Software development workload evaluation system and method based on large model

Abstract

The invention relates to the technical field of software development, and particularly discloses a software development workload evaluation system and method based on a large model, and the system comprises a data collection module which is used for collecting function point data information of software and software requirement specification data information; the data recording module is used for recording development workload data information of similar projects completed in the past; the expert scoring module is used for recording software development workload data estimated by different experts; the predicted software development workload is calculated through the data processing module in combination with data of the data acquisition module, the data recording module and the expert scoring module, and the accuracy of a calculation result can be improved through multi-dimensional data information; therefore, the accuracy and reliability of the evaluation result of the workload required by the software development project are improved, and the situation that the final evaluation result of the software development workload has a large error due to the fact that few data can be provided is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of software development, and in particular to a software development workload evaluation system and method based on a large model. Background Art

[0002] The software development workload estimation system is a comprehensive tool or platform specifically used to predict and estimate the workload required for software development projects. It is used to help project managers, development teams and stakeholders better understand the scale, complexity, resource requirements and potential risks of the project, so as to formulate reasonable schedules, budgets and plans.

[0003] When developing a certain software, a development workload assessment system will be used to predict and estimate the workload required for software development. The traditional method is to conduct statistical analysis based on historical project data, such as development time, resource input, task decomposition, etc., to provide a reference for the workload assessment of the current project, and combine industry standards and team experience to conduct trend forecasting and risk assessment, thereby realizing the assessment of development workload.

[0004] In the prior art, when estimating the software development workload, the project historical data is usually combined with industry standards and team experience for evaluation. However, in the early stages of software development, the amount of historical data on the project is small, so there is not much data available. This may lead to large errors in the final evaluation results of the software development workload, affecting the reliability of the data. Summary of the invention

[0005] The purpose of the present invention is to provide a software development workload assessment system and method based on a large model to solve the following technical problems: How to improve the accuracy of the final evaluation results of software development workload.

[0006] The purpose of the present invention can be achieved through the following technical solutions: A software development workload assessment system and method based on a large model, the system comprising: Data collection module, used to collect software function point data information and software requirement specification data information; Data recording module, used to record the development workload data information of similar projects completed in the past; Expert scoring module, used to record the software development workload data estimated by different experts; The data processing module is used to establish a big data calculation model and calculate the software development workload after the prediction by combining the data from the data collection module, the data recording module and the expert scoring module; The data evaluation module is used to compare the predicted software development workload calculated in the data processing module with the preset software development workload, and to evaluate the workload required for software development based on the comparison results and make corresponding decisions.

[0007] Furthermore, the function point data collected by the data collection module includes the number of internal logic files and the number of externally received files, and the software requirement specification data information collected by the data collection module includes the number of preset requirement specification changes, the number of requirement specification modifications, the amount of requirement specification information content lost, the amount of requirement specification information content ambiguity, the number of technical difficulties and the amount of available resources.

[0008] Furthermore, the processing process of the data processing module includes: By formula Calculate the work required to obtain the software requirements specification ; in, The number of changes to the preset requirements specification. is the influence coefficient of the number of modification times of the requirement specification, which is set based on experience fitting. is the amount of information lost in the software requirements specification, is the ambiguity of the information content in the requirement specification, is the total amount of information in the requirements specification, is the number of technical difficulties in the requirements specification, is the available resource amount, is the number of preset technical difficulties, is the preset amount of available resources, is the error influence coefficient, which is set according to empirical fitting.

[0009] Furthermore, the processing process of the data processing module also includes: By formula Calculate the coefficient of dispersion of the workload of similar projects completed in the past ; Where b is the total number of similar projects completed in the past, a is any similar project completed in the past, is the development workload of the ath similar project completed previously, For all The average value of g is the proportionality coefficient, which is set according to empirical fitting. For all The maximum value in is the weight coefficient of the ath similar project completed in the past.

[0010] Furthermore, the processing process of the data processing module also includes: By formula Calculate the software development effort after obtaining the forecast ; Where n is the total number of experts involved in software development workload estimation, i is any expert involved in software development workload estimation, is the software development workload estimated by the i-th expert, is the weight coefficient of the i-th expert, is the number of logical files in the software function point, is the number of externally received files in the software function point, is the total number of files, and 2 is the weight coefficient, To adjust the coefficient comparison table function, according to The impact of the range of values ​​in the medium on the software development workload is obtained based on testing.

[0011] Furthermore, the evaluation process of the data evaluation module includes: The predicted software development effort is obtained by calculating Compared with the preset development workload threshold Make a comparison; like ,The system determines that the workload required for the software development is high and requires additional allocation of more human resources, materials and technical support; like ,The system determines that the workload of software development is low and does not require additional allocation of human resources, materials and technical support.

[0012] Furthermore, the process of the expert scoring module includes: Multiple experts make a preliminary score for the software development workload based on information from multiple dimensions, including the project's functional requirements, non-functional requirements, technical difficulty, team capabilities, resource limitations, etc., and assign different weights based on different experts' domain knowledge, experience, and understanding of specific issues.

[0013] A software development workload assessment method based on a large model, the method comprising: S1: Collect software function point data information and software requirement specification data information through the data collection module; S2: Record the software development workload data estimated by different experts through the expert scoring module; S3: Calculate the work demand of the software requirement specification by combining the data processing module with the software requirement specification data information collected by the data collection module; S4: Calculate the discrete coefficient of the workload of similar projects completed in the past by combining the development workload data information of similar projects completed in the past recorded in the data recording module through the data processing module; S5: The predicted software development workload is calculated by combining the software function point data information and the software development workload data estimated by different experts, as well as the data calculated in S3 and S4, through the data processing module; S6: Compare the predicted software development workload calculated in the data processing module with the preset software development workload, and evaluate the software development workload based on the comparison result.

[0014] Beneficial effects of the present invention: (1) The present invention calculates the predicted software development workload by combining the data from the data processing module, the data collection module, the data recording module and the expert scoring module. The multi-dimensional data information can improve the accuracy of the calculation results, thereby improving the accuracy and reliability of the workload evaluation results required for the software development project, and avoiding the situation where the final evaluation results of the software development workload have large errors due to less data available.

[0015] (2) The present invention combines the dispersion coefficient of the workload of similar projects completed in the past Work requirements related to software requirements specification , and based on data such as the average software development workload estimated by all experts, the number of logical files in software function points and the number of externally received files, high-value and multi-dimensional data can improve the reliability of the calculation results, thereby improving the accuracy of the final evaluation results of the software development workload.

[0016] (3) The present invention calculates the predicted software development workload Compared with the preset development workload threshold By comparing, the workload required for the software development can be judged based on the comparison results, and because the predicted software development workload is calculated based on multi-dimensional and large amounts of data, its reliability is relatively high, thereby improving the accuracy and reliability of the judgment on the workload required for the software development, and avoiding the situation where the final evaluation result of the software development workload has large errors due to the small amount of data available.

[0017] (4) The present invention first calculates the discrete coefficients of the workload of the software requirement specification and the workload of similar projects completed in the past based on the collected different information and recorded data. Then, on this basis, the software function point data information and the software development workload data estimated by different experts can be combined to calculate the predicted software development workload. With the support of a large amount of multi-source data, the accuracy of the calculation results can be improved, thereby improving the accuracy and reliability of subsequent evaluation of the software development workload. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention will be further described below in conjunction with the accompanying drawings.

[0019] Figure 1 It is a schematic block diagram of a software development workload assessment system based on a large model in the present invention; Figure 2 It is a flow chart of a software development workload assessment method based on a large model in the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] See also Figure 1 As shown, in one embodiment, the present application provides a software development workload assessment system and method based on a large model, the system comprising: Data collection module, used to collect software function point data information and software requirement specification data information; Data recording module, used to record the development workload data information of similar projects completed in the past; Expert scoring module, used to record the software development workload data estimated by different experts; The data processing module is used to establish a big data calculation model and calculate the software development workload after the prediction by combining the data from the data collection module, the data recording module and the expert scoring module; The data evaluation module is used to compare the predicted software development workload calculated in the data processing module with the preset software development workload, and to evaluate the workload required for software development based on the comparison results and make corresponding decisions; Through the above technical solution, this embodiment provides a data collection module, which can be used to collect software function point data information and software requirement specification data information, and provides a data recording module and an expert scoring module, which are respectively used to record the development workload data information of similar projects completed in the past, and record the software development workload data estimated by different experts. When evaluating the software development workload, the data processing module can be combined with the data of the data collection module, the data recording module and the expert scoring module to calculate the predicted software development workload, and the data evaluation module can be used to compare the predicted software development workload calculated in the data processing module with the preset software development workload, and the required workload of the software development is evaluated according to the comparison result, and corresponding decisions are made; Through such a setting, the predicted software development workload is calculated through the data processing module combined with the data from the data acquisition module, the data recording module and the expert scoring module. The multi-dimensional data information can improve the accuracy of the calculation results, thereby improving the accuracy and reliability of the workload assessment results required for the software development project, and avoiding the situation where there are large errors in the final assessment results of the software development workload due to less data available.

[0022] The function point data collected by the data collection module includes the number of internal logic files and the number of externally received files. The software requirement specification data information collected by the data collection module includes the number of preset requirement specification changes, the number of requirement specification modifications, the amount of requirement specification information content loss, the amount of requirement specification information content ambiguity, the number of technical difficulties and the amount of available resources; Through the above technical scheme, this embodiment provides function point data collected by the data acquisition module, which reflects the software function complexity and data complexity. The workload of software development is evaluated based on the data, which can improve the accuracy of the evaluation results. The data acquisition module also collects software requirement specification data information, which can reflect the complexity of the software requirement specification. The data can also improve the accuracy of the software development workload evaluation results, thereby improving the accuracy of subsequent calculations of the predicted software development workload.

[0023] The processing process of the data processing module includes: By formula Calculate the work required to obtain the software requirements specification ; in, The number of changes to the preset requirements specification. is the influence coefficient of the number of modification times of the requirement specification, which is set based on experience fitting. is the amount of information lost in the software requirements specification, is the ambiguity of the information content in the requirement specification, is the total amount of information in the requirements specification, is the number of technical difficulties in the requirements specification, is the available resource amount, is the number of preset technical difficulties, is the preset amount of available resources, is the error influence coefficient, which is set according to empirical fitting; Through the above technical solution, this embodiment provides the software requirements specification work requirements , can be obtained by formula It is obvious that the more the preset number of changes to the requirements specification, the number of technical difficulties in the requirements specification, the amount of ambiguity in the information content of the requirements specification, and the amount of loss of information content in the requirements specification, and the less the available resources, the higher the demand for software requirements specification work will be. Conversely, the smaller the preset number of changes to the requirements specification, the number of technical difficulties in the requirements specification, the amount of ambiguity in the information content of the requirements specification, and the amount of loss of information content in the requirements specification, and the more available resources, the smaller the demand for software requirements specification work will be. By setting it this way, the demand for software requirements specification work can be calculated, and the data can improve the accuracy of the workload assessment results of software development, thereby improving the accuracy of subsequent calculations of the predicted software development workload.

[0024] The processing process of the data processing module also includes: By formula Calculate the coefficient of dispersion of the workload of similar projects completed in the past ; Where b is the total number of similar projects completed in the past, a is any similar project completed in the past, is the development workload of the ath similar project completed previously, For all The average value of g is the proportionality coefficient, which is set according to empirical fitting. For all The maximum value in is the weight coefficient of the ath similar project completed in the past; Through the above technical solution, this embodiment provides the discrete coefficient of the workload of similar projects completed in the past , can be obtained by formula The data obtained by calculation can reflect the volatility of the workload of similar projects completed in the past, that is, it represents the workload fluctuation of this type of project. If the value is low, it means that the development workload required for this type of project is roughly the same. Otherwise, it means that the development workload required for this type of project varies greatly. By calculating the discrete coefficient of the workload of similar projects completed in the past, another dimension of data can be provided for the evaluation of software development workload, thereby improving the accuracy of the final evaluation results of software development workload.

[0025] The processing process of the data processing module also includes: By formula Calculate the software development effort after obtaining the forecast ; Where n is the total number of experts involved in software development workload estimation, i is any expert involved in software development workload estimation, is the software development workload estimated by the i-th expert, is the weight coefficient of the i-th expert, is the number of logical files in the software function point, is the number of externally received files in the software function point, is the total number of files, and 2 is the weight coefficient, To adjust the coefficient comparison table function, according to The impact of the range of values ​​in the medium on the software development workload is obtained based on testing; Through the above technical solution, this embodiment provides the predicted software development workload , can be obtained by formula It is obvious that when the average software development workload estimated by all experts is higher, and the number of logic files in the software function points and the number of externally received files are larger, the workload required for the software development is higher. On the contrary, when the average software development workload estimated by all experts is lower, and the number of logic files in the software function points and the number of externally received files are smaller, the workload required for the software development is lower. Through this calculation method, this formula combines the discrete coefficient of the workload of similar projects completed in the past. Work requirements related to software requirements specification , and based on data such as the average software development workload estimated by all experts, the number of logical files in software function points and the number of externally received files, high-value and multi-dimensional data can improve the reliability of the calculation results, thereby improving the accuracy of the final evaluation results of the software development workload.

[0026] The evaluation process of the data evaluation module includes: The predicted software development effort is obtained by calculating Compared with the preset development workload threshold Make a comparison; like ,The system determines that the workload required for the software development is high and requires additional allocation of more human resources, materials and technical support; like ,The system determines that the workload of software development is low and does not require additional allocation of human resources, materials and technical support; Through the above technical solution, this embodiment calculates the predicted software development workload Compared with the preset development workload threshold By comparing, the workload required for the software development can be judged based on the comparison results, and because the predicted software development workload is calculated based on multi-dimensional and large amounts of data, its reliability is relatively high, thereby improving the accuracy and reliability of the judgment on the workload required for the software development, and avoiding the situation where the final evaluation result of the software development workload has large errors due to the small amount of data available.

[0027] The process of the expert scoring module includes: Multiple experts will conduct a preliminary scoring of the software development workload based on information from multiple dimensions, such as the project's functional requirements, non-functional requirements, technical difficulty, team capabilities, and resource constraints. Different weights will be assigned based on the domain knowledge, experience, and understanding of specific issues of different experts. Through the above technical scheme, this embodiment provides a process of the expert scoring module. Since experts are usually screened and invited, their opinions are based on their deep professional knowledge and rich practical experience, so the credibility of the experts' evaluation of the software development workload is relatively high. In addition, by assigning different weights to the evaluation results of different experts, the weights of different experts' evaluation results can be corrected according to the domain knowledge, experience and understanding of specific issues in which different experts are good at, thereby improving the accuracy and reliability of the experts' evaluation results of the software development workload, and then improving the accuracy of the final evaluation results of the software development workload.

[0028] See also Figure 2 As shown, a software development workload assessment method based on a large model comprises: S1: Collect software function point data information and software requirement specification data information through the data collection module; S2: Record the software development workload data estimated by different experts through the expert scoring module; S3: Calculate the work demand of the software requirement specification by combining the data processing module with the software requirement specification data information collected by the data collection module; S4: Calculate the discrete coefficient of the workload of similar projects completed in the past by combining the development workload data information of similar projects completed in the past recorded in the data recording module through the data processing module; S5: The predicted software development workload is calculated by combining the software function point data information and the software development workload data estimated by different experts, as well as the data calculated in S3 and S4, through the data processing module; S6: comparing the predicted software development workload calculated in the data processing module with the preset software development workload, and evaluating the software development workload according to the comparison result; Through the above technical scheme, this embodiment provides a software development workload evaluation method based on a large model, firstly, the function point data information of the software and the software requirement specification data information are collected through the data collection module, and the software development workload data estimated by different experts are recorded through the expert scoring module, and then the software requirement specification work requirement is calculated by the data processing module in combination with the software requirement specification data information collected by the data collection module, and the discrete coefficient of the workload of similar projects completed in the past is calculated by the data processing module in combination with the development workload data information of similar projects completed in the past recorded in the data recording module, and then the predicted software development workload can be calculated by the data processing module in combination with the software function point data information and the software development workload data estimated by different experts, and finally the predicted software development workload calculated in the data processing module is compared with the preset software development workload, and the software development workload is evaluated according to the comparison result; Through such a setting, we can first calculate the discrete coefficients of the workload of the software requirement specification and the workload of similar projects completed in the past based on the different information collected and the recorded data. Then, on this basis, combined with the software function point data information and the software development workload data estimated by different experts, the predicted software development workload can be calculated. With the support of large amounts of multi-source data, the accuracy of the calculation results can be improved, thereby improving the accuracy and reliability of subsequent evaluations of the software development volume.

[0029] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A software development workload evaluation system based on a large model, characterized in that: The system comprises: Data collection module, used to collect software function point data information and software requirement specification data information; Data recording module, used to record the development workload data information of similar projects completed in the past; Expert scoring module, used to record the software development workload data estimated by different experts; The data processing module is used to establish a big data calculation model and calculate the software development workload after the prediction by combining the data from the data collection module, the data recording module and the expert scoring module; The data evaluation module is used to compare the predicted software development workload calculated in the data processing module with the preset software development workload, and to evaluate the workload required for software development based on the comparison results and make corresponding decisions.

2. A software development workload assessment system based on a large model according to claim 1, characterized in that: The function point data collected by the data acquisition module includes the number of internal logic files and the number of externally received files, and the software requirement specification data information collected by the data acquisition module includes the preset number of requirement specification changes, the number of requirement specification modifications, the amount of requirement specification information content lost, the amount of requirement specification information content ambiguity, the number of technical difficulties and the amount of available resources.

3. A software development workload assessment system based on a large model according to claim 2, characterized in that: The processing process of the data processing module includes: By formula Calculate the work required to obtain the software requirements specification ; in, The number of changes to the preset requirements specification. is the influence coefficient of the number of modification times of the requirement specification, which is set based on experience fitting. is the amount of information lost in the software requirements specification, is the ambiguity of the information content in the requirement specification, is the total amount of information in the requirements specification, is the number of technical difficulties in the requirements specification, is the available resource amount, is the number of preset technical difficulties, is the preset amount of available resources, is the error influence coefficient, which is set according to empirical fitting.

4. A software development workload assessment system based on a large model according to claim 3, characterized in that: The processing process of the data processing module also includes: By formula Calculate the coefficient of dispersion of the workload of similar projects completed in the past ; Where b is the total number of similar projects completed in the past, a is any similar project completed in the past, is the development workload of the ath similar project completed previously, For all The average value of g is the proportionality coefficient, which is set according to empirical fitting. For all The maximum value in is the weight coefficient of the ath similar project completed in the past.

5. A software development workload assessment system based on a large model according to claim 4, characterized in that: The processing process of the data processing module also includes: By formula Calculate the software development effort after obtaining the forecast ; Where n is the total number of experts involved in software development workload estimation, i is any expert involved in software development workload estimation, is the software development workload estimated by the i-th expert, is the weight coefficient of the i-th expert, is the number of logical files in the software function point, is the number of externally received files in the software function point, is the total number of files, and 2 is the weight coefficient, To adjust the coefficient comparison table function, according to The impact of the range of values ​​in the medium on the software development workload is obtained based on testing.

6. A software development workload assessment system based on a large model according to claim 5, characterized in that: The evaluation process of the data evaluation module includes: The predicted software development effort is obtained by calculating Compared with the preset development workload threshold Make a comparison; like ,The system determines that the workload required for the software development is high and requires additional allocation of more human resources, materials and technical support; like ,The system determines that the workload of software development is low and does not require additional allocation of human resources, materials and technical support.

7. A software development workload assessment system based on a large model according to claim 1, characterized in that: The process of the expert scoring module includes: Multiple experts make a preliminary score for the software development workload based on information from multiple dimensions, including the project's functional requirements, non-functional requirements, technical difficulty, team capabilities, resource limitations, etc., and assign different weights based on different experts' domain knowledge, experience, and understanding of specific issues.

8. A software development workload assessment method based on a large model, characterized in that: The method adopts a software development workload assessment system based on a large model as described in claims 1-7, and the method comprises: S1: Collect software function point data information and software requirement specification data information through the data collection module; S2: Record the software development workload data estimated by different experts through the expert scoring module; S3: Calculate the work demand of the software requirement specification by combining the data processing module with the software requirement specification data information collected by the data collection module; S4: Calculate the discrete coefficient of the workload of similar projects completed in the past by combining the development workload data information of similar projects completed in the past recorded in the data recording module through the data processing module; S5: The predicted software development workload is calculated by combining the software function point data information and the software development workload data estimated by different experts, as well as the data calculated in S3 and S4, through the data processing module; S6: Compare the predicted software development workload calculated in the data processing module with the preset software development workload, and evaluate the software development workload based on the comparison result.

Images