JPH11224187A - Software quality evaluation system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a software quality evaluation system by which the quality of a software is sufficiently evaluated and then a quality reinforcing counterplan based on the quality evaluation is quickly provided. SOLUTION: This software quality evaluation system is provided with a quality evaluating part 1 for evaluating the quality of the software based on a test density being the ratio of a test item number as against the scale of the software to be used in a computer and an error detecting rate being the ratio of the number of errors which occur at the time of executing the software, a counterplan generating part 2 for previously generating the plural counterplans corresponding to plural quality evaluating results of be expected and an output part 4 for outputting the quality evaluating results evaluated by the quality evaluating part 1 and the counterplans corresponding thereto.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a software quality evaluation system for performing software quality evaluation based on a test result, and more particularly, to a uniform software quality evaluation and to respond to the quality evaluation result. Software quality evaluation system that can provide countermeasures.

[0002]

2. Description of the Related Art In today's society, it is impossible to live without being involved in computers, and ensuring the quality of software used in computers is an important issue.

[0005] Development in consideration of software quality is generally performed as follows. First, a design based on the specifications is performed. Next, software generation (coding) is performed based on this design. Subsequently, a test is performed on the generated software. At this time, a plurality of test items are set in consideration of the specifications, the scale of the software, and the like. Then, a test is performed based on the set test items. Based on the result of this test, the software quality is evaluated. Then, after this quality evaluation, if the quality is good, the software is shipped. On the other hand, if the quality is poor,
It is necessary to consider the cause of the failure and the measures to enhance the quality of the evaluation result. As a countermeasure, coding, review of specifications, etc. are available.

However, conventionally, there is no quantitative evaluation standard for software quality evaluation, and there is no quality evaluation system that can perform software quality evaluation in a unified manner. For this reason, unified software quality evaluation could not be performed.

[0005] Therefore, there are the following techniques as techniques for setting quantitative judgment criteria for software quality evaluation and performing unified software quality evaluation. That is, the technique is disclosed in Japanese Patent Application Laid-Open Nos. 5-216651 and 5-282178.

In the technique disclosed in Japanese Patent Application Laid-Open No. Hei 5-216661, the quality of software is evaluated as follows. That is, the test density, which is the ratio of the number of test items to the scale of the software, and the error detection rate, which is the ratio of the number of errors generated when the software executes the test of the number of test items with respect to the scale of the software, Defined as an index for software quality evaluation. Normally, the appropriate number of errors and the number of test items change according to the scale of the software. Therefore, as described above, the test density and the error detection rate are used as indexes for evaluating the quality of the software.

The test density area is divided into three stages based on a predetermined test density reference value. Similarly, the area of the error detection rate is divided into three stages based on the reference value of the predetermined error detection rate. Then, a quality evaluation matrix in which nine areas are set is generated by combining the test density area divided into three steps and the error detection rate area divided into three steps. Good and bad software quality are assigned to these nine areas.

Then, it is determined which of the nine areas of the quality evaluation matrix corresponds to the software to be evaluated from the test density and the error detection rate. Then, the evaluation (good,
(Bad), the quality of the software to be evaluated is evaluated.

In the technology disclosed in Japanese Patent Application Laid-Open No. 5-282178, software quality evaluation is performed as follows. That is, the test coverage indicating the degree of execution of the software by the test and the error detection rate, which is the ratio of the number of errors to the scale of the software, are defined as indices of software quality evaluation. The degree of execution of the software means that when a predetermined test is executed by the software, a predetermined number of functions (or the number of branch instructions, etc.) predicted to be executed in advance are actually executed. It shows the ratio of the number of functions (or the number of branch instructions, etc.).

[0010] Then, a plurality of reference ranges of error detection rates are set for a plurality of test coverages. When the test coverage and the error detection rate of the software to be evaluated are calculated, and the calculated error detection rate falls within the reference range of the error detection rate corresponding to the calculated test coverage, Evaluated the quality as good, and if not included, evaluated the quality as poor.

[0011]

However, the above-mentioned prior art has the following problems. That is, Japanese Patent Application Laid-Open Nos. Hei 5-216651 and Hei 5-28
In the technique according to Japanese Patent No. 2178, software quality is evaluated only in two stages of good and bad. However, although there is a difference in the degree of quality even when the evaluation is made as a defect, the conventional technology cannot perform an evaluation in consideration of the difference in the degree of quality. That is, the quality evaluation of the software was not sufficient.

Further, in the prior art, it was not possible to provide a worker with a measure for improving the quality with respect to the quality evaluation result. For this reason, the operator had to quickly consider a countermeasure based on the insufficient quality evaluation described above. As a result, the measures actually taken may differ depending on the experience, skill, and the like of the worker.

Therefore, the prior art has a problem in that it is not possible to sufficiently evaluate software quality and to promptly provide a worker with a measure based on the sufficiently performed quality evaluation.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and its purpose is to sufficiently evaluate the quality of software and quickly provide quality enhancement measures based on the quality evaluation. An object of the present invention is to provide a software quality evaluation system capable of performing the same.

[0015]

To achieve the above object, a software quality evaluation system according to a first aspect of the present invention includes a test density, which is a ratio of the number of test items to a scale of software used in a computer; Quality evaluation means for evaluating software quality based on an error detection rate which is a ratio of the number of errors generated when the software executes the test of the number of test items with respect to the scale of the software, Measure generation means for generating in advance a plurality of measure plans corresponding to the plurality of quality evaluation results, and output means for outputting the quality evaluation result evaluated by the quality evaluation means and the corresponding measure plan. Things.

According to a second aspect of the present invention, in the software quality evaluation system according to the first aspect, the quality evaluation means includes:
When respective reference values are set for the test density, the error detection rate, and a hit ratio that is a ratio of the error detection rate to the test density, a plurality of test densities and a plurality of errors are set based on these reference values. A quality evaluation matrix generating means for generating a quality evaluation matrix in which a detection rate, a plurality of hit rates, and a plurality of software quality states are associated with each other, and each quality evaluation corresponding to each software quality state is generated. From the quality evaluation generating means, the calculating means for calculating the hit rate from the test density and the error detection rate of the software to be evaluated, and the test density, the error detection rate and the hit rate of the software to be evaluated, according to the quality evaluation matrix, Determine the quality state of the software to be evaluated, generated by the quality evaluation generation means Of quality evaluation, a quality determination means for determining a quality evaluation corresponding to the quality state as the quality evaluation result, and it has a.

The software quality evaluation system according to a third invention is the software quality evaluation system according to the first or second invention, wherein the countermeasure plan generation means includes a plurality of countermeasure plans for each of the quality evaluations generated by the quality evaluation generation means. Each of the corresponding countermeasures is generated, and the output unit outputs a quality evaluation determined by the quality determining unit as the quality evaluation result, and a quality evaluation output unit that outputs the quality evaluation determined as the quality evaluation result. And a measure output means for outputting a measure corresponding to the quality evaluation determined by the quality determining means, among the measure plans generated by the measure plan generating means.

According to a fourth aspect of the present invention, in the software quality evaluation system according to the second or third aspect, the calculation unit is configured to input a test coverage indicating a degree of execution of the software to be evaluated by the test. Multiplying the test coverage by the test density of the software to be evaluated to calculate a new test density, and calculating the hit rate from the error detection rate of the software to be evaluated and the new test density, The determining means determines the quality state of the software to be evaluated from the error detection rate of the software to be evaluated, the new test density and the hit rate according to the quality evaluation matrix, and is generated by the quality evaluation generating means. Among the quality evaluations, a quality evaluation corresponding to the quality state is determined.

According to a fifth aspect of the present invention, in the software quality evaluation system according to the second or third aspect, the calculating means calculates the number of errors with respect to a test progress up to an intermediate stage of a test performed on software to be evaluated. From the relationship,
A growth prediction curve for the number of errors is calculated, and a prediction error detection rate is calculated from the number of errors after the test, which is predicted based on the growth prediction curve, and the prediction error detection rate and the test density after the test are completed. The hit rate is calculated, and the quality determining means determines a quality state of the software to be evaluated according to the quality evaluation matrix from the test density after the test, the prediction error detection rate, and the hit rate, The quality evaluation corresponding to the quality state is determined from the quality evaluations generated by the quality evaluation generation means.

According to a sixth aspect of the present invention, in the software quality evaluation system according to the first aspect, the quality evaluation means includes:
There is provided a quality re-evaluation means for re-evaluating the quality evaluation evaluated by the quality evaluation means.

According to a seventh aspect of the present invention, in the software quality evaluation system according to the sixth aspect, the quality re-evaluation means calculates the number of remaining errors that can be predicted not to be detected due to insufficient execution of software in the test. A residual error number calculation method generating means for generating a method for each quality evaluation result; an additional test method generating means for generating an additional test method for each quality evaluation result;
According to the re-evaluation information determination means for determining the remaining error number calculation method and the additional test method corresponding to the quality evaluation result evaluated by the quality evaluation means, and the remaining error number calculation method determined by the re-evaluation information determination means, A test based on the remaining error number calculating means for calculating the number of remaining errors, the remaining error number calculated by the remaining error number calculating means, and the additional test method determined by the re-evaluation information determining means, Recalculating means for recalculating the test density and error detection rate of the software to be evaluated in consideration of the results executed by the software of the software to be evaluated, and the software to be evaluated based on the recalculated test density and error detection rate. And a quality re-determining means for re-determining the quality.

According to an eighth aspect of the present invention, in the software quality evaluation system according to the first or sixth aspect, the measure generation means generates a plurality of quality enhancement measures as the plurality of measures.

[0023]

Preferred embodiments of the present invention will be described below with reference to the drawings.

Embodiment 1 (1) Configuration of Software Quality Evaluation System FIG. 1 is a diagram illustrating an overall configuration of a software quality evaluation system (hereinafter, referred to as a quality evaluation system) according to the first embodiment.

(1-a) Overall Configuration of Quality Evaluation System First, the overall configuration of the quality evaluation system will be described.

The quality evaluation system includes a test density, which is a ratio of the number of test items to a software scale, and an error detection rate, which is a ratio of the number of errors generated when a test of the number of test items is performed to the software scale. , And a measure plan generator 2 that generates a plurality of measure plans corresponding to a plurality of quality evaluation results expected to be evaluated by the quality evaluate unit 1 in advance. And a storage unit 3 for storing the quality evaluation result evaluated by the quality evaluation unit 1 and the countermeasure plan generated by the countermeasure plan generation unit 2, and the quality evaluation result evaluated by the quality evaluation unit 1 and corresponding thereto And a printer unit 5 that prints various data output from the output unit 4 on paper or the like and provides the data to an operator.

The quality evaluation system includes an input unit 6 (for example, a keyboard or the like) for inputting data or the like necessary for quality evaluation of software by an input operation of an operator, and a guide or the like necessary for the input operation by the input unit 6. A display unit 7 for displaying a result of each process in each unit, and each unit (the input unit 6, the display unit 7,
Quality evaluation unit 1, countermeasure plan generation unit 2, storage unit 3, output unit 4,
And a control unit 8 for controlling the printer unit 5).

Next, the main part of the quality evaluation system will be described in detail.

(1-b) Quality Evaluation Unit 1 The quality evaluation unit 1 calculates the test density, the error detection rate,
When respective reference values are set for the hit rate, which is the ratio of the error detection rate to the test density, a plurality of test densities, a plurality of error detection rates, and a plurality of hit rates are set based on these reference values. And a plurality of quality states in the software, a quality evaluation matrix generation unit 10 for generating a quality evaluation matrix in which the respective quality states are associated with each other.
Having.

The hit rate indicates the number of errors with respect to the test scale, and is obtained from the test density and the error detection rate. The hit ratio is calculated by (error detection ratio) / (test density).

A quality evaluation section 1 generates a quality evaluation corresponding to each quality state, and a calculation section 1 calculates a hit rate from a test density and an error detection rate.
2 and the test density, error detection rate, and hit rate of the software to be evaluated according to the above-described quality evaluation matrix,
A quality determining unit that determines a quality state of the software to be evaluated and determines a quality evaluation corresponding to the quality state among the quality evaluations generated in advance by the quality evaluation generation unit as a quality evaluation result. .

(1-c) Countermeasure Plan Generation Unit 2 The countermeasure plan generation unit 2 generates each of the plurality of countermeasure plans corresponding to the respective quality evaluations generated by the quality evaluation generation unit 11 as the plurality of countermeasure plans.

(1-d) Output Unit 4 The output unit 4 outputs a quality evaluation determined by the quality determining unit 13 as a quality evaluation result.
A countermeasure output unit that outputs a countermeasure corresponding to the quality evaluation determined by the quality determination unit 13 among the countermeasures generated by the countermeasure plan generation unit 2 as a countermeasure plan corresponding to the quality evaluation result. And 15.

(2) Operation of Quality Evaluation System The operation of the quality evaluation system configured as described above will be described below. The operation of the quality evaluation system includes (2)
-1) an operation of generating a quality evaluation matrix, an operation of generating each quality evaluation corresponding to each quality state set in this matrix, an operation of generating each measure plan corresponding to each quality evaluation,
(2-2) There is an operation of performing quality evaluation of software to be evaluated using each data generated in (2-1), and outputting a countermeasure plan corresponding to the quality evaluation result. First, the operation (2-1) will be described.

(2-1) An operation of generating a quality evaluation matrix, an operation of generating each quality evaluation, and an operation of generating each measure by the quality evaluation system according to the first embodiment will be described with reference to a flowchart shown in FIG. Will be explained.

In step S102, the operator inputs, through the input unit 6, a command to generate a quality evaluation matrix. As a result, the input unit 6 sends a message to that effect to the control unit 8. The control unit 8 causes the display unit 7 to display an instruction to instruct the operator to input the reference values of the test density and the error detection rate.

In step S104, the operator inputs the reference value of the test density and the reference value of the error detection rate to the input unit 6. The reference value of the test density and the reference value of the error detection rate are determined based on the scale and characteristics of the software to be evaluated. Then, the control unit 8 receives each reference value input by the input unit 6 and sends it to the calculation unit 12.

In step S106, the calculation unit 12 calculates a reference value of the hit rate from the reference value of the test density and the reference value of the error detection rate. At this time, the reference value of the hit rate is
It is calculated by (reference value of error detection rate) / (reference value of test density). Then, the calculation unit 12 notifies the control unit 8 that the reference value of the hit ratio has been calculated. The control unit 8 sends the reference value of the test density, the reference value of the error detection rate, and the reference value of the hit rate (hereinafter, these reference values are referred to as respective reference values) to the quality evaluation matrix generation unit 10 for the calculation unit 12. Tell them to send. As a result, the calculation unit 12 sends each reference value to the quality evaluation matrix generation unit 10.

In step S 108, the quality evaluation matrix generator 10 generates a quality evaluation matrix based on the reference values sent from the calculator 12.

(Method of Generating Quality Evaluation Matrix) The method of generating the quality evaluation matrix in step S108 will be described in detail with reference to the drawings (FIGS. 3 to 7).

When each reference value is sent from the calculation unit 12, the quality evaluation matrix generation unit 10 divides the area of the error detection rate based on the reference value of the error detection rate as shown in FIG.
It is divided into three stages: below the reference range, above the reference range, and above the reference range. Similarly, based on the test density reference value, the quality evaluation matrix generation unit 10 divides the test density area into three stages of a reference area, a reference area, a reference area, and a reference area, as shown in FIG.
As a result, as shown in FIG. 5, the quality evaluation matrix generation unit 10 generates an error detection rate area divided into three stages,
And the test density area divided into three stages,
A matrix in which nine areas (areas A to I in FIG. 5) are set is generated. Then, as shown in FIG. 6, the quality evaluation matrix generation unit 10 divides the data into the reference area and the outside of the reference area based on the reference value of the hit ratio. Thus, the area A, the area E, and the area I in FIG. 5 are further divided into two. As a result, as shown in FIG. 7, the quality in which 12 areas (each of these areas is referred to as each quality state) is set by a combination of the area of the error detection rate, the area of the test density, and the area of the hit rate. An evaluation matrix is generated.

In the above example, the number of quality states is 12
However, the number of quality states is not limited to 12, and may be another value. Also, the test density area, the error detection rate area, and the hit rate area are not limited to the above-described number of steps.

After the quality evaluation matrix generator 10 generates the quality evaluation matrix in this way, it sends to the controller 8 that the quality evaluation matrix has been generated. The control unit 8 instructs the quality evaluation matrix generation unit 10 to send each quality state of the generated quality evaluation matrix to the quality evaluation generation unit 11.

When each quality state is sent to the quality evaluation generator, in step S110, the quality evaluation generator 11 generates each quality evaluation corresponding to each quality state.

The manner in which the quality evaluation generating section 11 generates each quality evaluation will be described in detail below. That is, when each quality state is sent to the quality evaluation generation unit 11, the quality evaluation generation unit 11
First, the control unit 8 is instructed to send information indicating each quality evaluation (each quality evaluation information) corresponding to each quality state to the quality evaluation generation unit 11. As a result, the control unit 8 causes the display unit 7 to display an instruction to the operator to input each piece of quality evaluation information. When the operator inputs each piece of quality evaluation information, the input unit 6 sends each piece of quality evaluation information to the control unit 8. When each piece of quality evaluation information is sent from the control unit 8 to the quality evaluation generation unit 11, the quality evaluation generation unit 11 generates each quality evaluation corresponding to each quality state based on each piece of quality evaluation information.

(For each quality evaluation)
Each quality evaluation generated in 110 will be described in detail. As the quality evaluation, a detailed quality evaluation (detailed quality evaluation) and, for example, five levels (from the higher quality, ◎, ○, △,
Δ △, ×) are generated. The levels of the quality level are variously set according to specifications and the like, and can be set up to a maximum of 12 levels. Also,
The detailed quality evaluation also evaluates a qualitative prediction of the number of remaining errors that can be predicted not to be detected due to insufficient execution of software by a test. The following detailed quality evaluation is conceptually shown, and it is preferable to generate a more detailed quality evaluation according to an actual software product.

(A) Quality Evaluation 1 for Quality State <1> The detailed quality evaluation is as follows. That is, although the error detection rate is equal to or less than the reference area and the hit rate is within the reference area, the quality cannot be confirmed because the test density has not reached the reference area. It can be said that there is a residual error number. The quality level is △.

(B) Quality Evaluation 2 for Quality State <2> The detailed quality evaluation is as follows. That is, although the error detection rate is equal to or less than the reference area, the hit rate is outside the reference area, and it can be said that the number of errors is large with respect to the test scale and the software scale. It can be said that there is a residual error number. The quality level is △△.

(C) Quality Evaluation 3 for Quality State <3> The detailed quality evaluation is as follows. That is, the error detection rate has reached the reference range despite the lack of test density, and it can be expected that the error detection rate will increase when the test density is set in the reference range. It can be said that the number of remaining errors is large. The quality level is △△.

(D) Quality Evaluation 4 for Quality State <4> The detailed quality evaluation is as follows. That is, although the test density is insufficient, the error detection rate is equal to or higher than the reference range, and it can be said that the number of errors is very large with respect to the test scale and the software scale. It can be said that the number of remaining errors is very large. The quality level is x.

(E) Quality Evaluation 5 for Quality State <5> The detailed quality evaluation is as follows. That is, although both the test density and the error detection rate are in the reference range, the hit rate is outside the reference range, and it can be said that the number of errors is slightly larger than the test scale and the software scale. In addition, it can be said that there is a residual error number. The quality level is △.

(F) Quality Evaluation 6 for Quality State <6> The detailed quality evaluation is as follows. That is, although the test density is in the reference range, the error detection rate is equal to or higher than the reference range, so that the number of errors is large with respect to the test scale and the software scale. It can also be said that the number of remaining errors is large. The quality level is △△.

(G) Quality Evaluation 7 for Quality State <7> The detailed quality evaluation is as follows. That is, although the test density is equal to or higher than the reference area, the error detection rate is equal to or higher than the reference area. Therefore, it can be said that the number of errors is slightly larger than the test scale and the software scale. It can be said that there are some residual errors. The quality level is △.

(F) Quality Evaluation 8 for Quality State <8> The detailed quality evaluation is as follows. That is, since the test density is in the reference range and the error detection rate is equal to or lower than the reference range, it can be said that the quality is close to good. The quality level is ○.

(G) Quality Evaluation 9 for Quality State <9> The detailed quality evaluation is as follows. In other words, the test density and the error detection rate are both in the reference range, and the hit rate is within the reference range. The quality level is ○.

(G) Quality Evaluation 10 for Quality State <10> The detailed quality evaluation is as follows. That is, since the test density is not less than the reference range and the error detection rate is not more than the reference range, it can be said that the quality is good. The quality level is ◎.

(F) Quality Evaluation 11 for Quality State <11> The detailed quality evaluation is as follows. That is, the test density is equal to or higher than the reference range, and the error detection rate is in the reference range, which means that the quality is close to good. The quality level is ○.

(G) Quality Evaluation 12 for Quality State <12> The detailed quality evaluation is as follows. That is, although the error detection rate is equal to or higher than the reference area, the test density is equal to or higher than the reference area, and the hit rate is equal to or higher than the reference area. The quality level is ○.

FIG. 8 shows how each quality state corresponds to each quality level.

When the quality evaluation generation unit 11 informs the control unit 8 that each quality evaluation (the above-described quality evaluations 1 to 12) has been generated, the control unit 8
It instructs each generated quality evaluation to be sent to the measure plan generation unit 2. The quality evaluation generation unit 11 sends each quality evaluation to the measure plan generation unit 2.

In step S112, when each quality evaluation is sent to the measure plan generation unit 2, the measure plan generation unit 2 generates each measure plan corresponding to each quality evaluation of the software (the above-described quality evaluations 1 to 12). I do. The manner in which each measure proposal corresponding to each quality evaluation is generated will be described in detail below.

When each quality evaluation is sent to the countermeasure plan generator 2, the countermeasure generator 2 transmits to the control unit 8 an instruction indicating that information necessary for generating the countermeasure plan is sent to the countermeasure generator 2. The control unit 8 causes the display unit 7 to display a message to instruct the operator to input basic measure information, which is a basic measure.
This basic countermeasure information is a quality enhancement countermeasure for enhancing the quality, and includes, for example, countermeasure proposal items such as a review of a cause of an error, a review of a specification, a review of a test method, and a review. These basic countermeasure information is common regardless of the nature of the software. When the worker inputs basic countermeasure information through the input unit 6, the basic countermeasure information is held in the control unit 8.

Then, the control unit 8 displays the quality evaluation and the basic countermeasure information on the display unit 7, and displays on the display unit 7 an instruction to instruct the operator to select the basic countermeasure information corresponding to the quality evaluation. . When the operator selects the basic countermeasure information corresponding to the quality evaluation, the control unit 8 causes the display unit 7 to display an instruction to input additional countermeasure information which is information for limiting the basic countermeasure information in detail. This additional countermeasure information is countermeasure information that changes depending on the nature of software, the nature of specifications, and the like.

When the operator inputs additional countermeasure information corresponding to the selected basic countermeasure information through the input unit 6, the control unit 8
Sends the basic measure information and the additional measure information corresponding to the quality evaluation to the measure plan generating unit 2. The measure plan generation unit 2 generates a measure plan by compiling basic measure information and additional measure information corresponding to the quality evaluation. The above-described countermeasure plan generation operation is performed for all quality evaluations (quality evaluations 1 to 12).

A specific example will be described below.

For example, the display unit 7 displays basic countermeasure information (review of cause of error, review of specifications, implementation of review, etc.)
Is displayed. Then, the display unit 7 displays, for example, an instruction to select the basic measure information corresponding to the quality evaluation 3. When the operator selects, for example, a review of the cause of the error (a plurality of basic countermeasure information may be selected),
The control unit 8 causes the display unit 7 to display an instruction to input additional countermeasure information regarding the review of the cause of the error. For example, the operator inputs the specific contents of the review of the coding, the review of the test data, and the review of the test environment setting through the input unit 6. As a result, the control unit 8 sends the basic measure information and the additional measure information corresponding to the quality evaluation 3 to the measure plan generating unit 2. The measure plan generator 2 generates a measure plan as follows. That is, the countermeasure generation unit 2 generates a countermeasure so that the review of the cause of the error is a basic content, and the specific content is a review of the coding, a review of the test data, and a review of the test environment setting. The measure proposed for quality evaluation 3 is, for example, "It is necessary to review the causes of errors. The main points of review are review of coding, review of test data, and review of test environment settings. The specific method is.
For the specific method of reviewing test data, see. . . What is the specific method of reviewing the test environment settings? . . . ]. Then, a countermeasure plan corresponding to another quality evaluation is similarly generated. For example, in the case of quality evaluation 4, "review needs to be performed. In conducting the review, the items to be examined are ..., and the specific examination method for these items is ... Is generated.

In this way, the measure plan generator 2 generates each measure plan corresponding to each quality evaluation (the above-described quality evaluations 1 to 12). Then, each measure is a quality enhancement measure for enhancing quality.

When the countermeasure plan generation unit 2 notifies the control unit 8 that each countermeasure plan has been generated, the control unit 8 displays on the display unit 7 a quality evaluation matrix, each quality evaluation, and each countermeasure plan. Display that generation has been completed. Then, the control unit 8 includes matrix data indicating a quality evaluation matrix, quality evaluation data summarizing each quality evaluation, a matrix data name for identifying each measure plan data summarizing each measure plan, a quality evaluation data name, and a measure plan data name. Is displayed on the display unit 7 to instruct the operator to input the.

Then, in step S114, the operator inputs the matrix data name, the quality evaluation data name, and the countermeasure data name through the input unit 6.

As a result, these data names are sent to the control unit 8. The control unit 8 controls the quality evaluation matrix generation unit 1
0, matrix data, quality evaluation data, and countermeasure data are extracted from the quality evaluation generator 11 and the countermeasure generator 2.

Then, in step S116, the control unit 8
Causes the storage unit 3 to store the matrix data, the quality evaluation data, and the measure plan data in association with the matrix data name, the quality evaluation data name, and the measure plan data name, respectively. At this time, the matrix data name, the quality evaluation data name, and the measure proposal data name are associated with each other. Then, each quality state, each quality evaluation, and each measure are associated with each other.

(2-2) The quality evaluation of the software to be evaluated is performed using the quality evaluation matrix, each quality evaluation, and each countermeasure plan generated as described above, and the quality evaluation result and the countermeasure plan are obtained. The output operation will be described with reference to the flowcharts shown in FIGS.

In step S 202, the operator inputs, through the input unit 6, that the quality evaluation of the tested software to be evaluated is to be performed. As a result, the input unit 6 sends the fact to the control unit 8, and the control unit 8 causes the display unit 7 to display an instruction to instruct the operator to input the test density and the error detection rate.

In step S 204, the operator inputs the test density and the error detection rate of the evaluation target software through the input unit 6 according to the display on the display unit 7. Further, an evaluation target data name for identifying the evaluation target data including the test density and the error detection rate and the calculated hit rate is also input. The test density and the error detection rate input by the input unit 6 are sent to the calculation unit 12 via the control unit 8.

In step S206, the calculation unit 12 calculates a hit rate from the test density and the error detection rate. After calculating the hit ratio, the calculation unit 12 informs the control unit 8 of the calculation.

In step S208, the control unit 8 controls the display unit 7 to display that the calculation of the hit ratio has been completed, and to input the test density and the error detection ratio and the calculated hit ratio. Is stored in the storage unit 3 in association with the evaluation target data name.

In step S210, in response to an instruction from the control unit 8, the display unit 7 displays a message requesting the operator to determine whether to continue generating the evaluation target data. When the generation of the evaluation target data is continued, the operator inputs the fact to the input unit 6, and the operations of steps S204, S206, and S208 are performed again by each unit. On the other hand, when the generation of the evaluation target data is not to be continued, the operator inputs the fact to the input unit 6 and the operation of step S212 is performed.

In step S212, the control unit 8 displays on the display unit 7 a command to instruct the worker to input the data name of the matrix data corresponding to the quality evaluation of the evaluation target software. Let it.

In step S 214, the operator inputs a matrix data name using the input unit 6. As a result, the control unit 8 reads matrix data corresponding to the input matrix data name from the storage unit 3 and sends the matrix data to the quality determination unit 13. At the same time, the control unit 8 reads out the quality evaluation data (each of the quality evaluations 1 to 12) of the quality evaluation data name associated with the matrix data name from the storage unit 3 and sends it to the quality determination unit 13. At this time, the control unit 8 holds the matrix data name. The quality determining unit 13 sends the control unit 8 that the matrix data has been received.

In step S 216, the control unit 8 reads one piece of evaluation target data stored in the storage unit 3 and sends the read evaluation target data to the quality determination unit 13. In addition,
Reading is not performed again on the evaluation target data that has already been read.

In step S 218, the quality determining section 13
Is the quality state (each quality state <1> to <12>) from the evaluation target data (test density, error detection rate, and hit rate of the evaluation target software) according to the quality evaluation matrix.
One of them). And the quality determining unit 13
Indicates each quality evaluation of the quality evaluation data (each quality evaluation 1 to 1)
In 2), the quality evaluation (any of the quality evaluations 1 to 12) corresponding to the determined quality state (any of the quality states <1> to <12>) is determined. The quality determination unit 13 notifies the control unit 8 that the quality evaluation of the evaluation target data has been completed.

In step S220, the control unit 8 reads out the countermeasure data of the countermeasure data name corresponding to the held matrix data name from the storage unit 3, and then reads out the countermeasure data of the read countermeasure data. Then, a measure corresponding to the quality evaluation determined by the quality determining unit 13 (any one of the quality evaluations 1 to 12) is determined.

In step S222, the control section 8 controls the quality evaluation output section 14 to output the quality evaluation determined by the quality determination section 13 (any one of the quality evaluations 1 to 12) and to output the countermeasure plan. The unit 15 is controlled to output a measure (one of the measure plans) corresponding to the quality evaluation (one of the quality evaluations 1 to 12). The control unit 8 causes the display unit 7 to display these output results. Then, the control unit 8 instructs the operator to input an evaluation measure result data name for identifying the evaluation measure result data summarizing the quality evaluation and the measure plan.
To be displayed.

In step S224, the operator inputs
Input the evaluation measure result data name.

In step S226, the control unit 8 stores the evaluation measure result data in the storage unit 3 in association with the evaluation measure result data name.

In step S 228, control unit 8 determines whether there is any data to be evaluated that has not been read from storage unit 3. At this time, the evaluation target data to be read is limited to the evaluation target data that can be evaluated by the read quality evaluation matrix.

If there is data to be evaluated that has not been read from the storage unit 3, the operation after step S216 is performed again. On the other hand, all data to be evaluated are stored in the storage unit 3
, The operation of step S230 is performed.

In step S230, the control unit 8 reads out all the evaluation countermeasure result data stored in the storage unit 3, and instructs the printer unit 5 to print all the evaluation countermeasure result data.

In step S232, the printer section 5 prints all the evaluation measure result data. At this time, the printer unit 5 can print the evaluation measure result data for each evaluation measure result data. In addition, as shown in FIG. 11, all the evaluation measure result data can be printed in a table format.

In the prior art, the quality of the software can be evaluated only in two stages of good or bad. In the present embodiment, for example, the quality can be evaluated in 12 stages. For each quality evaluation, the quality level can be divided into, for example, five levels (up to 12 levels).
For example, in the prior art, software a (quality level △) in quality state 1 and software b (quality level x) in quality state 4 both of which have been evaluated as defective are:
It can be seen that there are two quality level differences. And
In the present embodiment, each countermeasure plan corresponding to each quality evaluation is output by the display unit 7 and the printer unit 5.

In the present embodiment, a sufficient quality evaluation can be performed as follows. That is, first, the quality evaluation matrix generation unit 10 generates a quality evaluation matrix based on the respective reference values of the test density, the error detection rate, and the hit rate. In this quality assessment matrix, for a combination of multiple test density zones, multiple error detection rate zones, multiple hit rate zones,
Each of them has a plurality of quality states (quality state <1> to <1
2>).

Then, the quality evaluation generating section 11 generates each quality evaluation (quality evaluations 1 to 12) corresponding to each quality state (quality states <1> to <12>). Then, the quality determination unit 13 determines, based on the test density, the error detection rate, and the hit rate of the software to be evaluated, the software to be evaluated among the quality states (quality states <1> to <12>) of the quality evaluation matrix. To determine which quality state the quality state corresponds to. Then, the quality determination unit 13 determines the determined quality state (any of the quality states <1> to <12>).
Determines which quality evaluation corresponds to each of the quality evaluations (quality evaluations 1 to 12) generated by the quality evaluation generation unit 11.

For this reason, based on the test density, the error detection rate, and the hit rate, a plurality of quality states are set sufficiently large,
If a plurality of quality evaluations corresponding to this are set sufficiently large, it is possible to perform a sufficient quality evaluation taking into account the difference in the degree of software quality in detail.

Further, in this embodiment, a countermeasure plan corresponding to the result of the sufficient quality evaluation performed as described above can be generated and output as follows.

That is, the measure plan generator 2 generates each measure plan corresponding to each quality state (quality states <1> to <12>). For this reason, the measure plan corresponding to the quality evaluation (any of the quality evaluations 1 to 12) determined by the quality determination unit 13 is generated in advance by the measure plan generation unit 2. The determined quality evaluation (any of the quality evaluations 1 to 12) can be output by the quality evaluation output unit 14, and the quality evaluation results (the quality evaluations 1 to 12) can be output by the measure proposal output unit 15. Can be output. As a result, a measure plan based on a sufficient quality evaluation can be promptly provided to the worker, so that a reliable measure can be promptly performed regardless of the worker's experience and skill.

Embodiment 2 In the quality evaluation system according to the second embodiment, with respect to the test density, which is the test result,
A new test density is generated in consideration of test coverage (test coverage).

Generally, when a test based on a predetermined number of test items is performed in software to be evaluated,
It can be expected that a predetermined number of functions (such as branch instructions) in the software will be executed.

However, when a certain function is executed during the test, an error may occur, and the test may end without being able to execute another function that could have been executed. For this reason, when the expected number of functions are not executed (when the test coverage is low) even if the test of the predetermined number of test items is executed, the test expected from the test density is actually executed. Will not be done. In such a case, even if the test density is input and quality evaluation is performed, accurate quality evaluation is not performed.

Therefore, in the present embodiment, in consideration of a case where a test having a low test coverage is performed, the test coverage as well as the test density and the error detection rate are obtained as test results. The true test density is calculated by multiplying the test density by the degree.

The quality evaluation system according to the second embodiment differs from the quality evaluation system according to the first embodiment in the following points.

That is, when the test coverage is input, the calculation unit 12 calculates a new test density by multiplying the test coverage by the test density of the software to be evaluated, and detects an error in the software to be evaluated. The hit ratio is calculated from the ratio and the new test density.

The quality determining unit 13 determines the quality state of the software to be evaluated according to the quality evaluation matrix from the error detection rate of the software to be evaluated, the new test density, and the hit rate. 11
Among the quality evaluations generated by the above, a quality evaluation corresponding to the determined quality state is determined.

The operation of the quality evaluation system according to the present embodiment will be described below.

The operation of generating the quality evaluation matrix, the operation of generating each quality evaluation, and the operation of generating each measure are the same as those described in (2-1) in the first embodiment.

The operation of performing the quality evaluation of the software to be evaluated in consideration of the test coverage and outputting the quality evaluation result and the corresponding measure plan will be described with reference to the flowchart shown in FIG.

In step S302, the input operation of the input unit 6 sends to the control unit 8 a message indicating that the quality of the tested software to be evaluated is to be evaluated. At this time, the input unit 6
Is input by taking into account the test coverage. As a result, the control unit 8 causes the display unit 7 to display an instruction to instruct the operator to input the test density, the error detection rate, and the test coverage.

In step S 304, the operator inputs the test density, the error detection rate, and the test coverage of the evaluation target software through the input unit 6 according to the display on the display unit 7. Further, the operator also inputs an evaluation target data name for identifying the evaluation target data including the test density and the error detection rate and the calculated hit rate. The test density, the error detection rate, and the test coverage input by the input unit 6 are sent to the calculation unit 12 via the control unit 8.

In step S306, the calculation unit 12 calculates a new test density by multiplying the test density by the test coverage. Then, the calculation unit 12 calculates a hit rate from the new test density and the error detection rate. When the hit ratio is calculated, the calculation unit 12 notifies the control unit 8 of the calculation.

In step S308, the control unit 8 displays on the display unit 7 that the calculation of the hit ratio has been completed, and stores the evaluation target data in the storage unit 3 in association with the evaluation target data name.

In step S310, in response to an instruction from the control unit 8, the display unit 7 displays a message requesting the operator to determine whether to continue generating the evaluation target data. When the generation of the evaluation target data is continued, the operator inputs the fact to the input unit 6, and the operations of steps S304, S306, and S308 are performed again by each unit. On the other hand, when the generation of the evaluation target data is not to be continued, the operator inputs the fact to the input unit 6 and the operation of step S212 is performed.

The following steps are the same as the steps S212 to S232 shown in FIGS.

In this embodiment, for example, when evaluating software having a test coverage of 85%, a value obtained by multiplying the test density by 0.85 is set as a new test density. As a result, the degree of test execution can be made uniform, so that quality evaluation can be performed using the quality evaluation matrix described in the first embodiment. For example, as shown in FIG. 13, when the test coverage is not considered, the software quality is determined from the positions of X, Y, and Z. The software quality is determined from the position as shown.

In the present embodiment, the hit ratio can be calculated from the test density without considering the test coverage as in the first embodiment. In this case, in step S302, in step S20 of the first embodiment,
The same input operation as in step 2 may be performed.

In the present embodiment, the calculating unit 12
A new test density is calculated from the test density and the test coverage as test results, and a hit rate is calculated from the new test density and the error detection rate as the test result. Then, the quality determination unit 13 determines a quality state (quality states <1> to <12>) from these new test densities, error detection rates, and hit rates.
Is determined, and the corresponding quality evaluation (any of the quality evaluations 1 to 12) is determined. Therefore, even if the expected number of functions are not executed by executing the test of the predetermined number of test items, the test density is multiplied by the test coverage to obtain a test corresponding to the number of functions actually executed. Density can be calculated. As a result, it is possible to accurately evaluate the quality of software having low test coverage.

Embodiment 3 In the quality evaluation system according to the third embodiment, when a test of a predetermined number of test items is not performed on software to be evaluated and a test of a predetermined number of test items is completed based on the number of errors up to the middle of the test. Is predicted, and the prediction error detection rate calculated from the prediction error number is set as the error detection rate when the test of a predetermined number of test items is completed.

The quality evaluation system according to the third embodiment differs from the quality evaluation system according to the first embodiment in the following points.

That is, the calculation unit 12 calculates a growth prediction curve of the number of errors from the relationship between the number of errors and the test progress up to the intermediate stage of the test performed on the software to be evaluated. The prediction error detection rate is calculated from the number of errors after the end of the test predicted based on the
The hit rate is calculated from the prediction error detection rate and the test density after the end of the test.

The calculation unit 12 stores a plurality of growth prediction curve expressions. Then, the calculation unit 12 determines an appropriate growth prediction curve expression among the plurality of growth prediction curve expressions from the relationship between the number of errors and the test progress up to the test intermediate stage.

The quality determination unit 13 determines the quality state of the software to be evaluated according to the quality evaluation matrix from the test density of the software to be evaluated after the test, the prediction error detection rate and the hit rate, and Among the quality evaluations generated by the evaluation generation unit 11, a quality evaluation corresponding to the determined quality state is determined.

The operation of the quality evaluation system according to the present embodiment will be described below.

The operation of generating the quality evaluation matrix, the operation of generating each quality evaluation, and the operation of generating each measure are the same as those described in (2-1) of the first embodiment.

The operation of performing the quality evaluation of the evaluation target software in consideration of the growth prediction curve and outputting the quality evaluation result and the corresponding measure plan will be described with reference to the flowchart shown in FIG.

In step S402, the operator inputs, through the input unit 6, an instruction to evaluate the quality of the software to be evaluated. At this time, the input unit 6 inputs that the growth prediction curve is to be considered. As a result, the control unit 8 displays on the display unit 7 a command to instruct the operator to input the test density, the number of errors for the test progress up to the intermediate stage, and the like.

In step S404, the operator inputs the test density of the software to be evaluated and the number of errors with respect to the test progress in the intermediate stage according to the display on the display unit 7.
Enter by. At this time, the software test density is a test density corresponding to the planned number of test items. The number of errors with respect to the test progress in the intermediate stage is a plurality of errors (accumulated errors) measured in the middle of the planned number of test items. Therefore, the operator inputs the number of errors and the number of test items corresponding to the number of errors.

The time required to execute each test item number is also input in association with each test item number. Then, an evaluation target data name for identifying the evaluation target data including the test density and the error detection rate and the calculated hit rate is also input. The test density input by the input unit 6, the number of errors in the intermediate stage and the number of test items corresponding thereto, the time corresponding to the number of test items, (hereinafter, the number of errors in the intermediate stage and the test item corresponding thereto) The number and the time corresponding to each test item number are referred to as error number data in the intermediate stage) are sent to the calculation unit 12 via the control unit 8.

In step S406, the calculation unit 12 calculates a growth prediction curve of the number of errors from the data of the number of errors in the intermediate stage of the test.

The method of calculating the growth prediction curve will be described in detail below. The calculation unit 12 plots the number of errors corresponding to each test item number up to the middle of the test of the planned number of test items in relation to time. When the plots are completed, the calculation unit 12 checks which of the plurality of growth prediction curve equations fits these plots. As the growth prediction curve formula, for example, there is a growth prediction curve formula based on an exponential NHPP model as shown in Expression 1.

[0128]

F (t) = a * (1−exp (−b * t)) where F (t) is the number of errors, t is time, and a and b are coefficients. By changing the coefficients for the curve equations, a growth prediction curve equation and the coefficient that best fit the relationship between the number of errors plotted and time are determined. Thus, a growth prediction curve can be calculated from the relationship between the number of errors and the test progress up to the intermediate stage of the test.

In step S408, the calculation unit 12 calculates the number of errors after the test based on the calculated growth prediction curve, and calculates the prediction error detection rate from the number of errors.
At this time, the calculation unit 12 determines the time required to execute the planned number of test items from the relationship of the number of test items up to the test intermediate stage, and determines the number of errors corresponding to this time based on the growth prediction curve. It is determined. The calculating unit 12 calculates a hit rate from the prediction error detection rate and the test density. When the hit ratio is calculated, the calculation unit 12 notifies the control unit 8 of the calculation.

In step S410, the control unit 8 controls the display unit 7 to display that the calculation of the hit ratio is completed, and stores the evaluation target data in the storage unit 3 in association with the evaluation target data name. .

In step S 412, in response to an instruction from the control unit 8, the display unit 7 displays a message requesting the operator to determine whether to continue generating the evaluation target data. When the generation of the evaluation target data is continued, the operator inputs the fact to the input unit 6 and the operations of steps S404, S406, S408, and S410 are performed again by each unit. On the other hand, when the generation of the evaluation target data is not to be continued, the operator inputs the fact to the input unit 6 and the operation of step S212 is performed.

The following steps are the same as the steps S212 to S232 shown in FIGS.

In this embodiment, the hit ratio can be calculated without considering the growth prediction curve as in the first embodiment. In this case, step S4
In 02, the same input operation as in step S202 of the first embodiment may be performed.

In the present embodiment, the calculation unit 12
A growth prediction curve for the number of errors is calculated from the relationship of the number of errors to the test progress up to the middle stage of the test with the planned number of test items. For this reason, the number of errors when the number of test items is completed can be predicted from the growth prediction curve without performing all tests for the number of test items planned. Then, the calculation unit 12 calculates a prediction error detection rate after the end of the test from the number of errors, and calculates a hit rate from the prediction error detection rate and the test density. And the quality determining unit 13
Determines the quality state from these test densities, prediction error detection rates, and hit rates, and determines the corresponding quality evaluations.

Therefore, it is not necessary to perform all the planned number of test items in order to obtain the error detection rate after the end of the test. Therefore, the quality evaluation of the software to be evaluated can be known in a short period of time, and the countermeasure can be taken quickly.

Embodiment 4 The quality evaluation unit according to the present embodiment includes a quality re-evaluation unit 20 that evaluates again the quality evaluation evaluated by the quality evaluation unit according to the first embodiment.

In general, when the quality of the software to be evaluated is evaluated by a single quality evaluation, the quality cannot be accurately known from the test results under insufficient testing (for example, There are quality evaluations 1, 2, and 3) among the quality evaluations. Further, even if the test density has reached a predetermined criterion, if the number of errors is large relative to the number of test items, there is a high possibility that the number of errors has not converged yet. , 6). In such a case, it can be said that the software execution by the test is insufficient, so it is necessary to perform additional tests and re-evaluate the quality based on the results.

For this reason, in the present embodiment, when it is expected that the evaluated quality is not accurate among the quality evaluations evaluated by the quality evaluation unit 1 (the quality evaluation determined by the quality determination unit 13), When it is desired to know the quality more accurately, the quality re-evaluation unit 20 can perform the quality evaluation again.

(4-1) Configuration of Quality Evaluation System FIG. 15 is a diagram showing a configuration of a software quality evaluation system according to the fourth embodiment. 15, the same components as those in FIG. 1 are denoted by the same reference numerals, and the description thereof will be omitted.

In this embodiment, the quality evaluation section 1 includes a quality re-evaluation section 20.

The quality re-evaluation unit 20 includes a remaining error number calculation method generation unit 21 that generates a method for calculating the number of remaining errors that can be predicted not to be detected due to insufficient execution of software by a test for each quality evaluation result. An additional test method generating unit 22 for generating an additional test method for each quality evaluation result.

At this time, the remaining error number calculation method generation unit 21 generates each remaining error number calculation method corresponding to each of the quality evaluations 1 to 12 (quality evaluation results). In addition, the additional test method generation unit 22 performs each quality evaluation (quality evaluations 1 to 1).
Each additional test method corresponding to 2) is generated.

The quality re-evaluation unit 20 performs re-evaluation information for determining a method of calculating the number of remaining errors and an additional test method corresponding to the quality evaluation result evaluated by the quality evaluation unit 1 for the software to be evaluated. It has a decision unit 23.

At this time, the re-evaluation information determination unit 23 calculates the remaining error number calculation method corresponding to the quality evaluation (any of the quality evaluations 1 to 12) determined by the quality determination unit 13 of the quality evaluation unit 1 and adds Determine the test method.

The quality re-evaluation section 20 has a remaining error number calculation section 24 for calculating the number of remaining errors according to the remaining error number calculation method determined by the re-evaluation information determination section 23.

The quality re-evaluation unit 20 performs a test based on the number of remaining errors calculated by the number of remaining errors calculation unit 22 and the additional test method determined by the re-evaluation information determination unit 23. The recalculation unit 25 recalculates the test density and the error detection rate of the software to be evaluated in consideration of the result of the execution of the software to be evaluated. And a quality re-determining unit 26 for re-determining the quality.

At this time, the recalculation unit 25 also recalculates the hit rate from the test density and the error detection rate of the software to be evaluated. The quality redetermining unit 26 redetermines the quality state and quality evaluation of the software to be evaluated from the recalculated test density, error detection rate, and hit rate according to the quality evaluation matrix.

(4-2) Operation of Quality Evaluation System The operation of the quality evaluation system includes (4-2-1) a generation operation of a residual error number calculation method by the residual error number calculation method generation unit 21 and an additional test method generation. An operation of generating an additional test method by the unit 23, and (4-2-2) an operation of providing an additional test information (the number of remaining errors and an additional test method) necessary for the quality re-evaluation of the software to be evaluated to the operator; (4-2-3) There is an operation of re-evaluating the quality of the software to be evaluated using the result of the additional test, and outputting the re-evaluation result and a countermeasure plan. First, (4-2-1) will be described.

(4-2-1) The generation of the residual error number calculation method by the residual error number calculation method generation unit 21 and the generation of the additional test method by the additional test method generation unit 23 will be described with reference to the flowchart shown in FIG. explain.

In step S502, the operator inputs, through the input unit 6, an instruction to generate a remaining error count calculation method and an additional test method. As a result, the input unit 6 sends a message to that effect to the control unit 8. As a result, the control unit 8 causes the display unit 7 to display an instruction to the operator to input a data name of desired quality evaluation data.

In step S504, the operator inputs the data name of the quality evaluation data including the quality evaluations (quality evaluations 1 to 12) through the input unit 6. As a result, the control unit 8
Reads out the quality evaluation data of the input quality evaluation data name from the storage unit 3. Then, the control unit 8 causes the display unit 7 to display a command to instruct the operator to input information indicating each remaining error count calculation method corresponding to each quality evaluation (quality evaluations 1 to 12).

In step S506, the input unit 6 is operated to input information indicating each quality evaluation and the corresponding remaining error number calculation method. As a result, the control unit 8 receives information indicating each remaining error number calculation method, and sends the information to the remaining error number calculation method generation unit 21 together with the quality evaluation data.

In step S508, the remaining error number calculation method generation unit 21 generates each remaining error number calculation method corresponding to each quality evaluation based on the transmitted information.

(Each residual error number calculation method) Step S
An example of a method of calculating the number of remaining errors generated in 508 will be described below. The number of remaining errors for each quality evaluation is calculated based on the following calculation formula.

The standard test scale described below is determined based on the specifications, and the standard error detection rate and the standard test density are determined based on the respective reference values.

(A) Method of calculating remaining error number for quality evaluation 1 (residual error number calculating formula) 1 This is calculated according to the following formula.

Number of residual errors = (standard error detection rate−error detection rate of test result) * (standard test scale) (B) Number of remaining errors calculation method for quality evaluation 2 (residual error number calculation formula) 2 Is calculated.

Number of remaining errors = (number of errors in test result) /
((Test density of test result) / (Standard test density))-
(Number of errors in test result) (C) Method for calculating remaining error number for quality evaluation 3 (residual error number calculating formula) 3 Calculated according to the following formula.

Number of remaining errors = (number of errors in test result) /
((Test density of test result) / (Standard test density))-
(Number of errors in test result) (D) Method of calculating remaining error number for quality evaluation 4 (residual error number calculating formula) 4 Calculated according to the following formula.

Number of remaining errors = (number of errors in test result) /
((Test density of test result) / (Standard test density))-
(Number of errors in test result) (E) Method of calculating remaining error number for quality evaluation 5 (residual error number calculating formula) 5 Calculated according to the following formula.

Number of remaining errors = (number of errors in test result) /
(Error detection convergence degree)-(Number of errors in test result) (F) Method of calculating remaining error number for quality evaluation 6 (Remaining error number calculating formula) 6 Calculated according to the following equation.

Number of remaining errors = (number of errors in test result) /
(Error detection convergence degree)-(Number of errors in test result) (G) Method for calculating remaining error number for quality evaluation 7 (Remaining error number calculation formula) 7 Calculated according to the following equation.

Number of remaining errors = (number of errors in test result) /
(Error detection convergence degree)-(Number of errors in test result) After the remaining error number calculation method generation unit 21 generates each remaining error number calculation method, the remaining error number calculation method generation unit 21 Tell 8 to that effect. The control unit 8
The display unit 7 displays an instruction to instruct the operator to input additional test information corresponding to each quality evaluation. The additional test information is, for example, a test item.

In step S510, the input unit 6
Each additional test information corresponding to each quality evaluation (quality evaluations 1 to 12) is input. As a result, the control unit 8 receives the additional test information and sends the information to the additional test method generation unit 23 together with the quality evaluation data (quality evaluations 1 to 12).

In step S512, the additional test method generator 23 generates each additional test method corresponding to each quality evaluation (quality evaluations 1 to 12) based on the transmitted additional test information.

The generation of this additional test method will be described in detail below. The additional test method generation unit 23 previously corresponds to each test item and each specific additional test method (for example, executing software based on a predetermined test environment using predetermined test data). A plurality of attached tables are stored. Each table corresponds to the scale of the software and the like. Then, the additional test method generation unit 23 determines a table for generating an additional test method based on the scale of the software to be evaluated. Then, the additional test method generation unit 23 refers to this table and refers to the additional test information (quality evaluation 1) transmitted from the control unit 8.
Test items) and the corresponding additional test method. Then, the additional test method generation unit 23 associates the determined additional test method with the quality evaluation 1. In this way, the additional test method generator 23 generates an additional test method corresponding to each quality evaluation (quality evaluations 1 to 12).

When the additional test methods corresponding to all the quality evaluations have been generated, the additional test method generator 23 informs the controller 8 that each additional test method has been generated.

In step S514, the control unit 8 transmits the remaining error number calculation method data summarizing the remaining error number calculation methods (residual error number calculation methods 1 to 12) and the additional test method data summarizing the respective additional test methods. Is stored in the storage unit 3 in association with the quality evaluation data. At this time, each quality evaluation is associated with each remaining error number calculation method and each additional test method.

(4-2-2) Next, when re-evaluating the software of the quality evaluation determined by the quality determining unit 13, an additional test is performed, and the re-quality evaluation is performed in consideration of the result. There is a need to.

The operation of determining and outputting the test information necessary for the additional test will be described below with reference to the flowcharts shown in FIGS.

In step S602, the operator sets the input unit 6
Is input to generate test information necessary for re-evaluating the quality of the evaluation target software. As a result, the control unit 8 prompts the operator to input the evaluation target data name, the quality evaluation data name used in the first quality evaluation, and the evaluation measure result data name generated in the first quality evaluation. The instruction to display the instruction is displayed on the display unit 7.

At step S604, the operator sets the input unit 6
Input the name of the data to be evaluated, the name of the quality evaluation data, and the name of the evaluation measure result data.

In step S606, the control section 8 reads out the data to be evaluated, the quality evaluation data, and the evaluation countermeasure result data, and also reads out the remaining error number calculation method data and the additional test method data corresponding to the quality evaluation data. Then, the control unit 8 transmits these data (evaluation countermeasure result data, remaining error number calculation method data, and additional test method data).
To the re-evaluation information determination unit 23.

In step S608, the re-evaluation information determination unit 23 evaluates the quality of the evaluation result data (quality evaluation 1 to quality evaluation 1).
12) and a method of calculating the number of remaining errors (any of the methods of calculating the number of remaining errors 1 to 12) and an additional test method. Then, the reevaluation information determination unit 23 notifies the control unit 8 that these pieces of information have been determined. The control unit 8
The data to be evaluated and the determined remaining error number calculation method are sent to the remaining error number calculation unit 22.

In step S610, the remaining error number calculation section 22 calculates the number of remaining errors from the evaluation target data and the determined remaining error number calculation method. Then, the remaining error number calculation unit 22 sends the calculated remaining error number to the control unit 8. The control unit sends the calculated number of remaining errors and the determined additional test method to the output unit 4 and instructs to output the information.

In step S614, the output unit 4 outputs additional test data summarizing the determined additional test method and the calculated number of remaining errors. The output unit 4 notifies the control unit 8 that the additional test data has been output. Control unit 8
Causes the display unit 7 to display the additional test data output from the output unit 4. The control unit 8 causes the display unit 7 to display an instruction to input an additional test data name to the operator.

At step S616, the operator sets the input
Input the additional test data name.

In step S618, the control unit 8 stores the output additional test data in the storage unit 3 in association with the additional test data name.

In step S620, in response to an instruction from control unit 8, display unit 7 displays a message requesting the operator to determine whether to continue generating the additional test data. In the case where the generation of the additional test data is continued, the operator inputs the fact to the input unit 6, and the operation after step S604 is performed again by each unit. on the other hand,
When the generation of the additional test data is not to be continued, the operator inputs the fact to the input unit 6 to thereby execute Step S62.
Operation 2 is performed.

In step S622, the control section 8 reads out all the additional test data stored in the storage section 3 and causes the printer section 5 to print. The control unit 8 associates each evaluation measure result data with each additional test data, and
By sending these data to the printer unit 5, the printer unit 5 can print as shown in FIG.

In this way, the test information (the number of remaining errors, the additional test method) necessary for the quality re-evaluation can be provided to the operator. Then, the operator performs an additional test on the software to be evaluated again based on the test information.

(4-2-3) The operation of re-determining the quality of the evaluation target software using the result of the additional test and outputting the re-determined quality and the countermeasure plan is shown in FIG.
This will be described below with reference to the flowchart shown in FIG.

At step S 702, the operator sets the input unit 6
Input that the quality is to be re-evaluated. As a result, the control unit 8 determines the result of the additional test (test density, error detection rate)
And instruct the operator to input the data name of the data to be evaluated (test density, error detection rate, hit rate) used in the first quality evaluation, and the data name of the re-evaluation object to be generated from this. Is displayed on the display unit 7.

In step S704, the operator sets the input unit 6
As a result, the name of the evaluation target data and the name of the reevaluation target data are input as a result of the additional test. As a result, as a result of the additional test, the evaluation target data name and the reevaluation target data name are sent to the control unit 8. As a result, the control unit 8 reads out the evaluation target data of the input evaluation target data name from the storage unit 3.
Then, the control unit 8 sends the result of the additional test and the data to be evaluated to the recalculation unit 25, and instructs the recalculation unit 25 to calculate the test density, the error detection rate, and the hit rate to be reevaluated.

At step S706, the recalculation unit 25
Data sent (results of additional tests, data to be evaluated)
The test density, error detection rate, and hit rate to be re-evaluated are calculated based on Then, the recalculation unit 25 notifies the control unit 8 that these data have been recalculated, and sends the recalculated data.

In step 708, the control unit 8 associates the re-evaluation target data summarizing the test density, the error detection rate, and the hit rate re-calculated by the re-calculation unit 25 with the above-described re-evaluation target data name and stores it in the storage unit. Store it in 3.

[0187] In step S710, in response to an instruction from the control unit 8, the display unit 7 displays a message requesting the operator to determine whether or not to continue generating the data to be re-evaluated. When the generation of the reevaluation target data is continued, the operator inputs the fact to the input unit 6, and the operations of steps S704, S706, and S708 are performed again by each unit. On the other hand, when the generation of the re-evaluation target data is not to be continued, the operator inputs the fact to the input unit 6 and the operation of step S712 is performed.

In step S712, the control unit 8 inputs to the worker the data name of the matrix data (same as the matrix data used for the first quality evaluation) corresponding to the quality evaluation of the software to be re-evaluated. Is displayed on the display unit 7 to indicate to the operator.

[0189] In step S714, the operator inputs the matrix data name using the input unit 6. As a result, the control unit 8 reads matrix data corresponding to the input matrix data name from the storage unit 3 and sends the matrix data to the quality determination unit 13. At the same time, the control unit 8 reads out the quality evaluation data of the quality evaluation data name associated with the matrix data name from the storage unit 3, and reads out the quality re-determination unit 2
Send to 6. At this time, the control unit 8 holds the matrix data name. The quality redetermining unit 26 sends the fact that the matrix data has been received to the control unit 8.

In step S716, the control unit 8 reads one piece of data to be re-evaluated stored in the storage unit 3 and sends the read-out data to be re-evaluated to the quality re-determining unit 26.
The re-evaluation target data that has already been read is not read again.

At step S718, the quality re-determining section 26
Re-determines the quality state from the reevaluation target data (test density, error detection rate, and hit rate of the reevaluation target software) according to the quality evaluation matrix. Then, the quality redetermining unit 13 redetermines the quality evaluation corresponding to the redetermined quality state among the quality evaluations of the quality evaluation data. The quality re-determining unit 26 notifies the control unit 8 that the quality evaluation of the data to be re-evaluated has been completed.

In step S720, the control unit 8 reads out the measure plan data of the measure plan data name corresponding to the held matrix data name from the storage unit 3, and then, among the measure plans of the read measure plan data, , Quality redetermining unit 13
To determine the quality evaluation determined by and the corresponding countermeasures.

In step S722, the control unit 8 controls the quality evaluation output unit 14 to output the quality evaluation determined by the quality redetermining unit 13, and controls the measure proposal output unit 15 to respond to the quality evaluation. And output the countermeasures to take.
The control unit 8 causes the display unit 7 to display these output results.
Then, the control unit 8 causes the display unit 8 to display an instruction to instruct the operator to input a re-evaluation measure result data name for identifying the re-evaluation measure result data indicating the quality evaluation and the measure plan. .

In step S724, the operator sets the input unit 6
Input the name of the re-evaluation measure result data.

In step S726, control unit 8 causes storage unit 3 to store the reevaluation countermeasure result data in association with the reevaluation countermeasure result data name.

In step S728, control unit 8 determines whether there is any data to be re-evaluated and not read from storage unit 3. At this time, the re-evaluation target data to be read is limited to the re-evaluation target data that can be evaluated by the read quality evaluation matrix.

If there is data to be re-evaluated that has not been read from the storage unit 3, the operation after step S716 is performed again. On the other hand, when all the reevaluation target data has been read from the storage unit 3, the operation of step S730 is performed.

In step S730, the control section 8 reads out all the reevaluation countermeasure result data stored in the storage section 3 and instructs the printer section 5 to print all the reevaluation countermeasure result data.

In step S732, the printer unit 5 prints all the re-evaluation measure result data. At this time, the printer unit 5 can also print the reevaluation measure result data for each reevaluation measure result data. In addition, all the re-evaluation countermeasure result data can be printed in a table format.

In this embodiment, re-quality evaluation is performed for software for which the quality could not be completely determined by the evaluation by the quality evaluation unit due to lack of testing or the like.
Quality can be completely determined.

In the present embodiment, at the time of the additional test, the test coverage may be measured, and the data to be re-evaluated may be generated in consideration of the test coverage. Further, at the time of the additional test, the re-evaluation target data may be generated in consideration of the growth prediction curve without performing all scheduled tests.

[0202]

As described above, according to the first aspect of the present invention, software quality is evaluated based on the test density and the error detection rate. At this time, for example, by making the quality evaluation correspond in detail according to the degree of test density and the degree of error detection rate of the software to be evaluated,
Sufficient quality evaluation can be performed.

Further, the measure plan generating means generates in advance a measure plan corresponding to the quality evaluation result expected to be evaluated by the quality evaluation means. For this reason, since the countermeasure plan corresponding to the quality evaluation result evaluated by the quality evaluation unit has been generated in advance by the countermeasure plan generation unit, the output unit may output the countermeasure plan corresponding to the quality evaluation result. it can. As a result, it is possible to promptly provide workers with measures based on sufficient quality evaluation,
Regardless of skill, unified measures will be taken.

According to the second aspect of the invention, the quality evaluation means can reliably perform a sufficient quality evaluation as follows. That is, first, the quality evaluation matrix generation means generates a quality evaluation matrix based on the respective reference values of the test density, the error detection rate, and the hit rate. Then, the quality evaluation generating means generates each quality evaluation corresponding to each quality state. Then, the quality determining means includes a test density of the software to be evaluated, an error detection rate,
From the hit rates, it is determined which of the quality states of the quality evaluation matrix corresponds to the quality state of the evaluation target software. And the quality determining means is
It decides which quality evaluation the determined quality state corresponds to among the quality evaluations generated by the quality evaluation generating means.

Therefore, based on the test density, the error detection rate, and the hit rate, a plurality of quality states are set sufficiently large,
If a plurality of quality evaluations corresponding to this are set sufficiently large, it is possible to perform a sufficient quality evaluation taking into account the difference in the degree of software quality in detail.

According to the third aspect of the present invention, the measure generating means generates each measure corresponding to each quality evaluation according to the second aspect. Therefore, a countermeasure plan corresponding to the quality evaluation determined by the quality determining unit has been generated in advance by the countermeasure plan generating unit. Then, the determined quality evaluation can be output by the quality evaluation output means, and the measure proposal corresponding to the quality evaluation result can be output by the measure proposal output means. As a result, a countermeasure based on a sufficient quality evaluation can be promptly provided to the worker, so that a unified countermeasure is performed regardless of the experience, skill, etc. of the worker.

According to the fourth aspect of the present invention, the test of a predetermined number of test items is executed, and even if the expected predetermined number of functions cannot be executed, the test density is multiplied by the test coverage. The test density corresponding to the number of functions actually executed can be calculated. As a result, it is possible to accurately evaluate the quality of software having low test coverage.

According to the fifth aspect of the present invention, the number of errors when the number of test items is completed can be predicted from the growth prediction curve without performing all tests for the number of test items planned. Then, the error detection rate after the end of the test is calculated from the number of errors, and the hit rate is calculated from the predicted error detection rate and the test density. And the quality determining means is
From these new test densities, error detection rates and hit rates, a quality state is determined and a corresponding quality evaluation is determined. Therefore, it is not necessary to perform all the planned number of test items in order to obtain the error detection rate after the end of the test.
Therefore, the quality evaluation of the software to be evaluated can be known in a short period of time, and the countermeasure can be taken quickly.

[0209] According to the invention described in claim 6, the quality evaluation means has quality re-evaluation means for re-evaluating the evaluated quality evaluation result. Therefore, even for software to be evaluated whose quality could not be completely determined due to insufficient testing or the like, the software to be evaluated can be reevaluated in consideration of, for example, the result of an additional test.

According to the invention described in claim 7, the quality reevaluating means can reevaluate the quality of the software to be evaluated as follows.

The remaining error number calculation method generating means generates a method for calculating the number of remaining errors for each quality evaluation result. Also,
The additional test method generation means generates an additional test method for each quality evaluation result. As a result, each quality evaluation result corresponds to each remaining error number calculation method and each additional test method.

Then, the re-evaluation information determining means determines a remaining error number calculation method and an additional test method corresponding to the evaluated quality evaluation result for the software to be evaluated. Then, the remaining error number calculating means calculates the number of remaining errors according to the determined method of calculating the number of remaining errors. Therefore, the number of remaining errors and the additional test method, which are information of the additional test necessary for the re-evaluation, are notified to the operator.

Further, the quality re-determining means determines a new test density and a new error rate detection rate calculated in consideration of a result of causing the software to be evaluated to execute a test based on the additional test information. Redetermine the quality of the software being evaluated. For this reason, re-quality evaluation is performed for software for which the quality could not be completely determined by the evaluation by the quality evaluation means due to insufficient test density or the like, so that the quality can be completely determined.

According to the invention described in claim 8, each measure is a measure for enhancing quality. As a result, the operator can quickly take measures to enhance the quality.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration of a software quality evaluation system according to a first embodiment.

FIG. 2 is a flowchart illustrating an operation of the software quality evaluation system according to the first embodiment;

FIG. 3 is a diagram for explaining an operation by a quality evaluation matrix generation unit.

FIG. 4 is a diagram for explaining an operation by a quality evaluation matrix generation unit.

FIG. 5 is a diagram for explaining an operation by a quality evaluation matrix generation unit.

FIG. 6 is a diagram for explaining an operation by a quality evaluation matrix generation unit.

FIG. 7 is a diagram for explaining an operation by a quality evaluation matrix generation unit.

FIG. 8 is a diagram for explaining an operation by a quality evaluation matrix generation unit.

FIG. 9 is a flowchart illustrating an operation of the software quality evaluation system according to the first embodiment;

FIG. 10 is a flowchart illustrating an operation of the software quality evaluation system according to the first embodiment;

FIG. 11 is a diagram illustrating a result of printing by the printer unit according to the first embodiment;

FIG. 12 is a flowchart illustrating an operation of the software quality evaluation system according to the second embodiment;

FIG. 13 is a diagram showing a specific example of the operation of the software quality evaluation system according to the second embodiment.

FIG. 14 is a flowchart illustrating an operation of the software quality evaluation system according to the third embodiment;

FIG. 15 is a diagram showing a configuration of a software quality evaluation system according to a fourth embodiment.

FIG. 16 is a flowchart illustrating an operation of the software quality evaluation system according to the fourth embodiment;

FIG. 17 is a flowchart illustrating an operation of the software quality evaluation system according to the fourth embodiment;

FIG. 18 is a flowchart showing an operation of the software quality evaluation system according to the fourth embodiment.

FIG. 19 is a diagram illustrating a print result by a printer unit according to the fourth embodiment.

FIG. 20 is a flowchart illustrating an operation of the software quality evaluation system according to the fourth embodiment;

FIG. 21 is a flowchart illustrating an operation of the software quality evaluation system according to the fourth embodiment;

[Explanation of symbols]

1 quality evaluation section, 2 measure plan generation section, 3 storage section, 4
Output unit, 5 printer unit, 6 input unit, 7 display unit, 8
Control unit, 10 Quality evaluation matrix generation unit, 11
Quality evaluation generation unit, 12 calculation unit, 13 quality determination unit, 2
0 quality re-evaluation unit, 21 remaining error number calculation method generation unit,
22 additional test method generation unit, 23 re-evaluation information determination unit,
24 remaining error number calculator, 25 recalculator, 26 quality redeterminer.

[Procedure amendment]

[Submission date] February 5, 1999

[Procedure amendment 1]

[Document name to be amended] Statement

[Correction target item name] Claims

[Correction method] Change

[Correction contents]

[Claims]

[Procedure amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0015

[Correction method] Change

[Correction contents]

[0015]

To achieve the above object, a software quality evaluation system according to a first aspect of the present invention includes a test density, which is a ratio of the number of test items to a scale of software used in a computer; Based on a two-dimensional quality state represented by an error detection rate, which is a ratio of the number of errors generated when the software executes the test of the number of test items with respect to the scale of the software, the one-dimensional Quality evaluation means for performing effective quality evaluation, measure proposal generation means for previously generating a plurality of measure plans corresponding to a plurality of expected quality evaluation results, and a quality evaluation result evaluated by the quality evaluation means and Output means for outputting a corresponding measure plan.

[Procedure amendment 3]

[Document name to be amended] Statement

[Correction target item name] 0016

[Correction method] Change

[Correction contents]

According to a second aspect of the present invention, in the software quality evaluation system according to the first aspect, the quality evaluation means includes:
When respective reference values are set for the test density, the error detection rate, and a hit rate that is a ratio of the error detection rate to the test density , based on these reference values, the test density and the error detection rate are determined. And two-dimensionally represented by
Classify the quality state to be generated and generate a quality evaluation matrix,
Said test density, the set of error detection rate and the hit ratio
Correspondence the segmented quality status of the software against
A quality evaluation matrix generating means for generating that quality evaluation matrix by classification of the quality evaluation matrix
A quality evaluation generating means for generating said each quality evaluation corresponding to each quality status of the software defined,
The calculating means for calculating the hit rate from the test density and the error detection rate of the software to be evaluated and the test density, the error detection rate and the hit rate of the software to be evaluated, and the quality of the software to be evaluated according to the quality evaluation matrix. And a quality determining means for determining a state and determining, as the quality evaluation result, a quality evaluation corresponding to the quality state among the quality evaluations generated by the quality evaluation generating means.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0018

[Correction method] Change

[Correction contents]

A software quality evaluation system according to a fourth aspect is the software quality evaluation system according to the second or third aspect, wherein the calculating means obtains a test coverage indicating a degree of execution of the software to be evaluated by the test. the, cunning multiply the test coverage to test the density of the evaluation target software
By correcting the test density, a hit rate is calculated from the error detection rate of the software to be evaluated and the new test density obtained by the correction , and the quality determination unit evaluates the hit rate. From the error detection rate of the target software, the new test density and the hit rate, determine the quality state of the evaluation target software according to the quality evaluation matrix, and determine the quality state of each quality evaluation generated by the quality evaluation generation unit. The quality evaluation corresponding to the quality state is determined.

[Procedure amendment 5]

[Document name to be amended] Statement

[Correction target item name] 0019

[Correction method] Change

[Correction contents]

According to a fifth aspect of the present invention, in the software quality evaluation system according to the second or third aspect, the calculating means calculates the number of errors with respect to a test progress up to an intermediate stage of a test performed on software to be evaluated. From the relationship,
The number of detection errors at the completion of the scheduled test
Biopsies predicts the intermediate stage to calculate the growth prediction curve, the predicted After testing on the basis of this growth prediction curve
The prediction error detection rate is calculated from the number of outgoing errors, and the hit rate is calculated from the prediction error detection rate and the test density after the end of the test. From the detection rate and the hit rate, the quality state of the software to be evaluated is determined according to the quality evaluation matrix, and among the quality evaluations generated by the quality evaluation generation unit, the quality evaluation corresponding to the quality state is determined. To decide.

[Procedure amendment 6]

[Document name to be amended] Statement

[Correction target item name] 0083

[Correction method] Change

[Correction contents]

In step S222, the control section 8 controls the quality evaluation output section 14 to output the quality evaluation determined by the quality determination section 13 (any one of the quality evaluations 1 to 12) and to output the countermeasure plan. The unit 15 is controlled to output a measure (one of the measure plans) corresponding to the quality evaluation (one of the quality evaluations 1 to 12). The control unit 8 causes the display unit 7 to display these output results. Then, the control unit 8, the display unit 7 that it will instruct the operator to enter the evaluation measures result data name that identifies the evaluation measures result data summarizing the quality evaluation and its countermeasure
To be displayed.

 ──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Shinya Kiso 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsui Electric Co., Ltd. (72) Inventor Toshikazu Yamazaki 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Rishi Electric Co., Ltd. (72) Inventor Shigeki Takano 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsui Electric Co., Ltd. (72) Inventor Ryo Iwai 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsui Electric (72) Inventor Masahiko Hiwatari 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsui Electric Co., Ltd. (72) Inventor Hitoshi Suzuki 2-3-2 Marunouchi, Chiyoda-ku, Tokyo Mitsubishi Electric Co., Ltd. In company

Claims (8)

[Claims] 1. A test density, which is a ratio of the number of test items to a scale of software used in a computer, and a ratio of an error number generated when the software executes the test of the number of test items to the scale of the software. A quality evaluation means for performing software quality evaluation based on a certain error detection rate; a measure proposal generation means for previously generating a plurality of measure plans corresponding to a plurality of expected quality evaluation results; and the quality evaluation means. Output means for outputting the quality evaluation result evaluated by the above and a countermeasure plan corresponding thereto. 2. The quality evaluation unit according to claim 1, wherein said test density, said error detection rate, and a hit rate, which is a ratio of said error detection rate to said test density, are set. Quality evaluation matrix generating means for generating a quality evaluation matrix in which a plurality of test densities, a plurality of error detection rates and a plurality of hit rates, and a plurality of software quality states are respectively associated with each other, Quality evaluation generating means for generating each quality evaluation corresponding to a quality state; calculating means for calculating a hit rate from a test density and an error detection rate of software to be evaluated; test density and error detection of the software to be evaluated From the rate and hit rate, according to the quality evaluation matrix,
Determine the quality state of the software to be evaluated, among the quality evaluations generated by the quality evaluation generation means,
The software quality evaluation system according to claim 1, further comprising: a quality determination unit configured to determine a quality evaluation corresponding to the quality state as the quality evaluation result. 3. The measure plan generating means generates, as the plurality of measure plans, respective measure plans respectively corresponding to the quality evaluations generated by the quality evaluation generating means, and the output means comprises: As a result, a quality evaluation output unit that outputs the quality evaluation determined by the quality determination unit, and, as a countermeasure plan corresponding to the quality evaluation result, among the countermeasure plans generated by the countermeasure plan generation unit, the quality 3. The software quality evaluation system according to claim 2, further comprising: a measure proposal output unit that outputs a measure proposal corresponding to the quality evaluation determined by the determination unit. 4. When the test coverage indicating the degree of execution of the software to be evaluated by the test is input, the calculation unit multiplies the test coverage by the test density of the software to be evaluated and generates a new test coverage. The test density is calculated, and the hit rate is calculated from the error detection rate of the software to be evaluated and the new test density. The quality determining means includes an error detection rate of the software to be evaluated, and the new test density. Determining the quality state of the software to be evaluated from the hit rate according to the quality evaluation matrix, and determining a quality evaluation corresponding to the quality state among the quality evaluations generated by the quality evaluation generation unit. 4. The method according to claim 2, wherein
Software quality evaluation system described in. 5. The calculation means calculates a growth prediction curve of the number of errors from a relation of the number of errors with respect to a test progress up to an intermediate stage of a test performed on software to be evaluated. The prediction error detection rate is calculated from the number of errors after the test, which is predicted based on the test error, and the hit rate is calculated from the prediction error detection rate and the test density after the test is completed. From the subsequent test density, the prediction error detection rate and the hit rate, the quality state of the software to be evaluated is determined according to the quality evaluation matrix, and among the quality evaluations generated by the quality evaluation generation means, 4. The software quality evaluation system according to claim 2, wherein a quality evaluation corresponding to the quality state is determined. 6. The software quality evaluation system according to claim 1, wherein said quality evaluation means includes quality re-evaluation means for re-evaluating the quality evaluation evaluated by said quality evaluation means. 7. The residual error number calculating method generating means for generating, for each quality evaluation result, a method for calculating a residual error number that can be predicted not to be detected due to insufficient execution of software by the test. An additional test method generating means for generating an additional test method for each quality evaluation result; a method for calculating the number of residual errors corresponding to the quality evaluation result evaluated by the quality evaluation means for software to be evaluated; and an additional test. Re-evaluation information determining means for determining a method; remaining error number calculating means for calculating the number of remaining errors according to the remaining error number calculating method determined by the re-evaluation information determining means; The test based on the number of remaining errors and the additional test method determined by the re-evaluation information determining means is executed on the software to be evaluated. Recalculation means for recalculating the test density and error detection rate of the software to be evaluated in consideration of the results of the evaluation, and redetermining the quality of the software to be evaluated based on the recalculated test density and error detection rate 7. The software quality evaluation system according to claim 6, comprising: 8. The software quality evaluation system according to claim 1, wherein the measure proposal generating means generates a plurality of quality enhancement measure plans as the plurality of measure plans.