한국전자통신학회논문지 (The Journal of the Korea institute of electronic communication sciences)

한국전자통신학회 (Korea Institute of Electronic Communication Science)
권호 표지
DOI QR코드

DOI QR Code

지수족 수명 분포를 이용한 소프트웨어 개발 모형의 성능 평가

Performance Evaluation of Software Development Model using Exponential Family Life Distribution

  • 배효정 (남서울대학교 드론공간정보공학과)
  • Hyo-Jeong Bae (Dept. Drone and GIS Engineering, Namseoul University)
  • 투고 : 2024.07.19
  • 심사 : 2024.10.12
  • 발행 : 2024.10.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 연구에서는 NHPP 기반 소프트웨어 개발 모형에 시스템 고장의 신뢰성 분석에 적합한 지수족 수명 분포를 적용한 후, 개발자가 요구한 고장시간 데이터를 대입하여 모형의 성능을 새롭게 평가하였다. 이에 따라, 첫째, 모형의 성능을 결정하는 m(t) 함수를 분석한 결과, 모든 모형이 참값을 예측하는 능력에서 과대하게 추정하는 속성을 보였다. 하지만, 가장 작은 오차를 보인 Burr-Hatke-Exponential 모형은 효율적이었다. 둘째, 모형의 성능에 관한 속성(m(t), 개발비용, 방출 시간)을 종합 평가한 결과, Lindley 모형이 가장 우수함을 확인하였다. 따라서, 본 연구를 통해 지수족 수명 분포의 성능을 새롭게 탐색하였고, 개발자는 관련 데이터를 활용함으로써 초기 단계에서 필요한 속성을 파악하고, 이를 개발 테스팅 단계에서 참조 데이터로 활용할 수 있다.

In this study, an exponential family life distribution suitable for reliability analysis of system failures was applied to the NHPP-based software development model, and then the performance of models were newly evaluated by substituting the failure time data requested by the developer. First, as a result of analyzing the m(t) function that determines the model performance, all models showed the property of overestimating the ability to predict the true value. However, the Burr-Hatke-Exponential model, which showed the smallest error, was efficient. Second, the Lindley model was found to be the best when total evaluating the properties of the model performance (m(t), development cost, release time). Therefore, through this study, the performance of the exponential family life distribution was newly explored, and by utilizing related data, developers can identify required properties at an early stage and use them as basic reference data during the development testing phase.

키워드

과제정보

이 논문은 2024년도 남서울대학교 학술연구비 지원에 의해 연구되었음.

참고문헌

  1. T. Yang, "Comparative study on the Attributes Analysis of Software Development Cost Model Based on Exponential-type Lifetime Distribution," Int. J. of Emerging Technology and Advanced Engineering, vol. 11, no. 10, 2021, pp. 166-176. https://doi.org/10.46338/ijetae1021_20 
  2. Y. Ra, "A Comparative Analysis on the Performance of Finite Failure NHPP Software Reliability Model Based on Rayleigh-type Lifetime Distribution," J. of Theoretical and Applied Information Technology, vol. 99, no. 24, 2021, pp. 6162-6172. 
  3. S. Chatterjee, J. B. Singh, A. Roy, and A. Shukla, "NHPP-based Software Reliability Growth Modeling and Optimal Release Policy for N-Version Programming System with Increasing Fault Detection Rate under Imperfect Debugging," Proc. of the National Academy of Sciences, vol. 90, 2020, pp. 11-26. https://doi.org/10.1007/s40010-018-0551-1 
  4. H. Pham and X. Zhang, "NHPP Software Reliability and Cost Models with Testing Coverage," European J. of Operational Research, vol. 145, issue. 2, 2003, pp. 443-454. https://10.1016/S0377-2217(02)00181-9 
  5. H. Okamura and T. Dohi, "Phase-type software reliability model: parameter estimation algorithms with grouped data," Annals of Operations Research, vol. 244, no. 1, 2016. pp. 77. https://10.1007/s10479-015-1870-0 
  6. T. Yang, "A Comparative Study on Reliability Attributes for Software Reliability Model Dependent on Lindley and Erlang Life Distribution," J. of Korea Institute of Information, Electronics, and Communication Technology, vol. 10, no. 5, 2017, pp. 469-475. http://dx.doi.org/10.17661/jkiiect.2017.10.5.469 
  7. S. Park, "Performance Attributes Analysis of NHP-Based Software Development Cost Model With Inverse-Type Distribution Properties," J. of Theoretical and Applied Information Technology, vol. 101, no. 19, 2023, pp.5929-5939. 
  8. T. Yang, "Performance Analysis on the Reliability Attributes of NHPP Software Reliability Model Applying Exponential and Inverse-Exponential Lifetime Distribution," J. of Theoretical and Applied Information Technology, vol. 100, no. 22, 2022, pp. 6645-6656. 
  9. H. Kim, "A Comparative Study on the Cost of Software Development Model Based on Burr-Hatke-Exponential Distribution," Int. J. of Engineering Research and Technology, vol. 12, no. 11, 2019, pp. 2036-2040. 
  10. H. Kim, "A Comparative Study on the Finite Failure Software Reliability Model with Modified Lindley Type Lifetime Distribution," Int. J. of Engineering Research and Technology, vol. 12, no. 6, 2019, pp. 760-764. 
  11. Y. Zhang and K. Wu, "Software Cost Model Considering Reliability and Time of Software in Use," J. of Convergence Information Technology, vol. 7, no. 13, 2012, pp. 135-142. https://10.4156/jcit.vol7.issue13.16 
  12. R. S. Prasad, K. R. H. Rao, and R. R. L. Kantha, "Software Reliability Measuring using Modified Maximum Likelihood Estimation and SPC," Int. J. of Computer Applications, vol. 21, no. 7, 2011, pp. 1-5. https://10.5120/2527-3440 
  13. S. Park, "Comparative Analysis on the Performance of NHPP Software Reliability Model with Exponential Distribution Characteristics," J. of The Korea Institute of Electronic Communication Sciences, vol. 17, no. 4, 2022. pp. 641-648. http://dx.doi.org/10.13067/JKIECS.2022.17.4.6.41 
  14. S. Lee, "A Routing Algorithm based on Deep Reinforcement Learning in SDN," J. of The Korea Institute of Electronic Communication Sciences, vol. 16, no. 6, 2021. pp. 1153-1160. http://dx.doi.org/10.13067/JKIECS.2021.16.6.1153 
  15. S. Park, "Comparative Study on the Performance of Finite Failure NHPP Software Development Cost Model Based on Inverse-Type Life Distribution," J. of The Korea Institute of Electronic Communication Sciences, vol. 18, no. 5, 2023. pp. 935-944. http://dx.doi.org/10.13067/JKIECS.2023.18.5.935