Homework 2 - ME 5895 - Spring 2024
Homework 2 - ME 5895 - Spring 2024
Homework 2
1 Suppose that continuous random variables X and Y follow a standard bivariate normal
distribution with the correlation coefficient being ρ. Answer the following questions:
[20 pts total]
(2) Find the conditional probability density functions (PDFs) fX|Y(x|y) and f Y|X(y|x).
[10 pts total]
2 Two electronic components are arranged to form a system in two configurations shown in Fig.
1. Suppose the time to failures (TTFs) of the two components, T1 and T2, follow the below
exponential distributions:
• TTF (T1) of component A: = , if t ≥ 0, and 0, otherwise.
• TTF (T2) of component B: = , if t ≥ 0, and 0, otherwise.
–1 –1
• λ1 = 0.0002 hr and λ2 = 0.0004 hr .
• Random variables T1 and T2 are mutually independent: the state of one component does
not affect that of the other.
Component A
Component A Component B
Component B
(a) (b)
Fig. 1 Reliability block diagrams of system configurations (a) and (b).
(2) Determine the system reliability at 100 hr (i.e., the probability of reliable system operation
for 100 hr) for either system configuration. [10 pts]
(3) Given that the system has operated reliably for 100 hr, determine the system reliability at
200 hr (i.e., the probability of reliable system operation for 200 hr) for either system
configuration. [10 pts]
1
ME 5895 Spring 2024
3 A number of finite element analysis (FEA) simulations are performed on a vehicle mechanical
component for the reliability-based design of the component. Through the design of
experiments (DOE) and FEA simulations, a surrogate model of the maximum stress σ (MPa)
is created with respect to three design variables X1, X2, and X3 as
σ ( X ) = β 0 + β1 X 1 + β 2 X 2 + β 3 X 3
where β0 = 540, β1 = 0.8, β2 = –0.3, β3 = –32, and Xi (i = 1, 2, 3) are random design variables,
whose statistical information is summarized in Table 1.
Table 1 Statistical information of the random variables Problem 3
Random variable X1 (length) X2 (width) X3 (thickness)
Distribution Normal Normal Normal
Mean 60 20 1
Standard deviation 12 4 0.2
(1) If the design requirement for the stress is σ ≤ 575 MPa. What is the probability that the
design satisfies the requirement? Use a look-up table for the standard normal cumulative
distribution function (CDF) to solve this problem (the use of MATLAB is not
recommended for this question). [10 pts]
(2) If the reliability (i.e., the probability of requirement satisfaction) has a target of 99.87%,
does the design exceed the target reliability level? If not, how do you suggest improving
the reliability of the design? [5 pts]