Weibull Lindley Distribution

D.A. Magfira, D. Lestaria) and S. Nurrohmah

Department of Mathematics and Natural Sciences (FMIPA), Universitas Indonesia,

Depok 16424, Indonesia

a)
Corresponding author: dian.lestari@sci.ui.ac.id

Abstract. In reliability systems there are known two types of systems namely series systems and parallel systems. In the
series system, failure will occur if any of the possible event happens. Applications of the series system analysis also varies
from inspecting the durability of manufactured products to examining diseases in human. Therefore, several distributions
have been introduced to model failure data in series system. However, these distributions cannot model data with bathtub
shaped hazard function even though it is the one mostly found in real life situation. As a result, distribution which can
model lifetime data in series system with bathtub-shaped hazard function has to be developed. Weibull Lindley distribution,
which was introduced by Asgharzadeh et al. (2016), is developed to solve the problem. Weibull Lindley distribution
describes lifetime data of an object that can experience failure caused by 2 possible events. It can model data with
increasing, decreasing and bathtub shaped hazard function. This paper discusses the process of forming the Weibull Lindley
distribution, its properties and parameter estimation using the maximum likelihood method. In addition, the application of
Weibull Lindley distribution in lifetime data of machine consists of two independent component paired in series also be
discussed.

Keywords: Bathtub Hazard, Compounding Distribution, Hazard Function, Maximum Likelihood Method, Series System

INTRODUCTION

Lifetime data tells about time needed for an event or failure to happen. For example, lifetime data describes
time until death for human or time until failure for machine. The analysis of lifetime data is very important in various
fields, such as biomedical science and engineering. In the lifetime data analysis, the object will experience only one
event that can lead to failure. But in real life the population is heterogeneous so it can cause more complicated
problems e.g. failure can occur due to several possible unknown events [1].

In reliability systems there are known two types of systems namely series systems and parallel systems. In
the series system, failure will occur if any of the possible event happens. In real life application of series system, inter-
event seems to compete each other to cause failure because if one event has occurred, the observation time is
considered finish so that another event cannot occur.

Applications of the series system analysis also varies from inspecting the durability of manufactured products
to examining diseases in human. Thus, several distributions have been introduced to model lifetime data in series
system. Some distributions that can be used to model lifetime data in series system are Exponential Geometric
distribution, Exponential Poisson distribution, Weibull Geometric distribution, Pareto Poisson-Lindley distribution
and so on [2]. However, these distributions cannot model data with bathtub-shaped hazard function even though it is
the one mostly found in real life situation. Bathtub shaped hazard function can be found in the data of machines
population observed since the beginning of usage until failure or human population observed since birth until death.
Therefore, distribution which can model lifetime data in series system with bathtub-shaped hazard function has to be
developed. Weibull Lindley distribution, which was introduced by Asgharzadeh et al. (2016), is developed to solve
the problem.
The Weibull Lindley distribution is formed by compounding Weibull distribution and Lindley distribution. It can
model increasing, decreasing, and bathtub shaped hazard function from lifetime data in series system. Weibull Lindley
distribution describes lifetime data of an object that can experience failure caused by 2 possible events. In medical
field, Weibull Lindley is suitable for modelling competing risk data with two risk factors. Whereas, in engineering
field, Weibull Lindley distribution can model a lifetime data of a machine which has two independent components
paired in series. This paper discusses the formation of Weibull Lindley distribution in a detailed manner, its properties,
and the estimation of its parameters.

METHODOLOGY

Formation of Weibull Lindley Distribution

The Weibull Lindley distribution is formed by compounding Weibull distribution and Lindley distribution. Let 𝑋1
denote a Lindley random variable with parameter 𝜆 > 0 and 𝑋2 denote a Weibull random variable with parameter
𝛼, 𝛽 > 0. Assume 𝑋1 and 𝑋2 are independent random variables. Then, Weibull Lindley random variables is defined
as 𝑋 = min⁡(𝑋1 , 𝑋2 ).

The survival function of Weibull distribution with parameter 𝛼, 𝛽 [3] known as follows:
𝛼
𝑆𝑋1 (𝑥) = 𝑒 −(𝛽𝑥) ⁡𝑓𝑜𝑟⁡𝑥 > ⁡0⁡⁡⁡⁡⁡⁡𝛼, 𝛽 > ⁡0⁡ (1)

The survival function of Lindley distribution with parameter 𝜆 [4] known as follows:

1 + 𝜆 + 𝜆𝑥 −𝜃𝑥
𝑆𝑋2 (𝑥) = ( 𝑒 ) 𝑓𝑜𝑟⁡𝑥 > ⁡0, 𝜆 > ⁡0⁡ (2)⁡
1+𝜆

Therefore, the distribution function of Weibull Lindley distribution is obtained by compounding two independent,
i.e. Weibull dan Lindley random variables based on compounding methods by Adamidis and Loukas [1], that is
defined by,
𝐹𝑋 (𝑥) = 1 − [𝑆𝑋1 (x)𝑆𝑋2 (x)⁡]

So, the distribution function of Weibull Lindley distribution based on Equation (1) and Equation (2) can be written
as follows:
1 + 𝜆 + 𝜆𝑥 −𝜆𝑥−(𝛽𝑥)𝛼
𝐹𝑋 (𝑥) = 1 − ( 𝑒 ) ,⁡⁡⁡⁡⁡⁡⁡⁡𝑥 > 0, 𝛼 > 0, 𝛽 > 0, 𝜆 > 0⁡ (3)
1+𝜆

Survival function of Weibull Lindley distribution based on Equation (3) can be written as follows:

1 + 𝜆 + 𝜆𝑥 −𝜆𝑥−(𝛽𝑥)𝛼
𝑆𝑋 (𝑥) = ( 𝑒 ) ; ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡𝑥 > 0, 𝛼 > 0, 𝛽 > 0, 𝜆 > 0⁡ (4)
1+𝜆

In general, the probability density function of Weibull Lindley distribution is:

1 𝛼
[𝛼𝜆(𝛽𝑥)𝛼 + 𝛼𝛽(1 + 𝜆)(𝛽𝑥)𝛼−1 + 𝜆2 (1 + 𝑥)]𝑒 −𝜆𝑥−(𝛽𝑥) ⁡𝑥 > 0, 𝛼 > 0, 𝛽 > 0, 𝜆 > 0⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡(5)
𝑓𝑋 (𝑥) = {1 + 𝜆
0⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡𝑜𝑡ℎ𝑒𝑟𝑠
⁡

Known from Figure 1 that the pdf of Weibull Lindley distribution is unimodal when 𝛼 > 𝛽 and 𝛼, 𝛽, 𝜆 > 1. When
𝛼 > 1, 𝜆 < 1 the pdf of Weibull Lindley distribution will be unimodal. Then, the pdf of Weibull Lindley distribution
will be decreasing when 𝛼 < 1.
 FIGURE 1. Pdf shapes of Weibull Lindley distribution

Therefore, the hazard function of Weibull Lindley distribution based on Equation (4) and Equation (5) can be
written as follows:
𝜆2 (1 + 𝑥)
ℎ𝑋 (𝑥) = 𝛼𝛽(𝛽𝑥)𝛼−1 + ; ⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡𝑥 > 0, 𝛼 > 0, 𝛽 > 0, 𝜆 > 0⁡ (6)
1 + 𝜆 + 𝜆𝑥

Shown in Figure 2 that the hazard function shape of Weibull Lindley distribution is increasing when 𝛼 > 1. When
0 < 𝛼 < 1 the hazard function shape of Weibull Lindley distribution will be decreasing. Then, the hazard function of
Weibull Lindley distribution will bathtub shaped when 𝛼 close to 0 or 𝛼 close to 1. In conclusion, the shape of Weibull
Lindley hazard function only depends on the value of 𝛼.

FIGURE 2. Hazard function shapes of Weibull Lindley distribution

The r-th moment function of Weibull Lindley can be written as follows:

∞ ∞
𝑟]
(−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 𝛼 + 𝑟 + 1)
𝛼
𝜆2 (−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 𝑟 + 2)
𝐸[𝑋 = 𝛼𝛽 ∑ 𝛼𝑖+𝛼+𝑟 [ + Γ(𝛼𝑖 + 𝛼 + 𝑟)] + ∑ 𝛼𝑖+𝑟+1 [Γ(𝛼𝑖 + 𝑟 + 1) + ]⁡ (7)
𝑖! 𝜆 (1 + 𝜆) 1+𝜆 𝑖! 𝜆 𝜆
𝑖=0 𝑖=0

Then, the first moment or mean and the second moment of Weibull Lindley distribution can be written as follows:
∞ ∞
(−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 𝛼 + 2)
𝛼
𝜆2 (−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 3)
𝐸[𝑋] = 𝛼𝛽 ∑ 𝛼𝑖+𝛼+1 [ + Γ(𝛼𝑖 + 𝛼 + 1)] + ∑ [Γ(𝛼𝑖 + 2) + ] (8)
𝑖! 𝜆 (1 + 𝜆) 1+𝜆 𝑖! 𝜆𝛼𝑖+2 𝜆
𝑖=0 𝑖=0
⁡⁡⁡
∞ ∞
2]
(−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 𝛼 + 3)
𝛼
𝜆2 (−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 4)
𝐸[𝑋 = 𝛼𝛽 ∑ 𝛼𝑖+𝛼+2 [ + Γ(𝛼𝑖 + 𝛼 + 2)] + ∑ 𝛼𝑖+3
[Γ(𝛼𝑖 + 3) + ]⁡ (9)
𝑖! 𝜆 (1 + 𝜆) 1+𝜆 𝑖! 𝜆 𝜆
𝑖=0 𝑖=0
 Variance of Weibull Lindley distribution based on Equation (8) and Equation (9) can be written as follows:
∞ ∞
(−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 𝛼 + 3)
𝛼
𝜆2 (−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 4)
𝑉𝑎𝑟(𝑋) = ⁡𝛼𝛽 ∑ 𝛼𝑖+𝛼+2 [ + Γ(𝛼𝑖 + 𝛼 + 2)] + ∑ 𝛼𝑖+3
[Γ(𝛼𝑖 + 3) + ]
𝑖! 𝜆 (1 + 𝜆) 1+𝜆 𝑖! 𝜆 𝜆
𝑖=0 𝑖=0
∞ ∞ 2
𝛼
(−1)𝑖 𝛽 𝛼𝑖
Γ(𝛼𝑖 + 𝛼 + 2) 𝜆 2
(−1)𝑖 𝛽 𝛼𝑖 Γ(𝛼𝑖 + 3)
⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡⁡− [𝛼𝛽 ∑ 𝛼𝑖+𝛼+1 [ + Γ(𝛼𝑖 + 𝛼 + 1)] + ∑ 𝛼𝑖+2
[Γ(𝛼𝑖 + 2) + ]] (10)
𝑖! 𝜆 (1 + 𝜆) 1+𝜆 𝑖! 𝜆 𝜆
𝑖=0 𝑖=0

Parameter Estimation of Weibull Lindley Distribution

Likelihood function for Weibull Lindley distribution is needed for maximum likelihood estimation. To obtain
likelihood function, first, let 𝑋1 , 𝑋2 , … , 𝑋𝑛 a random samples of size n from distribution with pdf (5), then the likelihood
function with observed sample 𝑥 is a multiplication of its pdf :
∞
1 𝑛 ∞ ∞ 𝛼
𝐿(𝛼, 𝛽, 𝜆; 𝑥) == ( ) ∏[𝛼𝜆(𝛽𝑥𝑖 )𝛼 + 𝛼𝛽(1 + 𝜆)(𝛽𝑥𝑖 )𝛼−1 + 𝜆2 (1 + 𝑥𝑖 )] 𝑒 −𝜆 ∑𝑖=1 𝑥𝑖 −∑𝑖=1(𝛽𝑥𝑖) ⁡ (11)
1+𝜆
𝑖=1

Log-likelihood function is needed to help in obtaining solution. The loglikelihood function can be obtained as:
∞ ∞ ∞

𝑙(𝛼, 𝛽, 𝜆; 𝑥) = ⁡ −𝑛 ln(1 + 𝜆) + ∑ 𝑙𝑛[𝛼𝜆(𝛽𝑥𝑖 )𝛼 + 𝛼𝛽(1 + 𝜆)(𝛽𝑥𝑖 )𝛼−1 +𝜆2 (1

+ 𝑥𝑖 )] − 𝜆 ∑ 𝑥𝑖 − ∑(𝛽𝑥𝑖 )𝛼 ⁡ (12)
𝑖=1 𝑖=1 𝑖=1

Therefore, the first derivative of (12) with respect to each parameter is needed for maximum likelihood estimation.
Estimated value of 𝛼, 𝛽 and 𝜆 are obtained by solving the equation as follows:

Where 𝑊(𝑥𝑖 ) = [𝛼𝜆(𝛽𝑥𝑖 )𝛼 + 𝛼𝛽(1 + 𝜆)(𝛽𝑥𝑖 )𝛼−1 + 𝜆2 (1 + 𝑥𝑖 )], Then

∞ ∞
𝜕𝑙(𝛼, 𝛽, 𝜆; 𝑥) 1
=∑ {[(𝜆 + 𝛼𝜆𝑙𝑛(𝛽𝑥𝑖 ))(𝛽𝑥𝑖 )𝛼 ] + [(𝛽(1 + 𝜆) + 𝛼𝛽(1 + 𝜆)𝑙𝑛(𝛽𝑥𝑖 ))(𝛽𝑥𝑖 )𝛼−1 ]⁡} − ∑(𝛽𝑥𝑖 )𝛼 𝑙𝑛(𝛽𝑥𝑖 ) = 0⁡(13)
𝜕𝛼 𝑊(𝑥𝑖 )
𝑖=1 𝑖=1

Equation (13) can be solved by numerical method that will produce an estimated value of 𝛼, which is 𝛼̂.⁡
∞ ∞
𝜕𝑙(𝛼, 𝛽, 𝜆; 𝑥) 1 𝛼
=∑ (𝛼 2 𝜆(𝛽𝑥𝑖 )𝛼 + 𝛼 2 𝛽(1 + 𝜆)(𝛽𝑥𝑖 )𝛼−1 ) − ∑(𝛽𝑥𝑖 )𝛼 = 0⁡ (14)
𝜕𝛽 𝛽𝑊(𝑥𝑖 ) 𝛽
𝑖=1 𝑖=1

Equation (14) can be solved by numerical method that will produce an estimated value of 𝛽, which is 𝛽̂ .⁡

∞ ∞
𝜕𝑙(𝛼, 𝛽, 𝜆; 𝑥) 1 𝑛
⁡= ∑ (𝛼(𝛽𝑥𝑖 )𝛼 ) + (𝛼𝛽(𝛽𝑥𝑖 )𝛼−1 ) + (2𝜆(1 + 𝑥𝑖 )) − ∑ 𝑥𝑖 − = 0⁡ (15)
𝜕𝜆 𝑊(𝑥𝑖 ) 1+𝜆
𝑖=1 𝑖=1
Equation (15) can be solved by numerical method that will produce an estimated value of 𝜆, which is 𝜆̂.⁡
 Data Illustration

In this section, the application of Weibull Lindley distribution in modelling lifetime data of a series system will be
illustrated. The data consists of 61 machines with two independent components paired in series and the machine will
fail if one of the components fails. The data sourced from Tsai-Hung Fan and Tsung-Ming Hsu [5]. The statistical
descriptions of the data are given in Table 1. In Figure 3, we see that the data has 2 modus or called bimodal data.
Therefore, Weibull Lindley distribution which has characteristics of bimodal shaped pdf is suitable for modelling
machine lifetime data. Moreover, several other distributions are also suitable for modelling machine lifetime data such
as Weibull distribution and Lindley distribution. Table 2 contains the estimated values of the parameters using the
maximum likelihood method by nlm function in RStudio version 1.2.5019.

TABLE 1. Descriptive statistics for machine lifetime data

Mean Variance Standard deviation Minimum Maximum

2.51603 2.164 1.4750 0.148 5.44

FIGURE 3. Histogram of machine lifetime data

TABLE 2. Maximum Likelihood Estimation of machine lifetime data

Distribution 𝝀̂ ̂
𝜶 ̂
𝜷

Weibull Lindley 0.50117 7.6775 0.2267

Lindley 0.6398 - -
Weibull - 17.147 0.3556

Figure 4 gives the empirical and fitted cumulative functions. In Figure 5 the graph of Weibull Lindley pdf is
bimodal which resembles the shape of machine data histogram. Otherwise, the pdf of Weibull is unimodal, and the
pdf of Lindley is increasing. In Figure 4. we can see that the cumulative function of Weibull Lindley distribution close
fits the empirical cumulative function.
FIGURE 4. The empirical and fitted cumulative functions FIGURE 5. The pdf and histogram of machine lifetime
data

For the comparison for fitted models distribution Weibull Lindley with the Weibull and Lindley distribution, we
used the Kolmogorov-Smirnov Test Statistics, the Akaike information criterion (AIC) defined by -2log L + 2q, and
the Bayesian information criterion (BIC) defined by -2log L +q log(n), where q is the number of estimated parameters
and n is the sample size. The best model would give by the lowest values from Kolmogorov-Smirnov Test Statistics,
AIC and BIC. From Table 3 we have Weibull Lindley is more suitable for modelling the machine lifetime data than
the other distribution. Figure 6 shows that the hazard function of Weibull Lindley distribution in machine lifetime data
is increasing.

TABLE 3. Summary of fitted model

Kolmogorov
Distribution Smirnov Test AIC BIC
Statistics
Weibull Lindley 0.067171 204.94 211.277
Lindley 0.119717 429.378 429.378
Weibull 0.128655 459.13 463.352

FIGURE 6. The hazard function of Weibull Lindley distribution

 RESULTS AND DISCUSSION

1. From data illustration, we can conclude that Weibull Lindley distribution is more suitable for modelling
machine lifetime data designed in series than Weibull distribution or Lindley distribution.
2. Theoretically, it is obtained that the pdf of Weibull Lindley distribution can be bimodal. It is also shown
in the machine lifetime data that the pdf shape is bimodal.
3. Theoretically, it is obtained that the hazard function of Weibull Lindley can be increasing, decreasing and
bathtub shaped. The shape of Weibull Lindley hazard function only depends on the value of 𝛼 as follows:
• The hazard function is increasing when 𝛼 > 1
• The hazard function is decreasing when 0 < 𝛼 < 1
• The hazard function is bathtub when 𝛼 being close enough to 0 or 𝛼 being close enough to 1.
This can be seen from the application of machine lifetime data that the estimated value of 𝛼 is 7.6775.
Because the estimated value of 𝛼 > 1 so the shape of hazard function is increasing.

CONCLUSION

In this paper, we compound Weibull distribution and Lindley distribution to create Weibull Lindley distribution
by defining the Weibull Lindley random variables as 𝑋 = min⁡(𝑋1 , 𝑋2 ) when 𝑋1 is Lindley random variables and 𝑋2
is Weibull random variables. Some properties of our proposed model Weibull Lindley distribution were derived, i.e.
the pdf can be unimodal, monotonically decreasing, or bimodal also the hazard rate shapes can be decreasing or
increasing, or bathtub shaped. The estimation of parameters was obtained by maximum likelihood method. The results
of simulation on machine lifetime data show that Weibull Lindley is more suitable for modelling the machine lifetime
data than the other distribution.

ACKNOWLEDGMENTS

Authors wishing to acknowledge assistance from various parties, including lecturers and friends at Faculty of
Mathematics and Natural Science of Universitas Indonesia. This work is funded by DRPM Universitas Indonesia.

REFERENCES

1. Adamidis K and Loukas S 1998 A lifetime distribution with decreasing failure rate Statistics & Probability
Letters 39(1) 35-42
2. Asgharzadeh A K, Nadarajah S and Sharafi F 2018 Weibull Lindley distribution REVSTAT Statistical Journal
(16) 87-113
3. H. Rinne, The Weibull distribution: a handbook (Taylor and Francis group, England, 2008).
4. Ghitany M E, Atieh B and Nadarajah S 2008 Lindley distribution and its application Mathematics and computers
in simulation 78(4) 493-506
5. Fan T H and Hsu T M 2004 Statistical inference of a two-component series system with correlated log-normal
lifetime distribution under multiple type-I censoring IEEE Transactions on Reliability 64(1) 376-85