Abstract
A method for computing the joint failure probability of a parallel system consisting of two linear limit state equations is given to indirectly obtain the value of the integral of the standard bivariate normal distribution by using the conclusion that the joint failure probability of the parallel system is equal to the value of the double integral. In the two-dimensional standard normal coordinate system, some circles whose centres are all at the coordinate origin are used to divide the two-dimensional standard sample space into a number of pairwise disjoint sub-sample spaces and obtain a number of pairwise disjoint sub-failure domains. According to the probability theory, the probability of any sub-failure domain can be expressed by using a sub-sample space probability and a conditional probability. Based on the total probability formula, the joint failure probability can be obtained by the sum of the probabilities of the sub-failure domains because the sub-sample spaces or the sub-failure domains are pairwise disjoint. By introducing a random variable obeying the Rayleigh distribution, it is possible to compute probabilities of the sub-sample spaces accurately. The formulae of computing the conditional probability are derived. The main parameters related to the computation of the joint failure probability, such as the minimum and maximum radii, and the number of the dividing circles, are discussed to make the computation process easy and the computed result meet a required precision. Examples show that it is possible and significant for the method in the paper to complete the computation of the standard bivariate normal distribution integral with high accuracy.
Similar content being viewed by others
References
Albers W, Kallenberg WCM (1994) A simple approximation to the bivariate normal distribution with large correlation coefficient. J Multivar Anal 49:87–96
Antia HM (2012) Numerical methods for scientists and engineers, 3rd edn. Hindustan Book Agency, New Delhi, pp 177–187
Cox DR, Wermuth N (1991) A simple approximation for bivariate and trivariate normal integrals. Int Stat Rev 59(2):263–269
Daley DJ (1974) Computation of bi-and tri-variate normal integrals. J R Stat Soc Ser C 23:435–438
Ditlevsen O (1979) Narrow reliability bounds for structural system. J Struct Mech 7(4):453–472
Divgi DR (1979) Calculation of univariate and bivariate normal probability functions. Ann Stat 7:903–919
Dong YG, Lu HT, Guo B (2013) A method for improving computational accuracy of reliability based on the theory of the minimum cut set. Chin J Mech Eng 49(20):184–191 ((in Chinese))
Dong YG, Zhang HM, He LG, Wang C, Wang MH (2019) The computation of standard normal distribution integral in any required precision based on reliability method. Commun Stat Theory Methods 48(6):1517–1528
Drezner Z (1978) Computation of the bivariate normal integral. Math Comput 32:277–279
Drezner Z, Wesolowsky GO (1990) On the computation of the bivariate normal integral. J Stat Comput Simul 35:101–107
Fayed HA, Atiya AF (2014) A novel series expansion for the multivariate normal probability integral based on Fourier series. Math Comput 83:2385–2402
Genz A (2004) Numerical computation of rectangular bivariate and trivariate normal and t probabilities. Stat Comput 14:251–260
Gong JX, Zhao GF (1996) An approximate algorithm for bivariate normal integral. Chin J Comput Mech 13(4):494–4999 ((in Chinese))
Harkness WL, Godambe AV (1976) Inequalities for tail probabilities for the multivariate normal distribution. Commun Stat Theory Methods A5(7):689–692
Hong HP (1999) An approximation to bivariate and trivariate normal integrals. Civ Eng Environ Syst 16:115–127
Hutchinson TP, Lai CD (1990) Continuous bivariate distributions: emphasizing applications. Rumsby, Adelaide
Kim J (2013) The computation of bivariate normal and t probabilities, with application to comparisons of three normal means. Comput Stat Data Anal 58:177–186
Lin JT (1995) A simple approximation for the bivariate normal integral. Probab Eng Inf Sci 9:317–321
Lozy MDME (1982) Simple computation of a bivariate normal integral arising from a problem of misclassification with applications to the diagnosis of hypertension. Commun Stat Theory Methods 11(19):2195–2205
Mee RW, Owen DB (1983) A simple approximation for bivariate normal probabilities. J Qual Technol 15(2):72–75
Milton RC (1972) Computer evaluation of the multivariate normal integral. Technometrics 14(4):881–889
Monhor D (2011) A new probabilistic approach to the path criticality in stochastic PERT. Cent Eur J Oper Res 19:615–633
Monhor D (2013) Inequalities for correlated bivariate normal distribution function. Probab Eng Inf Sci 27:115–123
Moskowitz H, Tsai HT (1989) An error-bounded polynomial approximation for bivariate normal probabilities. Commun Stat Simul Comput 18(4):1421–1437
Owen DB (1956) Tables for computing bivariate normal probabilities. Ann Math Stat 27:1075–1090
Pandey MD (1998) An effective approximation to evaluate multinormal integrals. Struct Saf 20:51–67
Rodgers JL, Nicewander WA (1988) Thirteen ways to look at the correlation coefficient. Am Stat 42(1):59–66
Sathe YS, Lingras SR (1980) A note on the inequalities for tail probability of the multivariate normal distribution. Commun Stat Theory Methods A9(7):711–715
Sándor Z, András P (2004) Alternative sampling methods for estimating multivariate normal probabilities. J Econ 120:207–234
Terza JV, Welland U (1991) A comparison of bivariate normal algorithms. J Stat Comput Simul 39:115–127
Verma AK, Ajit S, Karanki DR (2016) Reliability and safety engineering, 2nd edn. Springer, Heidelberg, pp 257–292
Vijverberg WPM (1997) Monte Carlo evaluation of multivariate normal probabilities. J Econ 76:281–307
Willink R (2004) Bounds on the bivariate normal distribution function. Commun Stat Theory Methods 33(10):2281–2297
Young JC, Minder CE (1974) Algorithm AS 76: an integral useful in calculating non-central t and bivariate normal probabilities. J R Stat Soc Ser C 23:455–457
Yuan XX, Pandey MD (2006) Analysis of approximations for multinormal integration in system reliability computation. Struct Saf 28:361–377
Author information
Authors and Affiliations
Corresponding authors
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Dong, Y., Xie, Q., Ding, S. et al. The evaluation of bivariate normal probabilities for failure of parallel systems. Stat Papers 63, 1585–1614 (2022). https://doi.org/10.1007/s00362-021-01282-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00362-021-01282-9