• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.377-377
• /
• 2000

Current software cost estimation models, such as the 1951 COCOMO, its 1987 Ada COCOMO update, is composed of nonlinear models, such as product attributes, computer attributes, personnel attributes, project attributes, effort-multiplier cost drivers, and have been experiencing increasing difficulties in estimating the costs of software developed to new lift cycle processes and capabilities. The COCOMO II is developed fur new forms against the current software cost estimation models. This paper provides a case-based analysis result of the cost driver in the software cost models, such as COCOMO and COCOMO 2.0 by fuzzy and neural network.

• Journal of Information Technology Applications and Management
• /
• v.13 no.4
• /
• pp.1-12
• /
• 2006

Many software reliability growth models(SRGM) have been proposed since the software reliability issue was raised in 1972. The technology to estimate and grow the reliability of developing S/W to target value during testing phase were developed using them. Most of these propositions assumed the S/W debugging testing efforts be constant or even did not consider them. A few papers were presented as the software reliability evaluation considering the testing effort was important afterwards. The testing effort forms which have been presented by this kind of papers were exponential, Rayleigh, Weibull, or logistic functions, and one of these 4 types was used as a testing effort function depending on the S/W developing circumstances. I propose the methology to evaluate the SRGM using least square estimator and maximum likelihood estimator for those 4 functions, and then examine parameters applying actual data adopted from real field test of developing S/W.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.5
• /
• pp.566-570
• /
• 2010

Use Case Point(UCP) is a measure of a software project size for software effort estimation based on use case. UCP measures the size of the software project based on the use case model. Because UCP is based on the use case model, it is intuitive and easy to obtain. Also, it does not require extra artifacts. On the other hand, UCP has some problems. UCP assumes every transaction has the same complexity. But, the number of operations and complexity of operations may affect complexity of transaction. In addition, UCP uses simple rating scale of complexity, but it may be inadequate for detailed estimates. To solve these problems, we suggest "Transaction Point(TP)", a size measure based on use case transaction. TP considers actors and operations in transaction. Complexity of transaction is based on the number of operations and complexity of operation, so it can support detailed estimation.

• Annual Conference of KIPS
• /
• 2019.10a
• /
• pp.701-703
• /
• 2019

현재의 소프트웨어는 매우 크고 복잡하다. 고품질 소프트웨어는 요구사항 단계부터 정확하게 분석해야 한다. 그러나 자연어 요구사항은 부정확하기 때문에 유스케이스 식별이 어렵다. 게다가, 기존 요구 사항 분석 방법은 객체와 용어 식별에 대한 정확한 기준이 없다. 따라서 분석 결과는 분석가마다 다르다. 본 논문에서는 자연어 요구사항으로부터 유스케이스 추출 및 소프트웨어 공수 산정 방법을 제안한다. 제안한 방법에서는 Goal Modeling과 Fillmore의 Case 메커니즘을 개선했다. 이 방법은 자연어 요구 사항을 단계별로 분석하여 유스케이스를 모델링한다. 또한 유스케이스 점수(Use Case Point)를 이용하여 소프트웨어의 공수를 산정(Effort Estimation)한다. 제안한 방법은 고품질 소프트웨어 개발을 위해 자연어 요구사항의 변경 없이 유스케이스 추출이 가능하다. 또한 추출된 유스케이스를 통해 UCP 기반의 공수 산정을 평가할 수 있다. 본 논문에서는 우체국 시스템의 사례에 적용하였다.

• Journal of Internet Computing and Services
• /
• v.8 no.4
• /
• pp.105-117
• /
• 2007

In order to succeed in a bid or development, the project manager should estimate its cost and schedule more accurately in the early stage of the project. Usually, the nominal schedule of most projects can be derived from rule of thumb, first-order estimation practice, or ball-park schedule estimation table. But the rule-of-thumb models for the nominal schedule estimation are so various, and the first-order estimation practice does not provide sufficient information. So they do not help much to decide on the proper development effort and schedule for a particular size of project. This paper presents a statistical regression model for deciding the development effort and schedule of a project using the ball-park schedule estimation table. First, we have redefined such words suggested in the ballpark schedule estimation table as shortest possible schedule, efficient schedule and nominal schedule, Next, we have investigated the relationship between the development effort and the schedule. Finally, we have suggested a model for estimating the development effort and the more accurate schedule of such particular sizes of software as are not presented in the ball-park schedule estimation table. The experimental results show that our proposed regression analysis model decreases the mean magnitude of relative error by 2% at maximum. Also this model can estimated the development effort and schedule for a particular size of software.

• Journal of Software Engineering Society
• /
• v.26 no.4
• /
• pp.77-92
• /
• 2013

To plan a software project, the manager (1)make a work breakdown structure (WBS), (2) estimate efforts for each task, (3) assign employee to each task, and (4) estimate overall schedule. When software project becomes complicated, the possible combination of WBS, effort, and employee assignments dramatically becomes larger. Software planning tool can help software project managers to deal with this complexity. In this research, we discuss with a group of experts who work in software industry, to elicit practical requirements that should be considered in the software planning technique. Considering these requirements, we develop a software project planning tool APP (Automatic Project Planner) which provide effort estimation based on historical knowledge data and automatic human resource allocation. Our technique can be the basis of reasonable and practical software project planing.

• Journal of Information Technology Applications and Management
• /
• v.11 no.2
• /
• pp.191-204
• /
• 2004

Many software reliability growth model(SRGM) have been proposed for past several decades. Most of these propositions assumed the S/W debugging testing efforts be constant or even did not consider them. A few papers were presented as the software reliability evaluation considering the testing effort was important afterwards. The testing effort forms which have been presented by this kind of papers were exponential, Rayleigh, Weibull, or Logistic functions, and one of these 4 types was used as a testing effort function depending on the S/W developing circumstances. We consider the methology to evaluate the SRGN using least square estimator(LSE) and maximum likelihood estimator(MLE) for those 4 functions, and then examine parameters applying actual data adopted from real field test of developing S/W.

• The KIPS Transactions:PartD
• /
• v.11D no.3
• /
• pp.597-604
• /
• 2004

Accurate software cost estimation is essential to both developers and customers. Most of the cost estimating models based on the size measure methods, such as LOC and FP, are obtained through size estimation. The accuracy of size estimation directly influences the accuracy of cost estimation. As a result, the overall structure of regression-based cost models applies the power function based on software size. Many growth phenomenon in nature such as the growth in living organism, performance of technology, and learning capability of human show an S-shaped curve. This paper proposes a model which estimates the developing effort by using the growth curve. The presented model assumes that the relation cost and size follows the growth curve. The appropriateness of the growth curve model based on Function Point, Full-Function Point and Use-Case Point, which are the general methods in estimating the software size have been confirmed. The proposed growth curve model shows similar performance with power function model. In conclusion, the growth curve model can be applied in the estimation of the software cost.

• The KIPS Transactions:PartD
• /
• v.11D no.1
• /
• pp.173-182
• /
• 2004

The research to estimate development effort of software has been progress. But, it is not easy gain that testing data for estimating of development effort. Also, if we get the testing data, it is important that analysis testing data. In this paper, we study the data analysis of software development effort using the 789 software development projects which developed in the 1990's. Software development scale and software development team site are various. Using the characteristic of factor, we have to study characteristic of data and we estimate the development effort step by step. First, we prove the difference of development effort with the 789 project data according to development type, development environment, the development language etc. Also, we execute the crosstabs analysis that team site and function point.

• The KIPS Transactions:PartD
• /
• v.11D no.4
• /
• pp.885-892
• /
• 2004

Weibull distribution Iincluding Rayleigh and Exponential distribution is a typical model to estimate the effort distribution which is committed to the software testing phase. This model does not represent standpoint that many efforts are committed actually at the test beginning point. Moreover, it does not properly represent the various distribution form of actual test effort. To solve these problems, this paper proposes the Sigmoid model. The sigmoid function to be applicable in neural network transformed into the function which properly represents the test effort of software in the model. The model was verified to the six test effort data which were got from actual software projects which have various distribution form and verified the suitability. The Sigmoid model nay be selected by the alternative of Weibull model to estimate software test effort because it is superior than the Weibull model.