# Allocation Problems at Unit Level Testing

• P. K. Kapur
• H. Pham
• A. Gupta
• P. C. Jha
Chapter
Part of the Springer Series in Reliability Engineering book series (RELIABILITY)

## Abstract

As mentioned in previous chapter reliability, scheduled delivery and cost are the three main quality attributes for almost all software. By determining the release time of the software optimally taking into consideration the various constraints and aspects of the software enables to best achieve these objectives. Many a time in the software release time problems we have seen keeping the cost minimization objective alone may leave us with a solution that the reliability achieved is low. On the other hand reliability maximization objective alone may require large budget. The two objectives simultaneously are conflicting and demand bounds on budget and achievable reliability. The release time problem by no means controls the consumption of the testing resources. The total software-testing cost as well as reliability depends largely on the consumption of the testing resources during the testing process.

## Keywords

Allocation Problem Module Testing Software Module Testing Effort Resource Allocation Problem
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

## Notation

pr {.}

Probability

{N(t), t ≥ 0}

Counting process representing the cumulative number of software faults detected in the time interval [0, t]

m(t)

Mean value function in the NHPP model, m(0) = 0

mf(t)

Mean value function of the failure process in the NHPP model, m f (0) = 0

mr(t)

Mean value function of the removal process in the NHPP model, m r (0) = 0

w(t), W(t)

Current testing-resource expenditures at testing time t for (w i (t), W i (t)) software (module i) and its integral form, i.e., $$W(t) = \int_{0}^{t} {w(x)dx} ;\quad W_{i} (t) = \int_{0}^{t} {w_{i} (x)dx} ,$$

a(ai)

Expected initial fault content in software (module i), a > 0

b (bi)

Constant fault detection/removal rate per remaining faults in software (module i) 0 < b<1

i

Subscript for each software-module number i = 1, 2, …, N

vi

Weighted importance for each software module, v i  > 0

zi, Z

The number of software faults remaining in each software module and the whole system

W

The specified total testing-resource expenditures before module testing, W > 0

R(s)

Software reliability which means that no software failures occur during the time interval (0, s](s ≥ 0) after the testing process

γ

Constant parameter related to the failure rate, γ > 0

R0

Objective value of the software reliability, 0 < R 0 < 1

*

Superscript that denotes the optimal solution of the test resource allocation problem

Ai

v i a i b i (detectability of module i)

L, λ

Lagrangian, Lagrange multiplier λ

## References

1. 1.
Ohtera H, Yamada S (1990) Optimal allocation and control problems for software testing resources. IEEE Trans Reliab 39(2):171–176
2. 2.
Kubat P, Koch HS (1983) Managing test-procedure to achieve reliable software. IEEE Trans Reliab 32(3):299–303
3. 3.
Yamada S, Hishitani J, Osaki S (1991) Test-effort dependent software reliability measurement. Int J Syst Sci 22(1):73–83
4. 4.
Yamada S, Ichimori T, Nishiwaki M (1995) Optimal allocation policies for testing-resource based on a software reliability growth model. Math Comput Model 22:295–301
5. 5.
Leung YW (1997) Dynamic resource allocation for software-module testing. J Syst Softw 37(2):129–139
6. 6.
Huang CY, Lo JH, Kuo SY, Lyu MR (2002) Optimal allocation of testing resources for modular software systems. In: Proceedings of 13th IEEE International Symposium on Software Reliability Engineering (ISSRE 2002), Nov 2002, Annapolis, Maryland, pp 129–138Google Scholar
7. 7.
Huang CY, Lo JH, Kuo SY, Lyu MR (2004) Optimal allocation of testing-resource considering cost, reliability, and testing-effort. In: Proceedings of 10th IEEE/IFIP Pacific Rim International Symposium on Dependable Computing, Papeete, Tahiti, French Polynesia, pp 103–112Google Scholar
8. 8.
Goel AL, Okumoto K (1979) Time dependent error detection rate model for software reliability and other performance measures. IEEE Trans Reliab R 28(3):206–211
9. 9.
Kubat P (1989) Assessing reliability of modular software. Oper Res Lett 8(1):35–41
10. 10.
Kapur PK, Jha PC, Bardhan AK (2004) Optimal allocation of testing resource for a modular software. Asia Pacific J Oper Res 21(3):333–354
11. 11.
12. 12.
Dur M, Horst R, Thoai NV (2001) Solving sum of ratios fractional programs using efficient points. Optimization 1:447–466
13. 13.
Bhatia D, Kumar N, Bhudhiraja RK (1997) Duality theorem for non differential multi-objective programs. Indian J Pure Appl Math 28(8):1030–1042Google Scholar
14. 14.
Geoffrion AM (1968) Proper efficiency and theory of vector maximization. J Math Anal Appl 22:613–630
15. 15.
Jha PC, Gupta D, Yang B, Kapur PK (2009) Optimal testing-resource allocation during module testing considering cost, testing effort and reliability. Comput Ind Eng 57(3):1122–1130
16. 16.
Jha PC, Gupta D, Anand S, Kapur PK (2006) An imperfect debugging software reliability growth model using lag function with testing coverage and related allocation of testing effort problem. Commun Dependability Qual Manag: Int J 9(4):148–165Google Scholar
17. 17.
Xie M, Zhao M (1992) The Schneidewind software reliability model revisited. In: Proceedings of 3rd International Symposium on Software Reliability Engineering, pp 184–192Google Scholar
18. 18.
Ignizio JP (1994) Linear programming in single and multiple objective functions. Prentice Hall, Englewood Cliffs, LondonGoogle Scholar
19. 19.
Xie M , Yang B ( 2000) Optimal testing time allocation for modular system. Int J Qual Reliab Manag 11(8):854–863Google Scholar
20. 20.
Bazaraa SM, Sherali HD, Setty CM (1993) Non linear programming: Theory and algorithm, John Wiley and Sons, New York.Google Scholar
21. 21.
Steuer RE (1989) Multiple criteria optimization: Theory, computation and application, Wiley, New York