Hindawi Publishing Corporation

Mathematical Problems in Engineering

Volume 2013, Article ID 594187, 12 pages
http://dx.doi.org/10.1155/2013/594187

Research Article
Bayesian Network Assessment Method for Civil Aviation Safety
Based on Flight Delays

Huawei Wang1 and Jun Gao2

1
College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China
2
Department of Management, Shijiazhuang Mechanical Engineering College, Shijiazhuang 050003, China

Correspondence should be addressed to Huawei Wang; [email protected]

Received 27 October 2013; Revised 28 November 2013; Accepted 29 November 2013

Academic Editor: Fang Zong

Copyright © 2013 H. Wang and J. Gao. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Flight delays and safety are the principal contradictions in the sound development of civil aviation. Flight delays often come up
and induce civil aviation safety risk simultaneously. Based on flight delays, the random characteristics of civil aviation safety risk
are analyzed. Flight delays have been deemed to a potential safety hazard. The change rules and characteristics of civil aviation
safety risk based on flight delays have been analyzed. Bayesian networks (BN) have been used to build the aviation operation safety
assessment model based on flight delay. The structure and parameters learning of the model have been researched. By using BN
model, some airline in China has been selected to assess safety risk of civil aviation. The civil aviation safety risk of BN model has
been assessed by GeNIe software. The research results show that flight delay, which increases the safety risk of civil aviation, can be
seen as incremental safety risk. The effectiveness and correctness of the model have been tested and verified.

1. Introduction aviation have realized that flight delays are an important

potential risk in the management of civil aviation safety. If the
The rapid increase of flight delays has already become a safety risk, which is concurrent and induced, is not controlled
prominent problem in the development of civil aviation in in time, the aviation accident and grievous potential accident
China. It not only affects the service quality and economic will take place under certain conditions. The aviation accident
benefit of civil aviation, but also reduces the civil aviation will bring about heavy losses.
safety level. The flight delays have attracted public attention The civil aviation safety level in China is in the state of
and become one of the key factors in impeding the develop- steady improvement. Now China has already become one
ment of civil aviation in China. of the countries, who have high civil aviation safety level in
The close relationship between flight delays and safety risk the world. However, the number of delay flights in China is
is the two principal contradictions, which seriously affects always very high. In 2012, there were 2,502,000 scheduled
the development of civil aviation in China. The complicated flights, which include 1,872,000 regular flights and 630,000
relation is principally due to the following reasons. Firstly, irregular flights, and the average flight punctuality rate is
the operation and management subject of flight regularity 75.69%. The flight delays are caused 38.5% by airlines, 25.0%
and civil aviation safety are coincident. Secondly, the factors, by flow control, and 21.6% by weather, so it is very severe
such as resource shortage and operation management behav- in the development of civil aviation. The main reason that
ior, can give rise to the flight delays and safety operation can affect civil aviation safety and flight delays is that the
simultaneously. Thirdly, the environmental factors, such as attributes of flight delays are subject to dominant indicators
severe weather and human disturbance, result in not only and the attributes of civil safety are subject to potential
flight delays but also safety risk. At the same time, flight delays indicators. Potential safety risk exists and is in the high state,
cannot only result in potential safety risk of civil aviation which will pose a threat to the safety of civil aviation. It may
but also come up simultaneously. Some specialists in civil lead to the fluctuation of safety level of civil aviation and even
2 Mathematical Problems in Engineering

worse the situation of civil aviation. Therefore, it is necessary severity, to passenger airline risk by analyzing three different
to analyze the effect of flight delays on the safety of civil sources of data from 1999 to 2008. Brooker [13] examined the
aviation in detail, assess the safety level of civil aviation based ability of Bayesian belief network-based techniques to make
on flight delays, and provide measures and suggestions for accurate aviation risk predictions. Hadjimichael [14] devel-
the collaborative management of flight delays and the civil oped a flight operation risk assessment system (FORAS),
aviation safety. which was a risk modeling methodology representing risk
Some scholars had already researched the related prob- factors and their interrelationships as a fuzzy expert system.
lems on flight delays and civil aviation safety. Airline main- A FORAS risk model provided a quantitative risk index
tenance operation affects not only flight delays but also representing an estimate of cumulative effects of potential
civil aviation safety. Sachon and Paté-Cornell [1] built a hazards on a single flight operation. Lee [15] developed a
probabilistic risk analysis model, represented by an influence quantitative model for assessing aviation risk factors as a
diagram, to quantify the effect of airline’s maintenance policy means of increasing the effectiveness of safety risk manage-
on flight delay, cancellation, and in-flight safety. McCrea et ment system by integrating the fuzzy linguistic scale method,
al. [2] made a novel severe-model paradigm to be applied for failure mode, effects and criticality analysis principle, and
the context of a large scale in specified probability threshold as low as reasonably practicable approach. Rose [16] studied
when encountering severe weather, subject to collision safety, the limitations and problems of trying to measure safety
airline equity, and sector workload considerations. More and operational risk and presented useful metrics from
research is to study flight delay and the aviation safety incident reporting data. Oster Jr. et al. [17] brought forward
problem, respectively. In the respect of flight delay, Zheng et that the next generation of safety challenge now required
al. [3] calculated the time series of after-affect delay spread development and understanding of new forms of data to
time using delay time distribution of the flights assigned gate improve safety in other segments of commercial aviation,
in airport and random delay in flights. Wong and Tsai [4] moving from a reactive, incident-based approach toward a
used coproportional hazards model to build departure and more proactive, predictive, and system-based approach.
arrival delay models that show how flight delay propagation From the above research literature, we understand that
can be formulated through repeated chain effects in aircraft there is little study in the systematic analysis of flight delays,
rotations. Yan and Tang [5] developed a heuristic approach which have an effect on civil aviation safety and safety
embedded in a framework designed to help the airport assessment.
authorities make airport gate assignments that are sensitive The complexity of civil aviation safety risk assessment
to stochastic flight delays. Pyrgiotis et al. [6] developed an based on flight delays shows in the following aspects.
analytical queuing and network decomposition model to
study the complex phenomenon of the propagation of delays (i) There are many factors that affect flight delays and
within a large network of major airports, which has been civil aviation safety risk. Coupling with the interac-
used to compute the delay due to local congestion at indi- tion between flight delays and civil aviation increases
vidual airports and captures the “ripple effect” that leads to the difficulty of analysis.
propagation of these delays. Santos and Robin [7] found that (ii) The mechanism of flight delays, which can erupt
four significant variables in explaining delays at European simultaneously and induce civil aviation safety,
airports were market concentration, slot coordination, hub remains to be further analyzed in detail.
airports, and hub airlines using flight data for the period (iii) The rate of civil aviation accident is low. But the
2000–2004. Using longitudinal data from a major airline and increase of potential safety hazard and the deviation
conditional difference-in-differences technique, Ferrer et al. of safety state may not arise in the shape of accident,
[8] analyzed the effects of flight delay on passengers’ future accident threat, or unsafe event immediately.
purchasing behavior. To develop a decision-support tool for
air traffic control, presented an algorithm for optimal arrival (iv) It is difficult to collect the direct statistics of the
flight sequencing and spacing in near-terminal, which used relevant factors. So, it is often completed with the help
discrete delay times as optimization variables. of subjective information by experts’ experience.
In the field of civil aviation safety, some scholars utilized Therefore, the difference between aviation safety risk
data on both accident and safety indicators. Shyur [9] based on flight delays and general aviation risk assessment
presented a specified proportional hazard model considering can be concluded.
the baseline hazard function as a quadratic spline function,
which had investigated and demonstrated its applicability in (i) Change of Assessment Elements. The general civil aviation
aviation risk assessment. Janic [10] described the main cause safety assessment only focuses on the elements of safety itself.
of aircraft accident, proposed a methodology for quantifying The safety assessment of civil aviation based on flight delays
risk and safety, and offered an assessment of risk and safety in is one of the assessment elements, which not only reflects
civil aviation. Chen et al. [11] ranked the significant threats the value change of safety elements, but also embodies the
and human errors affecting aviation safety and used the relationships between the civil aviation safety system and
analytical hierarchy process to calculate the weigh for each flight delay system.
criterion which were than ranked in order of importance.
Marais and Robichaud [12] investigated and quantified the (ii) The Difference of Definition the Analysis Scope. Because
contribution of maintenance, both in terms of frequency and both the flight delays system and civil aviation safety system
Mathematical Problems in Engineering 3

are big systems, it is difficult or insoluble for civil aviation Characteristics of aviation safety risk
safety assessment based on flight delays if all the elements of evaluation based on flight delays
flight delay system and civil aviation system are considered.
So, emphasis is given to civil aviation safety elements associ-
ated with flight delays, which is different from general civil
aviation safety assessment. The composition of aviation
safety risk based on flight delays Method selection
(iii) The Difference of Risk Evolution. The elements of flight
delay system are dynamic, which is real-time change. The risk
elements of civil aviation safety system are relatively stable. The risk composition of aviation
safety based on flight delays Aviation safety risk
So, the civil aviation safety assessment based on flight delays evaluation based on flight
is dynamic, while the general civil aviation safety assessment delay using Bayesian
is relatively stable. The risk mechanism of aviation networks (BN)
The main contribution of the paper can be concluded as safety based on flight delays
follows.
(i) The composition of civil safety risk based on
flight delays has been analyzed systematically, which BN structure of aviation
includes the aggregation and the transmission of civil safety risk evaluation based
on flight delays
aviation safety risk;
(ii) The change rules of civil aviation safety risk based
Data collection and expert
on flight delays have been analyzed, which include information extraction
the increase of potential safety risks leading to civil
aviation safety risk, the common causes of safety risk
and flight delays leading to aviation safety risk, and
flight delays leading to the increase in civil aviation
BN structure learning BN parameters learning
safety risk degree as an inducement;
(iii) The safety assessment model of civil aviation by
BN, which considered the composition of civil safety
risk based on flight delays, the randomness of civil BN inference for aviation
aviation safety risk variation, the change rules of civil operation safety risk based on
aviation safety risk based on flight delays. flight delays

Therefore, according to the characteristics of civil avia- Aviation operation safety

tion, the paper presents an assessment model of civil aviation evaluation based on flight
delays
safety risk based on flight delays. The remaining part of the
paper is organized as follows. Section 2 introduces the frame Figure 1: The diagram of civil aviation safety risk assessment based
structure of algorithm. Section 3 analyzes the characteristics on flight delays.
and change regulation of civil aviation safety risk based
on flight delay. Section 4 builds an assessment model of
civil aviation operation safety risk by the Bayesian network. (iii) The action mechanism and evolution law between
Section 5 uses example to validate the effectiveness of the civil aviation safety risk, which is induced by flight
proposed method. Finally, some concluding remarks are delays, and civil aviation operation system itself
made. should be analyzed.
Figure 1 shows the diagram of civil aviation safety risk
2. The Risk Assessment Framework of Civil assessment based on flight delays.
Aviation Safety Based on Flight Delays
There are three criteria in building risk assessment framework 3. The Civil Aviation Safety Risk Based on
of civil aviation safety in the paper. Flight Delays
(i) Considering the concurrence of subjective and objec-
3.1. The Composition of Civil Aviation Safety Risk Based on
tive data in risk assessment of civil aviation safety, the
mechanism of action and dynamic change reflected Flight Delays
by some factors in civil aviation system operation
should be considered in building the model. (i) The Aggregation of Civil Aviation Safety Risk. The aggrega-
(ii) Flights delay should be seen as an incremental risk tion of civil aviation safety risk means that the actual risk of
and be analyzed based on the risk analysis of civil civil aviation safety consists of coinstantaneous and induced
aviation safety. risks under flight delays. There are three main types of risk:
4 Mathematical Problems in Engineering

Risk convergence
Direct risk
Independent safety Safety risk degree led
Safety risk degree
influence factors by influence factors
Direct risk
Common factors
Common factors lead to Transitive risk
Safety risk degree
flight delays and safety

Risk propagation
The conditional risk probability
Flight delays level based on the flight delays rate and
spread range

Independent influence The flight delays rate and

Flight delays level
factors of flight delays propagation range

Figure 2: The entire constitute of civil aviation safety risk.

one is caused by the independent influence factors of civil (iv) As a special safety risk, the influence of flight delays
aviation safety; the second is caused by the common factors on civil aviation safety risk is random. That is to say,
which lead to flight delays and civil aviation safety risk; the the potential safety hazard which supervenes safety
third is caused by independent influence factors of flight risk with hidden dangers is random. The concurrence
delays. probability is random, which is based on different
delay rate and risk level.
(ii) The Transmission of Civil Aviation Safety Risk. The trans-
(v) As a special safety risk, there is a causal relationship
mission of risk is from two aspects: one is the variation of the
between different potential safety hazards to some
probability distribution of civil aviation safety risk according
extent. That is to say, as a potential safety hazard, it
to different flight delays; the other is that the surging number
is uncertain and random that flight delays can induce
of flight delay is caused by flight delays accumulation. The
and cause the variation of other dangers.
incensement of flight delays rate leads to the accumulation
of civil aviation safety risk. Obviously, the civil aviation safety (vi) The influence of flight delays on civil aviation safety
risk induced by flight delays changes with the rate and spread risk can be regarded as an increment risk, which
range of flight delays. belongs to the interaction between gradual accumu-
The entire constitute of civil aviation safety risk is shown lation and sudden surge of different random factors.
in Figure 2.
In conclusion, the source of random risk can be summa-
rized as follows:
3.2. The Randomness of Civil Aviation Safety Risk Variation.
The civil aviation safety risk based on flight delays is random. (i) the factors which induce flight delays and civil avia-
The randomness is reflected as follows. tion safety are random;
(ii) the conditions which lead to flight delays and civil
(i) Because the aviation accident is random, the civil avi- aviation safety are random;
ation safety risk assessment is supposed to obey
(iii) the state of flight delays and civil aviation safety is
statistical probability distribution.
random.
(ii) The appearance of flight delays is occasional. The randomness between factors, conditions, and states
Although flight delays can be controlled to some can be expressed in Figure 3.
extent, some flight delays caused by bad weather are
generally unforeseeable and unavoidable. Therefore,
some factors causing flight delay are random. 3.3. The Analysis on Change Rules of Civil Aviation Safety Risk
Based on Flight Delays
(iii) The occurrence of potential safety hazard factor is
random. The potential safety hazard often appears as 3.3.1. The Increase of Potential Safety Risks Leading to Civil
hidden factor, which is difficult to detect and will lead Aviation Safety Risk. Flight delays are regarded as a kind of
to safety risk when it is transformed into the dominant new safety risk. The characteristics of flight delays are that it
factor under some conditions. exists when flight delay exists; otherwise, it disappears.
Mathematical Problems in Engineering 5

The random The random The random

influence factor on state of civil aviation triggering condition
civil aviation led to safety risk
safety safety

The random
influence factors on The random influence
on civil aviation safety
flight delays and led by flight delays
safety risk

The random The random

influence factors on The random of triggering condition
flight delays led to flight delays
flight delays

Figure 3: The diagram of random risk of civil aviation safety based on flight delay.

SRn SRn

SR2
SR2
ASR
SR1 ASR
SR1

ASRCFD

SR: safety risk

ASR: aviation safety risk
ASRCFD: aviation safety risk based on flight delays

Figure 4: The change rule of civil aviation safety risk considering flight delays as potential safety hazard.

Assuming that there are 𝑛 hidden risks (𝑛 risk sources) (iv) As the change rule of potential safety risk in Figure 4,
when there are no flight delays, there are 𝑛 + 1 hidden risks the paper has taken the randomness of risk change
when flight delays occur. Even assuming that flight delays into account and assumes that the variation of risk
do not affect other hidden risks (e.g., civil aviation safety obeys normal distribution.
risk degree, which is caused by 𝑛 hidden dangers, does not
vary), the 𝑛 + 1 hidden risks, including flight delay and safety 3.3.2. The Common Causes of Safety Risk and Flight Delays
risk induced by flight delays, will increase the degree of civil Leading to Aviation Safety Risk. The factor which causes flight
aviation safety risk. delays has an impact on civil aviation safety at the same
The change rules of safety risk degree are shown in time. Assuming that there are 𝑛 potential safety risks (𝑛 risk
Figure 4. sources) when flight delays do not occur. The common factors
not only lead to flight delays but also have an influence on the
(i) Assuming that potential safety risk and flight delays risk variation rules of primary potential safety risk 𝑛. That is
are independent when delays flights are acting as a to say, flight delays are not the direct reason which leads to
potential safety risk; the increase of civil aviation safety risk.
(ii) The reason that safety risk increases is that flight The mechanism of risk is shown in Figure 5.
delays act as an independent potential safety risk; According to Figure 5, the explanations include the fol-
(iii) Regular flight is a normal state, and flight delays are lowing.
abnormal state. The existence of flight delays can lead (i) The risk caused by flight delays which act as an
to the increase of civil aviation risk inevitably; this is independent potential safety hazard should not be
a systemic risk; taken into account.
6 Mathematical Problems in Engineering

SRn
SRn
SR1 SR2 SR1 SR2
ASR ASR

SR: safety risk

ASR: aviation safety risk

Figure 5: The change rule of civil aviation safety risk caused by the common causes of civil aviation and flight delays.

SRn
SRn
SR2
SR1 SR2 SR1
ASR
ASR

ASRCFD ASRCFD

SR: safety risk

ASR: aviation safety risk
ASRCFD: aviation safety risk based on flight delays

Figure 6: The change rule of civil aviation safety risk under flight delay acting as inducement.

(ii) Figure 5 aims to analyze the influence on common overlapped between the above two kinds of influence and
factors of safety risk and flight delay on civil aviation the superposition may lead to a sound increase in safety risk
safety risk. Compared with flight delay, the indica- degree objectively.
tions of safety risk (accident, accident symptom, and Flight delays acted which as inducement increased the
unsafe events) are much less sensitive to flight delay. risk degree of civil aviation safety, and the change rule of risk
That is to say, flight delay is the explicit index of is showed in Figure 6.
common factors, and safety risk is shown as implicit From Figure 6, we can draw some conclusions.
index, usually.
(i) Flight delays as an inducement in Figure 6 directly
(iii) The common factors of safety risk and flight delay have an impact on potential safety risk, which acted
lead to various change rules of risk on 𝑛 safety as the occurrence of flight delay and stimulates and
potential safety hazard. Although the degree, time, triggers the change of safety risk degree.
and variation rules are different, the entire trend is to
increase the civil aviation risk. (ii) As the change rules of different potential safety risk
in Figure 4, the paper has taken the randomness of
(iv) Considering the randomness of risk change, the paper risk change into account and assumes that risk change
assumes that risk change obeys normal distribution obeys the normal distribution.
and the civil aviation safety risk sources are composed
of different risk sources. It is worth mentioning that the risk change rule of civil
aviation safety in Figures 4 and 6 has been assumed to
3.3.3. Flight Delays Leading to the Increase of Civil Aviation obey normal distribution. The assumption is from general
Safety Risk Degree as an Inducement. In addition to the knowledge about civil aviation safety risk, which is not
common factors, flight delay as an inducement will further carefully calculated. The detailed calculation about safety risk
have an effect on civil aviation. The actual safety risk is of civil aviation can be seen in Section 4.
Mathematical Problems in Engineering 7

4. Risk Assessment Model of reflect the actual civil aviation safety. The nodes determined
Civil Aviation Safety Based on BN in Step 1 are connected with causality. So, the structure of
BN consisted of causality chain, which can be required from
4.1. Overview of BN. BN is a directed acrylic graph for logic analysis and expert experience. The determination of
reasoning under uncertainty in which the nodes represent BN structure is the qualitative modeling in BN. The analyses
variables and are connected by means of directed arcs. The of Sections 2 and 3 are used to build the BN structure of civil
arcs denote dependencies or causal relationship between the aviation safety risk.
link nodes and the conditional probability table assigned to
the nodes determining how the linked nodes are dependent Step 3 (instantiate the BN with probabilities). First, prior
on each other [18]. probabilities are assigned to the root nodes in the BN for
Based on the conditional independency theorem and the civil aviation safety risk assessment. To root node 𝑋𝑖 , 𝑃(𝑥𝑖 =
chain rule, BN factorizes the joint probability distribution of a 𝑅) and 𝑃(𝑥𝑖 = 𝑁) represent risk probability and normal
set of random variables 𝑈 = {𝐴 1 , 𝐴 2 , . . . , 𝐴 𝑛 } by considering probability of node 𝑥𝑖 , respectively, and 𝑃(𝑥𝑖 = 𝑅) + 𝑃(𝑥𝑖 =
local dependencies. The joint probability distribution can be 𝑁) = 1.
decomposed as the product of the probabilities of the nodes Next, conditional probabilities are assigned to other
given their immediate parents [19]: nodes. To a child node with 𝑛 parents, 2𝑛 conditional
probabilities should be assigned. It is difficult to collect much
𝑛
information in actual civil aviation operation and manage-
𝑃 (𝑈) = ∏𝑃 (𝐴 𝑖 | 𝑃𝑎 (𝐴 𝑖 )) , (1) ment. Experts also have some difficulty in providing so much
𝑖=1
information. To overcome the problem, it is necessary to
where 𝑃(𝑈) is the joint probability distribution of variables study the BN for civil aviation safety risk assessment for
and 𝑃𝑎(𝐴 𝑖 ) is the parent set of variable 𝐴 𝑖 . simplifying the conditional probabilities assignment.
BN takes advantage of the Bayes theorem to update the
probability of events given a new observation, called evidence Step 4 (learn BN structure). There are two contents in BN
𝐸, to yield the posterior probability: structure learning. One way is to decide BN structure by data
reasoning. The other way is to verify structure of BN and
𝑃 (𝑈, 𝐸) 𝑃 (𝑈, 𝐸) delete weak connections between nodes by data if the initial
𝑃 (𝑈 | 𝐸) = = . (2) structure of BN has been known. In the paper, the initial
𝑃 (𝐸) ∑𝑈 𝑃 (𝑈, 𝐸)
BN structure of civil aviation safety risk assessment has been
Using (2) probability can be updated by versatile type of 𝐸. decided based on the study of Sections 2 and 3. The emphasis
BN is the method integrated by quality analysis and quan- of BN on civil aviation safety risk assessment is to simplify BN
titative analysis, which can utilize many kinds of information. structure. Markov blanket is selected to BN in civil aviation
BN has the function of learning itself, backward reasoning, operation risk.
and inference in incomplete data sets. The above advantages
make BN superior to other artificial intelligence methods in Step 5 (learn BN parameters). Bayes method is the theory
safety evaluation of civil aviation based on flight delays. basis for BN. Bayes method uses prior density and posterior
BN has been widely applied in engineering domain, such density to learn and assess parameters. BN also uses the above
as risk analysis [20–22] and fault diagnosis [23, 24] and process to learn parameters after collecting and accumulating
prediction [25, 26]. relevant data. In the practical application, the parameters
learning of BN also use conjugate prior to simplify parameters
learning.
4.2. The BN Model for Civil Aviation Safety Risk Assessment.
There are six steps in BN model to assess the civil aviation Step 6 (infer BN). Civil aviation safety risk can be assessed
safety risk. by BN if the BN structure has been known and the nodes
have been assigned. In the paper, GeNIe has been selected to
Step 1 (extract the key factors causing flight delays and civil compute because of large amount of calculation.
aviation safety risk and determine the BN nodes). Determi-
nation of nodes is the foundation and key for determining
the structure of BN. There are four categories of nodes to 4.3. The BN Structure of Civil Aviation Safety Risk Assessment.
be taken into account. First, the common cause factors for The analysis on designing BN structure of civil aviation safety
flight delay and civil aviation safety should be taken into risk assessment bases on flight delay.
account. Second, the independent causing factors for flight
delays and civil aviation safety should be taken into account. (i) Civil aviation safety system is the complex large
Third, the factors of flight delays which induced safety risk of system, which includes flight subsystem, maintenance
civil aviation operation should be taken into account. Forth, subsystem, air traffic control subsystem, and airport
the prorogation and spread of flight delay itself should be subsystem. These subsystems are affected by flight
taken into account. delays in different degree. These subsystems also have
an impact on civil aviation safety risk in different
Step 2 (determine the risk assessment BN structure of civil mode. So, it is necessary to use independent analysis
aviation operation based on flight delays). The model should for different subsystems.
8 Mathematical Problems in Engineering

Table 1: The description of node characteristics. 4.4. BN Learning for Civil Aviation Safety Risk Assessment
Meaning Abbreviation 4.4.1. BN Structure Learning of Civil Aviation Safety Risk
Weather WE Assessment. A novel idea of BN of significant feature selec-
Flow control FC tion is the Markov blanket (MB). In order to simplify BN
ME structure of civil aviation safety risk assessment, Markov
Maintenance and engineering
blanket is selected in BN structure learning.
Airplane plan AP
MB is defined as the set of input features because all
Airport design AD the other features are probabilistically independent of target
Airport order AO features. In a Bayesian network, the Markov blanket [27] of a
Scheduled flight time SFT node 𝑋𝑖 (which can be represented as MB(𝑋𝑖 ) in short) is the
ATC safety risk ATSR set of nodes which is composed of its parent nodes, its child
ALSR nodes, and parent nodes of its child nodes. The important
Airline safety risk
property of Markov blanket is that Markov blanket of variable
Airport safety risk APSR
𝑋𝑖 is the set of nodes which makes 𝑋𝑖 independent in the
Maintenance safety risk MSR network. That is to say, the nodes belonging to the Markov
Airport size AS blanket of 𝑋𝑖 are the most relevant to 𝑋𝑖 ,
Food supply FS
𝑃 (𝑋𝑖 | 𝑋1 , . . . , 𝑋𝑖−1 , 𝑋𝑖+1 , . . . , 𝑋𝑛 ) = 𝑃 (𝑋𝑖 | MB (𝑋𝑖 )) .
Other reason OR
(3)
Traveler TR
Aircraft type AT The probability of the target node is influenced only by
FD its Markov blanket. Therefore, MB (target node) is the most
Flight delay
informative and relevant to risk assessment of civil aviation
Station flight delay SFD
safety.
Flight delay rate FDR For example,
Risk rate of aviation operation safety AOSRR
SRBFD 𝑃 (AOSRR | MB (AOSRR))
Safety operation risk based on flight delay
ATC safety hidden danger ATSHD = 𝑃 (AOSRR | SFD, WE, FFD, ALSR, APSR, ARBR, MSR)
Airline safety hidden danger ALSHD
𝑃 (AOSSR, SFD, WE, FFD, ALSR, APSR, ARBR, MSR)
Airport safety hidden danger APSHD =
𝑃 (SFD, WE, FFD, ALSR, APSR, ARBR, MSR)
Maintenance safety hidden danger MSHD
Prohibited fly PF ∏𝑋𝑖 ∈{SFD,WE,FFD,ALSR,APSR,ARBR,MSR} 𝑃 (𝑋𝑖 | 𝜋 (𝑋𝑖 ))
=
Air crew AC 𝑃 (SFD, WE, FFD, ALSR, APSR, ARBR, MSR)
Security check SC
JC
= 𝑃 (AOSRR | 𝜋 (AOSRR))
Joint check
Transport service TS
⋅ ∏ 𝑃 (𝑋𝑖 | 𝜋 (𝑋𝑖 ))
𝑋𝑖 ∈Children(AOSRR)

(ii) Flight subsystem, maintenance subsystem, air traf- ⋅ ∏ 𝑃 (𝑋𝑗 | 𝜋 (𝑋𝑗 ))

fic control subsystem, and airport subsystem are 𝑋𝑗 ∉AOSRR∩𝑋𝑗 ∉Children(AOSRR)
all affected by person, machine, environment, and
management. To avoid large BN structure, modular × (𝑃 (SFD, WE, FFD, ALSR, APSR, ARBR, MSR))−1 .
structure has been introduced in the paper. The nodes (4)
of hidden risk and safety risk have been designed
in different subsystems, the value of which can be 4.4.2. BN Parameters Learning for Aviation Safety Risk Assess-
calculated by BN inference. ment. In the BN of civil aviation safety risk assessment,
assuming that node has a multinomial distribution, we can
(iii) In order to analyze the flight delays which propagated parameterize it with the parameter vector 𝜃1 , 𝜃2 , . . . , 𝜃𝑆 , where
safety risk, the flight delays modeling should be 𝑆 is the number of states of variable 𝑥 and 𝜃(𝑘) = 𝑃(𝑥 = 𝑥𝑘 |
emphasized, which include influence factors such as 𝑝), for 1 ≤ 𝑘 ≤ 𝑆.
flight delays, flight delays propagation, flight delays (1) 𝜃 possess the Dirichlet distribution
rate, and risk assessment based on flight delays.
𝜃 ∼ 𝐷 [𝛼1 , 𝛼2 , . . . , 𝛼𝑆 ] (5)

The initial risk assessment BN of civil aviation safety is 𝛼𝑖 > 0; 𝑖 = 1, . . . , 𝑆, and ∑𝑆𝑖=1 𝜃𝑖 = 1, we can act 𝛼𝑖 as
shown in Figure 7. The meaning of nodes in Figure 7 is shown presenting counts of past cases which are stored as a summary
in Table 1. of experience.
Mathematical Problems in Engineering 9

SFT AT FDR

SFD APSHD MSHD

SC FS ME AD AO AS APSR MSR

JC FFD WE AOSRR SRBFD

OR AC PF FC TS AP ALSR ATSR

TR ALSHD ATSHD

Figure 7: BN structure of civil aviation safety operation assessment based on flight delay.

(2) Let 𝛼0 = ∑𝑆𝑖=1 𝛼𝑖 and let expected value, variance, and The data comes from the following sources.
covariance of Dirichlet distribution be defined as follows: (i) Some data is obtained by statistics analysis, such as
𝛼
𝐸 (𝜃𝑖 ) = 𝑖 , FC node. The FC percentage can be calculated by
𝛼0 statistics data.
𝛼𝑖 (𝛼 0 −𝛼𝑖 ) (ii) Some data is difficult to collect directly, which is
Var (𝜃𝑖 ) = , (6) inferred by historical data and expert information.
𝛼02 (𝛼0 + 1)
(iii) Some data is obtained by other ways and analysis
𝛼𝑖 𝛼𝑗 methods, such as hinder risk of airport. Because com-
Cov (𝜃𝑖 ) = − .
𝛼02 (𝛼0 + 1) putation on hinder risk of airport is very complex, the
result of airport safety audit of last year can be used to
We assume that 𝜃 prior distribution is Dirichlet distribu- reduce the computation difficulty.
tion; then posterior distribution is also Dirichlet distribution;
they are conjunction distribution. In the paper, calculation of time interval is a month. The
When the samples are 𝑁 = (𝑁1 , 𝑁2 , . . . , 𝑁𝑆 ), the data of 2011 can be used as prior information. After collecting
posterior expected value of 𝜃𝑘 with 𝑘 = 1, 2, . . . , 𝑆 is given the data of January 2012, the marginal posterior probability
by of each node in the risk assessment BN of civil aviation
safety can be calculated for January 2012. Structure learning
𝛼𝑖 + 𝑁𝑖 and parameter learning of BN are in the whole calculation
𝐸 (𝜃𝑖 | 𝛼, 𝑁) = . (7)
∑𝑆𝑖=1 (𝛼𝑖 + 𝑁𝑖 ) process. Then posterior probability of each node in January
2012 can be used as the prior information of February 2012.
5. Examples After collecting data of February 2012, the marginal posterior
probability of each node in the risk assessment BN of civil
The paper selects an airline company in China to verify aviation safety can be calculated for February 2012. In the
the model in the paper. The relevant data in Figure 7 were same way, the posterior marginal probability of each node in
collected in 2012. every month of 2012 can be calculated.
10 Mathematical Problems in Engineering

AT FDR SRBFD
SFT
ME High 10%
Delay 10% Delay 34% Risk 4%
Delay 5% Medium 30%
Normal 90% ✓ Normal 66% ✓ Normal 96% ✓
Normal 95% ✓ Low 60% ✓

APSHD MSHD
FS SFD Risk 3% Risk 2%
Delay 24%
Delay 0% Normal 97% ✓ Normal 98% ✓
Normal 100% ✓ Normal 76%

AS AO AD
SC High 0% APSR MSR
High 1% High 0%
Delay 1% Medium 1% Risk 4% Risk 7%
Medium 1% Medium 1%
Normal 99% ✓ Low 98% ✓ Low 99% ✓ Low 99% ✓ Normal 96% ✓ Normal 93% ✓

OR FFD WE AOSRR
Delay 2% Risk Risk 8% Risk 3%
24%
Normal 98% ✓ Normal 76% ✓ Normal 92% ✓ Normal 97% ✓

AC TS AP ALSR ATSR
Delay 3% Delay 2% Delay 3% Risk 3% Risk 6%
Normal 97% ✓ Normal 98% ✓ Normal 97% ✓ Normal 97% ✓ Normal 94% ✓

TR FC ATSHD
High 12%
JC PF ALSHD
High 5%
Delay 2% Delay 3% Risk 2% Risk 5%
Medium 13%
Medium 10%
Low 75% ✓ Normal 98% ✓ Normal 97% ✓ Normal 98% ✓
Normal 95% ✓
Low 85% ✓

Figure 8: Posterior marginal probability of each node in January 2012.

4.5 40
4 35
Flight delays rate

3.5 30
3 25
Risk rate

2.5 20
2 15
1.5
10
1
5
0.5
0 0
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12
Month Month

Figure 9: The safety risk probability of civil aviation operation in Figure 10: The civil aviation safety risk rate based on flight delay in
2012. 2012.

For example, the posterior margin probability of nodes in the same. When flight delays increase dramatically, safety risk
January 2012 is shown in Figure 8 probability greatly increases. So it is necessary to pay close
From Figure 8, 𝑃(AOSRR = 𝑅) = 2.960% represents the attention to safety risk of civil aviation operation caused by
safety risk of civil aviation operation system itself. 𝑃 flight delays.
(SRBFD = Risk) = 4.241% represents the safety risk of civil The following conclusions can be drawn by the above
aviation system based on flight delays. From the result in results to strength the safety management of civil aviation in
Figure 8, fight delays substantially increase in the safety risk China.
probability of civil aviation, which increase the safety risk
probability by 43.77%.
The safety risk probability of civil aviation for each (i) Fight delays are not related to the service quality and
month in 2012 has been calculated in Figure 9. Flight delays operation efficiency of civil aviation but do have a
probability of civil aviation for each month in 2012 has been relatively great impact on civil aviation safety. The
calculated in Figure 10. From Figures 9 and 10, the results trend of flight delay rate and civil aviation risk rate
demonstrate that there is a closer relationship between flight is the same, which illustrates that flight delays have
delays and safety risk of civil aviation; the trend is basically effect on civil aviation safety.
Mathematical Problems in Engineering 11

(ii) Partial factors which caused flight delays are control- [7] G. Santos and M. Robin, “Determinants of delays at European
lable. The other factors which caused flight delays airports,” Transportation Research B, vol. 44, no. 3, pp. 392–403,
are not only uncontrollable but also unpredictable, 2010.
which determines the difficulty of managing flight [8] J.-C. Ferrer, P. Rocha e Oliveira, and A. Parasuraman, “The
delays in China. With the development of Chinese behavioral consequences of repeated flight delays,” Journal of
civil aviation, the number of arrival and departure Air Transport Management, vol. 20, pp. 35–38, 2012.
flights grows rapidly. If the managers cannot manage [9] H.-J. Shyur, “A quantitative model for aviation safety risk
the flight delays effectively, the rate of flight delays assessment,” Computers and Industrial Engineering, vol. 54, no.
will increase and the effect of flight delays will spread, 1, pp. 34–44, 2008.
which will lead to huge pressure on civil aviation [10] M. Janic, “An assessment of risk and safety in civil aviation,”
safety. Journal of Air Transport Management, vol. 6, no. 1, pp. 43–50,
2000.
(iii) It is an efficient way for strengthening the civil avi- [11] C.-C. Chen, J. Chen, and P.-C. Lin, “Identification of significant
ation by details management in China. The element threats and errors affecting aviation safety in Taiwan using
state change in civil aviation safety system should be the analytical hierarchy process,” Journal of Air Transport
avoided and improved the safety management level of Management, vol. 15, no. 5, pp. 261–263, 2009.
civil aviation in China. [12] K. B. Marais and M. R. Robichaud, “Analysis of trends in
aviation maintenance risk: an empirical approach,” Reliability
Engineering and System Safety, vol. 106, pp. 104–118, 2012.
6. Conclusions
[13] P. Brooker, “Experts, Bayesian belief networks, rare events and
The paper analyzes the relationship between flight delays aviation risk estimates,” Safety Science, vol. 49, no. 8-9, pp. 1142–
and civil aviation safety risk. The problem of flight delays 1155, 2011.
accompanied and propagated with civil aviation safety risk [14] M. Hadjimichael, “A fuzzy expert system for aviation risk
has been presented. The problem of flight delays leading to assessment,” Expert Systems with Applications, vol. 36, no. 3, pp.
propagation and superposition of civil aviation safety risk 6512–6519, 2009.
has been analyzed. The BN has been used to model safety [15] W.-K. Lee, “Risk assessment modeling in aviation safety man-
risk assessment of civil aviation safety. The example of some agement,” Journal of Air Transport Management, vol. 12, no. 5,
pp. 267–273, 2006.
airlines in China has demonstrated the effectiveness and
correctness of method. The countermeasures and suggestions [16] A. Rose, “Measuring operational safety in aviation,” Aircraft
Engineering and Aerospace Technology, vol. 78, no. 1, pp. 26–31,
have been put forward to solve the problems of Chinese civil
2006.
aviation safety in the process of rapid development.
[17] C. V. Oster Jr., J. S. Strong, and C. Kurt Zorn, “Analyzing
aviation safety: problems, challenges, opportunities,” Research
Acknowledgments in Transportation Economics, vol. 43, no. 1, pp. 148–164, 2013.
[18] F. V. Jensen and T. D. Nielson, Bayesian Networks and Decision
This work was supported by the National Natural Science Graphs, Springer, New York, NY, USA, 2nd edition, 2007.
Foundation and Aviation Fund of China (no. 60879001 and [19] J. Cheng, R. Greiner, J. Kelly, D. Bell, and W. Liu, “Learning
no. U1233115) and the Qinglan Project. Bayesian networks from data: an information-theory based
approach,” Artificial Intelligence, vol. 137, no. 1-2, pp. 43–90,
References 2002.
[20] P. Weber, G. Medina-Oliva, C. Simon, and B. Iung, “Overview
[1] M. Sachon and E. Paté-Cornell, “Delays and safety in airline on Bayesian networks applications for dependability, risk analy-
maintenance,” Reliability Engineering and System Safety, vol. 67, sis and maintenance areas,” Engineering Applications of Artificial
no. 3, pp. 301–309, 2000. Intelligence, vol. 25, no. 4, pp. 671–682, 2012.
[2] M. V. McCrea, H. D. Sherali, and A. A. Trani, “A probabilistic [21] D. Zhang, X. P. Yan, Z. L. Yang et al., “Application of Bayesian
framework for weather-based rerouting and delay estimations belief networks in navigation risk estimation of the Yangtz
within an Airspace Planning model,” Transportation Research river,” Reliability Engineering and System Safety, vol. 118, pp. 93–
C, vol. 16, no. 4, pp. 410–431, 2008. 105, 2013.
[3] P. Zheng, S.-J. Hu, and C. Zhang, “Chaotic phenomena and [22] M. J. Akhtar and I. B. Utne, “Human fatigue’s effect on the
quantitative analysis of aftereffect delay spread time on flights,” risk of maritime groundings—a Bayesian network modeling
Journal of Transportation Systems Engineering and Information approach,” Safety Science, vol. 62, pp. 427–440, 2014.
Technology, vol. 10, no. 4, pp. 68–72, 2010. [23] Y. Zhao, F. Xiao, and S. W. Wang, “An intelligent chiller fault
[4] J.-T. Wong and S.-C. Tsai, “A survival model for flight delay detection and diagnosis methodology using Bayesian belief
propagation,” Journal of Air Transport Management, vol. 23, pp. network,” Energy and Buildings, vol. 57, pp. 278–288, 2013.
5–11, 2012. [24] C. H. Lo, Y. K. Wong, and A. B. Rad, “Bond graph based
[5] S. Yan and C.-H. Tang, “A heuristic approach for airport gate Bayesian network for fault diagnosis,” Applied Soft Computing
assignments for stochastic flight delays,” European Journal of Journal, vol. 11, no. 1, pp. 1208–1212, 2011.
Operational Research, vol. 180, no. 2, pp. 547–567, 2007. [25] S. Y. Sohn and A. S. Lee, “Bayesian network analysis for
[6] N. Pyrgiotis, K. M. Malone, and A. Odoni, “Modelling the dynamic prediction of early stage entrepreneurial activity
delay propagation within an airport network,” Transportation index,” Expert Systems with Applications, vol. 40, no. 10, pp.
Research C, vol. 27, pp. 60–75, 2013. 4003–4009, 2013.
12 Mathematical Problems in Engineering

[26] Y. F. Xu, J. Choi, S. Dass et al., “Efficient Bayesian spatial

prediction with mobile sensor networks using Gaussian Markov
random fields,” Automatica, vol. 49, no. 12, pp. 3520–3530, 2013.
[27] J. Pearl, Probabilistic Reasoning in Intelligent Systems, Morgan
Kaufmann, 1988.
Advances in Advances in Journal of Journal of
Operations Research
Hindawi Publishing Corporation
Decision Sciences
Hindawi Publishing Corporation
Applied Mathematics
Hindawi Publishing Corporation
Algebra
Hindawi Publishing Corporation
Probability and Statistics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

The Scientific International Journal of

World Journal
Hindawi Publishing Corporation
Differential Equations
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

http://www.hindawi.com

International Journal of Advances in

Combinatorics
Hindawi Publishing Corporation
Mathematical Physics
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of Journal of Mathematical Problems Abstract and Discrete Dynamics in

Complex Analysis
Hindawi Publishing Corporation
Mathematics
Hindawi Publishing Corporation
in Engineering
Hindawi Publishing Corporation
Applied Analysis
Hindawi Publishing Corporation
Nature and Society
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

International
Journal of Journal of
Mathematics and
Mathematical
Discrete Mathematics
Sciences

Journal of International Journal of Journal of

Hindawi Publishing Corporation Hindawi Publishing Corporation Volume 2014

Function Spaces
Hindawi Publishing Corporation
Stochastic Analysis
Hindawi Publishing Corporation
Optimization
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014