, including: Randomly select the pre-processed data of a target project from the performance data table; Obtaining a user core level based on the developer data in the pre-processed data; Determine the priority of each sub-requirement of the pre-processed data and the statistical score corresponding to each priority based on the MoSCoW principle; Combining the priority of each sub-requirement and the user's core level, the requirement level of the target project corresponding to the pre-processed data is obtained; Return the pre-processed data of a randomly selected target item from the performance data table until all target items in the performance data table are selected, and obtain the demand level of each target item.

[0010] Preferably, the pre-processed data is input into the evaluation model, the evaluation model is combined with the pre-processed data to calculate the demand level of the target project, and the corresponding relationship between the demand level and the project cycle is continuously learned to obtain a trained evaluation model, which also includes: selecting pre-processing data of a target project from the performance data table and obtaining a demand level of the target project; establishing a coupling relationship between the project data and the demand level, and taking the project data, the demand level, and the coupling relationship between the project data and the demand level as a training sample; inputting the training sample into the evaluation model, so that the evaluation model continuously learns the corresponding relationship between the pre-processing data and the demand level and the project data; returning to selecting pre-processing data from the performance data table and obtaining a demand level of the pre-processing data until all pre-processing data in the performance data table are input into the evaluation model, obtaining a trained evaluation model; the trained evaluation model has the ability to automatically evaluate the demand level of the input performance base data according to the performance base data.

[0011] Preferably, the priority of each sub-demand of the pre-processing data is determined based on the MoSCoW principle, including: setting multiple demand priorities and statistical scores corresponding to each demand priority; obtaining each sub-demand contained in the pre-processing data; dividing the demand priority of the sub-demand according to the type of each sub-demand; assigning a corresponding statistical score based on the demand priority of each sub-demand.

[0012] Preferably, the demand priority includes must-have level, should-have level, can-have level, and need-not level.

[0013] Preferably, real-time performance base data of a real-time target project is collected, and the project data of the real-time target project is adjusted by the trained evaluation model combined with the real-time performance base data, including: collecting real-time performance base data of a real-time target project; inputting the real-time performance base data into the trained evaluation model; obtaining real-time demand level output by the trained evaluation model, and estimating real-time project period based on the demand level.

[0014] Preferably, the real-time demand level output by the trained evaluation model is obtained, and the real-time project period is estimated based on the demand level, including: obtaining real-time demand level; obtaining project data corresponding to the demand level of the real-time demand level, and recording the obtained project data as target data; assigning the project period of the target data to the real-time demand level, and recording the real-time project period required by the real-time demand level as the real-time project period.

[0015] The present application relates to a system and method for evaluating the performance of a software development life cycle. The system collects multiple basic performance data of a target project, pre-processes each basic performance data to obtain pre-processed data, then creates an evaluation model, and inputs the pre-processed data into the evaluation model. The evaluation model is combined with the pre-processed data to calculate the demand level of the target project, and the correspondence between the demand level and the project cycle is continuously learned to obtain a trained evaluation model. Finally, the real-time basic performance data of the real-time target project is collected, and the project data of the real-time target project is adjusted by combining the trained evaluation model with the real-time basic performance data. The present application calculates the demand level of the target project by combining customer data with the demand priority of sub-demands, thereby accurately controlling the core requirements of software development, improving the accuracy and efficiency of demand analysis for software development, and making the full performance evaluation of the software development life cycle more reasonable. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the structure of a software development life cycle performance evaluation system provided in one embodiment of the present application.

[0017] Figure 2 A flowchart of a method for evaluating the effectiveness of the entire software development life cycle provided in one embodiment of the present application.

[0018] Reference numerals: 100, acquisition component; 200, evaluation component. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application more clearly understood, the present application is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0020] On the one hand, the present application provides a software development life cycle full-process performance evaluation system.

[0021] like Figure 1 As shown, in one embodiment of the present application, a software development life cycle full-process performance evaluation system includes a collection component 100 and an evaluation component 200. The collection component 100 collects basic performance data, and the evaluation component 200 is communicated with the collection component 100. The evaluation component 200 combines the basic performance data to perform performance evaluation on the software development life cycle.

[0022] This embodiment relates to a system for evaluating the performance of a software development life cycle. The acquisition component 100 collects basic performance data of a target project, and all collected data is input into the evaluation component 200. The evaluation component 200 combines the basic performance data of the target project to complete the performance evaluation of the target project.

[0023] In one embodiment of the present application, the performance basic data includes demand data, developer data, customer data, and project data; Specifically, project data includes project start time, project end time, and project implementation period; Customer data includes customer type, industry, and customer information.

[0024] It should be noted that the life cycle effectiveness of software development refers to improving the quality, efficiency and customer satisfaction of software by optimizing the management and execution of each stage in the entire software development process. The software development life cycle usually includes the following main stages: requirements analysis, design, coding, testing, deployment, operation and maintenance.

[0025] Among these stages, demand analysis is the most important one. Therefore, in this application, the project requirements of different target projects are divided into different levels by combining the performance basic data, so as to facilitate users to carry out production management based on the demand level of the target project. For example, for target projects with high demand levels, more developers can be arranged, while for target projects with low demand levels, the number of developers can be appropriately reduced.

[0026] like Figure 2 As shown, on the other hand, this application also provides a method for evaluating the effectiveness of the entire software development life cycle, including: S100, collecting multiple basic performance data of the target project, preprocessing each basic performance data to obtain preprocessed data; S200, creating an evaluation model; S300: Input the preprocessed data into the evaluation model, calculate the demand level of the target project through the evaluation model combined with the preprocessed data, and continuously learn the corresponding relationship between the demand level and the project cycle to obtain a trained evaluation model; S400 , collecting real-time performance basic data of the real-time target project, and adjusting the project data of the real-time target project by combining the trained evaluation model with the real-time performance basic data.

[0027] It should be noted that by collecting multiple performance basis data of the target project, preprocessing each performance basis data to obtain preprocessed data, then creating an evaluation model and inputting the preprocessed data into the evaluation model, calculating the demand level of the target project through the evaluation model combined with the preprocessed data, and continuously learning the corresponding relationship between the demand level and the project cycle to obtain the trained evaluation model, finally collecting real-time performance basis data of the real-time target project, adjusting the project data of the real-time target project through the trained evaluation model combined with the real-time performance basis data, the demand level of the target project is calculated by combining the customer data and the demand priority of the sub-demand, so as to accurately control the core demand of software development, improve the accuracy and efficiency of demand analysis of software development, and make the overall performance evaluation of software development life cycle more reasonable.

[0028] In an embodiment of the present application, S100 comprises: S110, creating a performance data table; S120, collecting multiple performance basis data of multiple target projects respectively, and putting all the collected performance basis data into the performance data table; S130, randomly selecting one performance basis data from the performance data table, and judging whether the selected performance basis data has duplicate data; S140, if there is duplicate data, deleting any data in the duplicate data to obtain de-duplicated data; S150, judging whether there is missing data in the de-duplicated data; S160, if there is missing data, filling the missing data by mean value to obtain preprocessed data; S170, returning to step S130 until each performance basis technology in the performance data table is selected, and multiple preprocessed data are obtained.

[0029] It should be noted that by collecting multiple performance basis data of multiple different projects, and preprocessing these performance basis data, the duplicate data or invalid data in the data is removed, so that the obtained preprocessed data is more complete and reliable, and the trained evaluation model obtained by training these preprocessed data is more accurate.

[0030] In an embodiment of the present application, S300 comprises: S310, randomly selecting preprocessed data of a target project from the performance data table; S320, obtaining a user core level based on the developer data in the preprocessed data; Specifically, the user core level is calculated by formula 1; Formula 1;

[0031] in, is the customer's user core level, is the j-th customer information of the customer, is the weight corresponding to the j-th customer information of the customer, and M is the total number of customer information contained in the customer data;

[0032] For different customers, we combine their customer data to determine their user core level. Customers with higher user core levels have higher priorities, and the level of demand they raise is higher.

[0033] Before executing step S320, it is necessary to assign corresponding scores and weights to each piece of customer information to facilitate calculation. Since the weights can be adjusted in real time with different project types, the calculation of the customer's user core level is more flexible and the weights can be adjusted according to needs to ensure the reliability of the user core level. S330, determining the priority of each sub-requirement of the pre-processed data and the statistical score corresponding to each priority based on the MoSCoW principle; S340, determining the priority of each sub-requirement of the pre-processed data and a statistical score corresponding to each priority based on the MoSCoW principle; Specifically, the statistical score corresponding to each demand priority is calculated using Formula 2; Formula 2;

[0034] in, is the requirement level of the target project, is the user core level corresponding to the customers of the target project, is the statistical score corresponding to the i-th sub-requirement of the target project, and N is the total number of sub-requirements included in the target project;

[0035] S350, returning to step S310, until all target items in the performance data table are selected, and the demand level of each target item is obtained;

[0036] It should be noted that the user core level of each customer is calculated separately by combining the customer data in the preprocessed data. The project demands proposed by customers of different levels are handled differently. For example, the project demands proposed by customers with higher user core levels should be met first. Therefore, the user core level is linked to the project demand level, so that customer data can also be used as one of the influencing factors when calculating the demand level of the target project, making the final demand level of the target project more reasonable.

[0037] After obtaining the customer's user core level through formula 1, the demand level of the target project is calculated by combining the user core level with the demand priority of each sub-demand through formula 2. This allows the calculation of the demand level to take into account every sub-demand in the project requirements to ensure the comprehensiveness of the demand level. After obtaining the demand level of the target project, the personnel or project cycle of the target project can be adjusted according to different demand levels, thereby improving the development efficiency of the target project.

[0038] In one embodiment of the present application, S300 further includes: S360, selecting pre-processed data of a target project from the performance data table and obtaining the demand level of the target project; S370, establishing a coupling relationship between project data and requirement levels, and using the project data, requirement levels, and the coupling relationship between the project data and requirement levels as a training sample; S380, inputting the training samples into the evaluation model so that the evaluation model continuously learns the correspondence between the pre-processed data, the demand level, and the project data; S390, return to step S360, until all pre-processed data in the performance data table are input into the evaluation model to obtain a trained evaluation model; the trained evaluation model has the ability to automatically evaluate the demand level of the performance basic data based on the input performance basic data.

[0039] It should be noted that after obtaining the demand levels of multiple target projects, these target projects are input into the evaluation model as training samples, thereby continuously improving the database of the evaluation model until the trained evaluation model is obtained. Then, when the user uses it, the real-time performance basic data of the real-time target project can be directly input into the trained evaluation model to directly obtain the real-time demand level of the real-time target project, and the real-time demand level can be combined to adjust the personnel needs or project cycle of the real-time target project, thereby improving the efficiency of the software development life cycle performance evaluation.

[0040] In one embodiment of the present application, S340 includes: S341, setting multiple demand priorities and the statistical score corresponding to each demand priority; Specifically, the higher the priority of a sub-demand, the higher its corresponding statistical score; S342, obtaining each sub-requirement included in the pre-processed data; S343, dividing the demand priorities of the sub-demands according to the type of each sub-demand; S344: Assign a corresponding statistical score based on the priority of each sub-demand.

[0041] It should be noted that the MoSCoW principle is a classic method for prioritizing requirements. By dividing requirements into four categories, it helps teams clarify the tasks that "must be done" and "can be left undo" when resources are limited, thereby optimizing project delivery efficiency.

[0042] In one embodiment of the present application, the priorities of the sub-requirements include a must-have level, a should-have level, a possible-have level, and a no-need-level.

[0043] It should be noted that, by arranging them from high to low priority, we can get the must-have level, should-have level, possible-have level and unnecessary level. The must-have sub-requirements are the core requirements of the project requirements, which are necessary for interactive products or to achieve goals, such as the payment function of the e-commerce platform. The should-have sub-requirements are important requirements of the project requirements, and their priority is slightly lower than the must-have sub-requirements. The possible-have sub-requirements are the requirements for improving the experience in the project requirements, and their priority is slightly lower than the must-have level. The unnecessary level is the lower-value requirement in the project requirements, and has the lowest priority. It may be a pseudo-requirement in the project requirements. Therefore, the unnecessary sub-requirements should be deleted or excluded.

[0044] In one embodiment of the present application, S400 includes: S410, collecting real-time performance basic data of real-time target projects; S420, inputting the real-time performance basic data into the trained evaluation model; S430: Obtain the real-time demand level output by the trained evaluation model, and estimate the real-time project cycle based on the demand level.

[0045] It should be noted that after the evaluation model is trained, the user can directly input the real-time performance basic data of the real-time target project into the trained evaluation model, so that the trained evaluation model can combine the real-time performance basic data to calculate the real-time demand level of the real-time target project, and finally adjust the number of developers or project cycle length of the real-time target project based on the real-time demand level.

[0046] In one embodiment of the present application, S430 includes: S431, obtaining real-time demand level; S432, obtaining project data of a demand level corresponding to the real-time demand level, and recording the obtained project data as target data; S433: Assign the project cycle of the target data to the real-time demand level and record it as the real-time project cycle required by the real-time demand level.

[0047] It should be noted that after obtaining the real-time demand level through the trained evaluation model, the project data in the historical target project with the same level as the real-time target project can be used as a reference, so that users have a reference standard to facilitate users to adjust developer data and project cycle data.

[0048] The various technical features of the above-described embodiments can be combined arbitrarily, and the execution order of the method steps is not restricted. In order to make the description concise, not all possible combinations of the various technical features in the above-described 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 specification.

[0049] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A software development life cycle performance evaluation system, characterized by: include: A collection component for collecting basic performance data; An evaluation component is in communication with the acquisition component, and is used to perform an effectiveness evaluation on the software development life cycle in combination with the basic effectiveness data.

2. The software development life cycle performance evaluation system according to claim 1, characterized in that: The performance basic data includes demand data, developer data, customer data and project data.

3. A software development lifecycle performance evaluation method, applied to the software development lifecycle performance evaluation system according to any one of claims 1 to 2, characterized in that: include: Collect multiple basic performance data of the target project, pre-process each basic performance data to obtain pre-processed data; Create an evaluation model; The pre-processed data is input into the evaluation model. The evaluation model combines the pre-processed data to calculate the demand level of the target project, and continuously learns the corresponding relationship between the demand level and the project cycle to obtain a trained evaluation model. Collect the real-time performance basic data of the real-time target project, and adjust the project data of the real-time target project through the trained evaluation model combined with the real-time performance basic data.

4. The software development life cycle performance evaluation method according to claim 3, characterized in that: Collect multiple basic performance data of the target project, pre-process each basic performance data, and obtain pre-processed data, including: Create performance data tables; Collect multiple performance basic data of the target project and put all the collected performance basic data into the performance data table; Randomly select a performance basic data from the performance data table, and determine whether the selected performance basic data contains duplicate data; If duplicate data exists, delete any of the duplicate data to obtain deduplicated data; Determine whether there is missing data in the deduplicated data; If there are missing data, the mean is used to fill in the missing data to obtain preprocessed data; Return and randomly select a basic performance data from the performance data table until every basic performance technology in the performance data table is selected, thereby obtaining a plurality of pre-processed data.

5. The software development life cycle performance evaluation method according to claim 4, characterized in that: The preprocessed data is input into the evaluation model. The evaluation model combines the preprocessed data to calculate the demand level of the target project. The corresponding relationship between the demand level and the project cycle is continuously learned to obtain a trained evaluation model, including: Randomly select the pre-processed data of a target project from the performance data table; Obtaining a user core level based on the developer data in the pre-processed data; Determine the priority of each sub-requirement of the pre-processed data and the statistical score corresponding to each priority based on the MoSCoW principle; Combining the priority of each sub-requirement and the user's core level, the requirement level of the target project corresponding to the pre-processed data is obtained; Return the pre-processed data of a randomly selected target item from the performance data table until all target items in the performance data table are selected, and obtain the demand level of each target item.

6. The software development life cycle performance evaluation method according to claim 5, characterized in that: The preprocessed data is input into the evaluation model. The evaluation model combines the preprocessed data to calculate the demand level of the target project, and continuously learns the correspondence between the demand level and the project cycle to obtain a trained evaluation model, which also includes: Selecting pre-processed data of a target item from the performance data table and obtaining the demand level of the target item; Establish a coupling relationship between project data and requirement levels, and use the project data, requirement levels, and the coupling relationship between project data and requirement levels as a training sample; Input the training samples into the evaluation model so that the evaluation model continuously learns the correspondence between the pre-processed data, the demand level and the project data; Return to select a preprocessed data from the performance data table and obtain the requirement level of the preprocessed data, until all the preprocessed data in the performance data table are input into the evaluation model to obtain a trained evaluation model; the trained evaluation model has the ability to automatically evaluate the requirement level of the performance basic data based on the input performance basic data.

7. The software development life cycle performance evaluation method according to claim 6, characterized in that: Determine the priority of each sub-requirement of the pre-processed data based on the MoSCoW principle, including: Set multiple demand priorities and the corresponding statistical scores for each demand priority; Obtain each sub-requirement contained in the preprocessed data; Prioritize sub-requirements based on the type of each sub-requirement; Assign a corresponding statistical score based on the priority of each sub-requirement.

8. The software development life cycle performance evaluation method according to claim 7, characterized in that: The demand priorities include must-have, should-have, possible-have and not-needed.

9. The software development life cycle performance evaluation method according to claim 8, characterized in that: Collect the real-time performance basic data of the real-time target project, and adjust the project data of the real-time target project by combining the trained evaluation model with the real-time performance basic data, including: Collect real-time performance basic data of real-time target projects; Input the real-time performance basic data into the trained evaluation model; Obtain the real-time demand level output by the trained evaluation model and estimate the real-time project cycle based on the demand level.

10. The software development life cycle performance evaluation method according to claim 9, characterized in that: Obtain the real-time demand level output by the trained evaluation model and estimate the real-time project cycle based on the demand level, including: Get real-time demand levels; Acquire project data of a demand level corresponding to the real-time demand level, and record the acquired project data as target data; The project cycle of the target data is assigned to the real-time demand level and recorded as the real-time project cycle required by the real-time demand level.

Images