Implementation of Weibull’s Model for

Determination of Aircraft’s Parts Reliability

and Spare Parts Forecast

Nataša Kontrec, Milena Petrović, Jelena

Vujaković, Hranislav Milošević
Faculty of Natural Sciences and Mathematics, University of
Priština, Kosovska Mitrovica
 Content
• Introduction
• Numerical analysis and presentation of
approach
• Method evaluation
• Conclusion
 Introduction
• Optimized maintenance can be used as a key factor in
organizations efficiency and effectiveness.
• Adequate spare parts management in the aircrafts
maintenance system improves the aircraft availability
and reduces downtimes.
• In aviation industry some methods described in papers
[Croston, Rao, Synthetos-Boylan, etc] found their
application but due to stochastic nature of demand they
often failed to provide accurate results.
• In recent times, spare parts forecasting with the respect
to techno-economical issues (reliability, maintainability,
life cycle costs) have been extensively studded [12-14]
but not in aviation industry.
• Here, we present the methodology to forecast
the spare parts needs for non-repairable
aircraft parts.
• The approach is based on stochastic nature of
the parts failure times.
• In order to determine the mean time to failure
for each aircraft part, in this paper we are
using Weibull’s model.
 Weibull’s model
• The Weibull probability density function (pdf) is
given by: 1 t


 w  
 
f ( w)    exp (1)
  
f  w  0, w  0,   0,   0.

w denotes flight hours β denotes shape parameter or slope, η

denotes scale parameter or characteristic life.
• Based on previous, the cumulative distributive function (cdf)
can be determined as given in:

 w
 
 
F ( w)  1  exp (2)
• Reliability function of Weibull model can be calculated as
follows: 
 w
 

R( w)  exp (3)
• The mean time to failure (MTTF) of Weibull pdf can be
determined as in:
1 
MTTF      1 (4)
 
 
where is Gamma function evaluated at  1  1
 

• Failure rate function is given as:

f  w
1
 w
  w     (5)
R  w    
• In order to calculate reliability characteristic of an
aircraft part it is necessary to estimate the
parameters of Webull model. There are several ways
to achieve that, but in the case when we have limited
historical data on previous failures, it is best to
perform rank regression on Y.

• The idea is to bring our function to linear line. In

order to achieve that we are taking natural algorithm
of the both sides of the Eq (2).
 ln 1  F ( w)   1  exp   w / 


ln  ln 1  F ( w)    ln  w /  (6)

Now let:
y  ln[ ln 1  F  w  ]
a   ln  
b (7)

We are getting linear equation and after regression analysis we

can derive following equations:
N N

x y N N

y x
N i i

x y i i  i 1 i 1
i i
bˆ  i 1 N
2
(8) aˆ  i 1
 bˆ i 1 ˆ
 y  bx (9)
 N N N
 i  x
xi2   i 1 
N


i 1 N
According to Eqs. (8) and (9) we can obtain yi
and xi

yi  ln[ ln 1  F  wi  ]
xi   ln  wi  (10)

The F(wi) are values determined from the

median ranks, and after we calculate b̂and â ,
be can easily estimate parameters η and β.
Numerical analysis
First by using Weibull probalbility ploting it was concluded, as
can been seen on Fig1. that data are following Weibull
distribution.
Since the table provide the sample size less than 15 (failed
times), rank regression on y has been used for parameter
estimation. It has been calculated that shape parameter (β) is
4.86 and characteristic life (η) is 6,572.98.
Failure rate
 Method evaluation
• The major contribution of this paper is to determine a
number of the spare parts that should be kept on stock in
interval [0,w]. In order to achieve that we are using an
approach presented in paper [Konterc et al] where the
number of part exposed to failure in certain time frame was
calculated.
• Pdf of Weibull distributed failure time is given by Eq (1), while
the pdf of Rayghly distribute failure time is:
  2 
f ()  2 exp   2  (10)
  2 
• where w in Eq(1) denotes Weibull’s random variable and
denotes Rayleigh’s random variable.
According to above stated equations it can be
 
concluded that  
2 and
w
2
In order to create relation between these models we
are using following transformation:
 2  2 1 
pw,w  w, w  p  w , w w  J (11)
 2 
where J presents Jacobian transformation of random
variables given by following equations:
d d
dw d  2 2
J   w (12)
d d 4
dw dw
So the Eq(11) further transforms into:
2 2
pw, w  w, w  w p  ,   (13)
4
By capitalizing the random nature of failure time of an aircraft
part we are observing the expected number of variations of
Rayleigh’s random variable a within an interval (μ, μ+𝑑μ), for a
given slope within a specified open neighborhood 𝑑μ. The
number of parts that will be exposed to failure can be
determined as:

n  p  ,   d 
0


 2 2
  2 1  2
   2e 2
e 2
d (14)
0
 2 2


  wp  w, w dw
0
ww
According to previous Eqs number of spare parts exposed to
failure in time w can be finally determined as:

w
4 2 w  22
n e (15)

When we calculate average number of parts that are exposed
to failure in interval [0-w], we can determine the number of
parts that should be on inventory. We are using the approach
presented in paper [Kontrec et al.] where:
F  w
q (16)
n
Number of spare part Igniter plug
• The underage costs are difficult to determine due to their
nature. In this paper we are using the well known
Newsvendor method in order to calculate these costs.
 cu 
1
q   (17)
c 
 u oc

• Φ-1 presents inverse distribution function (complementary

error function), cu are underage costs and co are overage
costs, which in our case is spare part price.
• Figure 6 presented the underage costs for aircraft part Ignition
plug. The overage costs for this part are are $1.925,00 and it
can be noticed that the underage costs are growing
exponentially in relation to time.
Underage cost of the part Igniter plug
 Conclusion
• This paper presents an approach to determined reliability
parameters of each aircraft part.
• A new methodology for calculation of number of parts that
are exposed to failure in observed period of time is presented.
• According to obtain number, further has been calculated the
quantity of the aircraft spare parts that should be kept on
stock in order to avoid necessary costs.
• Newsvendor model has been used in order to assess the
potential underage costs in certain time period.
• All these calculations have an aim to support the decision
making process in planning aircraft maintenance activates and
spare parts needs.
References
 Implementation of Weibull’s Model for
Determination of Aircraft’s Parts Reliability
and Spare Parts Forecast

Nataša Kontrec, Milena Petrović Jelena

Vujaković, Hranislav Milošević
Faculty of Natural Sciences and Mathematics, University of
Priština, Kosovska Mitrovica