International Journal of Environmental Research and Public Health
Article Improving Aviation Safety through Modeling Accident Risk Assessment of Runway
Yaser Yousefi 1, Nader Karballaeezadeh 2,3 , Dariush Moazami 4, Amirhossein Sanaei Zahed 3,5, Danial Mohammadzadeh S. 3,4,6,7 and Amir Mosavi 8,9,10,*
1 School of Civil Engineering, Iran University of Science and Technology, Tehran P.O. Box 16765-163, Iran; yaser.yousefi@ymail.com 2 Faculty of Civil Engineering, Shahrood University of Technology, Shahrood P.O. Box 3619995161, Iran; [email protected] 3 Department of Elite Relations with Industries, Khorasan Construction Engineering Organization, Mashhad P.O. Box 9185816744, Iran; [email protected] (A.S.Z.); [email protected] (D.M.S.) 4 Department of Civil Engineering, Faculty of Montazeri, Khorasan Razavi Branch, Technical and Vocational University (TVU), Mashhad P.O. Box 9176994594, Iran; [email protected] 5 Toos Institute of Higher Education, Khorasan Razavi, Mashhad P.O. Box 9188911111, Iran 6 Department of Civil Engineering, Ferdowsi University of Mashhad, Mashhad P.O. Box 9177948974, Iran 7 Department of Civil Engineering, Mashhad Branch, Islamic Azad University, Mashhad P.O. Box 9187147578, Iran 8 Faculty of Civil Engineering, Technische Universität Dresden, 01069 Dresden, Germany 9 Institute of Research and Development, Duy Tan University, Da Nang 550000, Vietnam 10 Department of Informatics, J. Selye University, 94501 Komarno, Slovakia * Correspondence: [email protected]
Received: 18 June 2020; Accepted: 19 August 2020; Published: 21 August 2020
Abstract: The exponential increase in aviation activity and air traffic in recent decades has raised several public health issues. One of the critical public health concerns is runway safety and the increasing demand for airports without accidents. In addition to threatening human lives, runway accidents are often associated with severe environmental and pollution consequences. In this study, a three-step approach is used for runway risk assessment considering probability, location, and consequences of accidents through advanced statistical methods. This study proposes novel models for the implementation of these three steps in Iran. Data on runway excursion accidents were collected from several countries with similar air accident rates. The proposed models empower engineers to advance an accurate assessment of the accident probability and safety assessment of airports. For in-service airports, it is possible to assess existing runways to remove obstacles close to runways if necessary. Also, the proposed models can be used for preliminary evaluations of developing existing airports and the construction of new runways.
Keywords: aviation safety; airport safety; runway; accident risk assessment; airport design; runway excursion; accident research; digital health; public health; mobility; transportation; airplane collision; accident injuries prevention; accident modeling; air traffic
1. Introduction Progress of the aviation industry and airport services is considered essential for economic and social development [1]. Despite the numerous social and economic benefits, the expansion of air transport is confronted with several public health challenges [2]. The aviation accident is one of these problems. In addition to threatening human lives, accidents are often associated with severe
Int. J. Environ. Res. Public Health 2020, 17, 6085; doi:10.3390/ijerph17176085 www.mdpi.com/journal/ijerph Int. J. Environ. Res. Public Health 2020, 17, 6085 2 of 37 environmental and pollution consequences [3]. Therefore, the environment and safety have always been of interest to engineers in this industry [1–4]. Process optimization can be effective in solving these problems and their impact on public health [2,5]. Accordingly, there is an increasing demand for the development of environmentally friendly and sustainable airports with minimal environmental impact [6]. On the other hand, the rapid transition of passengers and goods and developments of countries are closely linked to the efficiency of their transportation systems. For example, transportation activities allocate itself more than 9.1% of the Gross Domestic Product (GDP) and involve close to 3 million jobs in Iran. In various human societies, aviation is regarded as one of the key industries in terms of features such as security, speed, and the attractiveness of the airline industry tourism [7]. As the space of take-off and landing aircraft, the runway is one of the most important parts of an airport. Any accident at the runway will result in threaten public health, reducing airport functionality, and resulting in service disruption. Also, accidents can harm the environment in the surrounding of the runway [8,9]. The idea of getting air accident rates down to zero is very naive and unbelievable, but trying to prevent them will increase air transportation safety. Therefore, identifying obstacles, examining the location constraints, and determining the optimum location of the runway are among the essential requirements of each country’s aviation organization. It is expected that there will be perfectly safe conditions when landing and taking-off for each aircraft. Factors such as inclement weather, poor visibility, obstacles, lack of familiarity with the airport, confusion, fatigue, and disruption to Air Traffic Control (ATC) may cause an accident at the airport [10,11]. In general, runway accidents are divided into two main categories [8,12–14]: Runway Incursion (RI) • These accidents occur due to the improper positioning of a plane, vehicle, or human in a protected area for landing and take-off [15,16]. In other words, RI is an accident that occurs when a plane crashes into another plane or vehicle, or human at the runway. The three main causes of RI accidents are non-compliance with ATC instructions, lack of familiarity with the airport, and non-compliance with standard operating procedures [17,18]. Runway Excursion (RE) • RE means leaving the aircraft from the runway. This type of accident is likely to occur during both the take-off and the landing [19]. According to the International Air Transport Association Safety Report, REs caused 22% of all accidents over the 2010–2014 period [20]. According to aviation authorities, about 80 percent of RE accidents occur in landing operations [3]. Several factors contribute to a RE accident, the most important of which are weather, pilot, airport, and aircraft [19,21]. Depending on the exit area of the runway, RE accidents are divided into two groups [22,23]: Veer-Off (VO) and Over-Run (OR). In this paper, two types of RE accidents have been considered by the authors: Landing Overrun (LDOR) and Landing Veer-off (LDVO). Because of the geographical extent of Iran and the underdevelopment of the rail and road transport systems, the appeal of using air transport has been increasing in this country during the last years. Currently, the Iran aviation industry suffers from poor service quality, long service life, lack of advanced navigation system, lack of proper planning, etc. Due to the exhaustion of the Iranian air fleet, the need to use risk assessment models at the airports of this country is very urgent. Therefore, the main purpose of this paper is to develop the final models for risk assessment in Iranian airports. The method of air accident risk assessment in scientific literature consists of three parts [22]: Modeling the probability of accidents, Modeling the location of accidents, and Modeling the consequences of accidents. For this purpose, modeling of probability, location, and consequences of accidents has been performed. Due to the limited number of data, all data were used in the model construction phase, and validation of models was performed with the help of statistical tests: Analysis of Variance (ANOVA) and Residual Diagnostic. For a practical examination of the models, two airports in Iran were evaluated by these models, Mehrabad in Tehran and Hasheminejad in Mashhad. By using these models, engineers can accurately analyze the existing runways and remove obstacles for reducing the number of accidents. Therefore, proposed models can help to increase safety and public health. Also, in new airports, Int. J. Environ. Res. Public Health 2020, 17, 6085 3 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 3 of 36 thescenarios models to help select engineers the optimal to evaluate location proposed of runways. scenarios As a toresult, select they the optimalcan consider location environmental of runways. Asaspects a result, and they choose can considerthe location environmental where the aspects future and probably choose theaccidents location will where have the less future damage probably to accidentsenvironmental will have elements less damage such as to wildlife environmental parks, old elements trees, historical such as wildlife monuments, parks, oldetc. trees,The rest historical of the monuments,paper is organized etc. The as follows: rest of the Section paper 2 ispresents organized the asbackground follows: Section of the 2study. presents Section the background3 introduces ofthe the materials study. Section and methods3 introduces of this the study. materials This and section methods includes of this all study. concepts This sectionrelated includesto accident all conceptsmodeling, related how to to collect accident data modeling, and introduce how the to collectairports data under and study. introduce Then, the we airports show the under results study. and Then,discuss we them show in theSection results 4. Finally, and discuss Section them 5 presents in Section our4. Finally,conclusion. Section 5 presents our conclusion.
2. Literature Review Risk evaluations areare usedused inin many parts of aviation. Little information is available for evaluating thethe riskrisk ofof accidentsaccidents atat airportsairports oror nearnear them.them. PreviousPrevious relevant studies can be classifiedclassified intointo fourfour partsparts [[22]:22]: operationaloperational risk,risk, facilityfacility risk,risk, airport airport design, design, and and third-party third-party risk. risk. The The focus focus of of this this paper paper is onis on third-party third-party risk. risk. The The most most important important and well-knownand well-known examples examples of third-party of third-party risk are risk the Airportare the CooperativeAirport Cooperative Research Research Program Program (ACRP) reports(ACRP) (Reportsreports (Reports 3 [22] and 3 [22] 50 [ 24and]). 50 A review[24]). A of review relevant of studiesrelevant is studies provided is provided below. In below. 2001, In the 2001, Norwegian the Norwegian Civil Aviation Civil Aviation Authority Authority evaluated evaluated the risk the of veer-orisk offf .veer-off. They found They that found the probability that the functionprobability formation function for formation veer-off accidents for veer-off is exponential accidents (see is Equationexponential (1)) (see [25 ]:Equation (1)) [25]: axn P(x) = e (1) P(x) =e− (1) where P is the probability that the aircraft at the end of the veer-off is x meters away from the where P is the probability that the aircraft at the end of the veer-off is x meters away from the runway runway centreline, while a and n are coefficients that depend on the conditions at the examined centreline, while a and n are coefficients that depend on the conditions at the examined contour. The contour. The distance x is calculated from the center of gravity of the aircraft. Moretti et al. calculated distance x is calculated from the center of gravity of the aircraft. Moretti et al. calculated the coefficient the coefficient constant of Equation (1) for two conditions, including take-off and landing (see constant of Equation (1) for two conditions, including take-off and landing (see Equations (2) and (3)) Equations (2) and (3)) [26]: [26]: 0.0219x Plandings = e− (2) . P =e (2) 0.0143x Ptake offs = e− (3) − . P =e (3) In another study, Moretti et al. selected an airport in Italy as a case study and calculated a veer-off risk assessmentIn another forstudy, 1500 Moretti points aroundet al. selected that airport’s an airport runway in Italy [27]. as Their a case modeling study and results calculated are presented a veer- inoff Table risk 1assessment. In this table, for P 1500 is the points frequency around of an that aircraft airport’s running runway beyond [27]. a certain Their distance,modeling x, results measured are frompresented the runway in Table centreline, 1. In this table, according P is the to thefrequency Cartesian of an system aircraft presented running in beyond Figure 1a. certain distance, x, measured from the runway centreline, according to the Cartesian system presented in Figure 1. Table 1. Veer-off modeling. Table 1. Veer-off modeling. Model Model Application Model Model Application 7 0.0219x P = 1.37 10− e− valid for landings on instrument runways P = 1.37 ×× 10 ×e× . valid for landings on instrument runways P = 9.72 10 7 e 0.0219x valid for landings on non-instrument runways P = 9.72 ×× 10 − ×e× . − valid for landings on non-instrument runways P = 6.96 10 8 e . 0.0143x valid for take-offs from instrument runways P = 6.96 ×× 10 − ×e× − valid for take-offs from instrument runways 9 . 0.0143x PP == 6.826.82 × 1010−×ee− validvalid forfor take-otake-offsffs non-instrumentnon-instrument runwaysrunways × ×
Figure 1. The defined defined Cartesian system for veer-off veer-off modeling in Table1 1..
Kirkland et al. proposed a methodology for risk assessment of runway landing overruns. Their proposed methodology included three main models as described in [28]:
Int. J. Environ. Res. Public Health 2020, 17, 6085 4 of 37
Int. J. Environ. Res. Public Health 2020, 17, x 4 of 36 Kirkland et al. proposed a methodology for risk assessment of runway landing overruns. Their• proposedModeling methodology the probability included of a landing three overrun main models as described in [28]: • Modeling the wreckage location Modeling the probability of a landing overrun • • Modeling the damage consequences of overruns Modeling the wreckage location • Table 2 shows the details of this methodology. Modeling the damage consequences of overruns • Table 2. The proposed methodology for risk assessment of runway landing overruns. Table2 shows the details of this methodology. Model Formula Description Table 2. The proposedP(landing methodology overrun) for risk assessment of runway landing overruns. A. Modeling the 1 P is the probability of an overrun, D is either the percent of = probabilityModel of an 1+e ( ) Formulaexcess distance available or percent Description of maximum allowable P(no landing overrun) P is the probability of an overrun, D is either the overrun weight,1 m and n are constants to be determined. A. Modeling the probability of P(landing overrun) = (m+nD) percent of excess distance available or percent of =1−P(landing overrun) 1+e− an overrun P(no landing overrun) = 1 P(landing overrun) maximum allowable weight, m and n are constants to −L is the normalized bewreckage determined. location relative to the end of the normalized requiredL is the land normalizeding distance wreckage expressed location relative as a to the B. Modeling the percentage of the requiredend of the land normalizeding distance, required landingE is the distance excess L=15 + 1.05 E expressed as a percentage of the required landing B.wreckage Modeling thelocation wreckage location L = 15 + 1.05distance E between thedistance, end of E isthe the required excess distance landing between distance the end of and the runway endthe expresse requiredd landing as a percentage distance and the of runway the end required landing distance.expressed as a percentage of the required landing distance. P(sd) is the probabilityP(sd) of is the probability aircraft suffering of the aircraft substantial suffering 1 damage or being destroyed,substantial damageP(nm) oris beingthe probability destroyed, P(nm) of the is the P(sd) = ( ) = 1 ( P) sd (g+hB) probability of the aircraft suffering no damage or e e− aircraft suffering no damage or minor damage, B signifies if P(nm) = 1 P(sd) minor damage, B signifies if the aircraft strikes an C. Modeling the P(nm)=1−P(sd) −the aircraft strikes anobstacle obstacle beyond beyond the runway the endrunway (yes = 1,end no =(yes0), g C. Modeling the damage and h are parameters. consequencesdamage of overruns * = 1, no = 0), g and h are parameters. consequences of P(d) is the probabilityP(d) of is the probability aircraft being of the aircraft destroyed, being destroyed, P(s) is P(s) is the probability of the aircraft being seriously overruns * 1 1 2 P(d) = (i+jBthe+kS) probability of thedamaged, aircraft B2 beingsignifies seriously if the aircraft damaged, struck a second B P(d) = e− 2 ( P()s) = 1 P(d) obstacle beyond the runway end (yes = 1, no = 0), S is e − signifies if the aircraft struck a second obstacle beyond the P(s) =1−P(d) the runway exit speed in meters per second, and i, j, runway end (yes = 1,and no k are= 0), parameters. S is the runway exit speed in * The most reliable predictors of whether the aircraft willmeters suff erper substantial second, and damage i, j, and or be k destroyedare parameters. are whether the aircraft* The has most struck reliable a second predictors obstacle beyondof whether therunway the aircra end,ft andwill thesuffer runway substantial exit speed, damage so the secondor be destroyed stage model is similarare whether but with the two aircraft variables. has struck a second obstacle beyond the runway end, and the runway exit speed, so the second stage model is similar but with two variables. In ACRP Report 3, Federal Aviation Administration (FAA) develops models for risk assessment of the probability,In ACRP location, Report 3, and Federal consequences Aviation Administra of landingtion overruns (FAA) (LDOR), develops landing models undershootsfor risk assessment (LDUS), andof take-o the probability,ff overruns location, (TOOR) and accidents consequences [22]. For of this landing report, overruns the study (LDOR), period landing was from undershoots 1982 to 2006 and(LDUS), in addition and take-off to the US, overruns data from (TOOR) similar accidents countries [22]. (Canada, For this Australia,report, the France,study period Britain, was New from Zealand, 1982 Singapore,to 2006 and Ireland, in addition and Spain) to the were US, data used. from In this simi report,lar countries the three-part (Canada, process Australia, shown France, in Figure Britain,2 is usedNew to modelZealand, the Singapore, risk of accidents. Ireland, and Figure Spain)2 was were adapted used. In from this ACRP report, Report the three-part 3 [22]. process shown in Figure 2 is used to model the risk of accidents. Figure 2 was adapted from ACRP Report 3 [22].
Three-Part Risk Model
Event Probability Location Probability
(Operating conditions) (Operating conditions, terrain)
RISK Accident Consequences (Location probability, Obstacle probability, Size and type, aircraft type)
Figure 2. Risk modeling process in ACRP Report 3. Figure 2. Risk modeling process in ACRP Report 3.
Int. J. Environ. Res. Public Health 2020, 17, 6085 5 of 37
In this report, logistic regression and exponential function were used to model even probability and location probability, respectively. For the third part, the probability of severe consequences modeling was performed using frequency and location models [22]. Using the results of the ACRP Report 3, Wong et al. conducted a risk assessment for the Runway Safety Area (RSA) and focused on accident probability modeling. They investigated LDOR, LDUS, TOOR, and Crash after Takeoff (TOC) accidents at US airports. The study period was from 1982 to 2002 and a total of 440 accidents were selected, including 199 LDOR cases, 122 LDUS cases, 52 TOOR cases, and 67 TOC cases [29]. In 2011, the FAA published ACRP Report 50. This report presented an improved model of the probability of accidents for LDOR, LDVO, LDUS, TOOR, and take-off veer-offs (TOVO). The study used 1031 accidents and 383 incidents at airports of US and similar countries. Normal operation data (NOD) from US airports, operations that did not cause any accidents or incidents, were also used in this study [24]. In addition to adding the LDVO and TOVO accidents, the variables and their levels have been modified in ACRP Report 50. Based on the ACRP Reports 3 and 50, Ayres et al. (2013) have focused on two issues [30]:
Presenting an airport risk assessment model to assess air traffic-related hazards at the airport • Managing the RSA as a risk reduction measure. • In their study, five types of runway accidents including LDOR, LDVO, LDUS, TOOR, and TOVO were analyzed and modeled. In terms of the status of accident distribution in the end and edges of the runway, they used the exponential form to model the accident location [30]. The modeling results performed by Ayres et al. are presented in Table3.
Table 3. Coefficients of accident location model.
Type of Accident Type of Data Model R2 0.00321x0.98494 X P d > x = e− 0.998 LDOR { } 0.20983x0.48620 Y P d > y = e− 0.939 0.00148x0.75150 X P d > x = e− 0.987 LDUS { } 0.02159x0.77390 Y P d > y = e− 0.986 0.02568x0.80395 LDVO Y P d > y = e− 0.915 0.00109x1.06764 X P d > x = e− 0.992 TOOR { } 0.04282x0.65957 Y P d > y = e− 0.987 0.01639x0.86346 TOVO Y P d > y = e− 0.942
Jeon et al. investigated the impact of rainfall on runway accidents. Due to not considering the effect of rainfall (rain and snow) on the RSA risk assessment in the existing studies, they attempted to develop a new location model to account for this effect [31]. The results of their research for the types of accidents are presented in Table4.
Table 4. Advanced location model.
Type of Accident Type of Data Model R2 0.00321x0.984932 X P d > x = e− 0.998 LDOR { } 0.20983x0.48621 Y P d > y = e− 0.942 0.01481x0.749897 X P d > x = e− 0.986 LDUS { } 0.02159x0.78789 Y P d > y = e− 0.989 0.2568x0.803984 LDVO Y P d > y = e− 0.994 0.00109x1.06754 X P d > x = e− 0.989 TOOR { } 0.04282x0.661398 Y P d > y = e− 0.991 0.01639x0.874332 TOVO Y P d > y = e− 0.943 Int.Int. J. J. Environ. Environ. Res. Res. PublicPublic Health 20202020,, 1717, ,x 6085 6 of 36 6 of 37
The runway risk assessment studies have always faced a major challenge called the variety of conditionsThe runway affecting risk risk assessment analysis. Various studies havevariables, always including faced aair major traffic, challenge aircraft type, called and the quality, variety of conditionsbuild quality affecting of Infrastructures, risk analysis. budget, Various maintenance variables, includingquality, weather air tra fficonditions,c, aircraft etc., type, vary and from quality, buildregion quality to region. of Infrastructures, Because of these budget, changes, maintenance engineers quality, are forced weather to model conditions, for each etc., region vary with from the region tospecific region. conditions Because ofof thesethat region. changes, Due engineers to the lack are of forceda comprehensive to model for and each practical region risk with assessment the specific conditionssystem for of runways that region. at Iranian Due toairports, the lack the of authors a comprehensive were encouraged and practical to address risk this assessment gap. Therefore, system for runwaysthe main at purpose Iranian of airports, the present the study authors is to were evaluate encouraged the RSA torisk address concerning this RE gap. accidents Therefore, at Iranian the main purposeairports. of In the terms present of health study and is to environmental evaluate the RSA problems risk concerning due to air RE accidents, accidents risk at Iraniananalysis airports.and Indetermination terms of health of andoptimum environmental location for problems constructing due to or air developing accidents, runways risk analysis can andhelp determinationreduce these of optimumproblems. location One of forthe constructing most important or developing and valid methods runways of can runway help reduce risk assessment these problems. is the method One of the mostpresented important in the and United valid States, methods ACRP of runwayReport 50. risk After assessment reviewing is the collected method presenteddata and comparing in the United States,their dispersion ACRP Report pattern 50. to After the American reviewing method, the collected the authors data and decided comparing to base theirtheir dispersionmodeling process pattern to theon Americanthis method. method, the authors decided to base their modeling process on this method.
3.3. Materials Materials andand MethodsMethods InIn this this study, study, riskrisk assessmentassessment modeling for for LD LDOROR and and LDVO LDVO accidents accidents is performed is performed in three in three parts:parts: Modeling Modeling thethe probabilityprobability of of accidents, accidents, Modeling Modeling the the location location of ofaccidents, accidents, and and Modeling Modeling the the consequencesconsequences of of accidents. accidents. AA comprehensivecomprehensive flowchart flowchart representing representing the the framework framework of of this this research research is presentedis presented in Figure in Figure3. In the3. In continuation the continuation of this of section, this section, the concepts the concepts related torelated runway to riskrunway assessment, risk assessment, data collection process, and specifications of the airports under study are presented in data collection process, and specifications of the airports under study are presented in detail. detail.
Comparison of air accident rates in different Selecting target Zoning of world zones and determining zones with rates countries to study and (based on IATA) similar to the MENA (Including Iran) investigate accidents
Determination of target countries
Basic Information Model of Probability
Aircraft and Airport Information Model of Location
Model of Weather Information Consequences Air accident data preparation Modeling the accidents
Mehrabad International Airport ANOVA
Residual diagnostic Hasheminejad International Airport
Implementing models Validation
Figure 3. Research flow chart. Figure 3. Research flow chart.
Int. J. Environ. Res. Public Health 2020, 17, 6085 7 of 37
Int. J. Environ. Res. Public Health 2020, 17, x 7 of 36 3.1. Risk Assessment of Runway Incidents 3.1. Risk Assessment of Runway Incidents Runway Safety Area (RSA), Figure4, is the sum of the area adjacent to the runway and the end area ofRunway the runway, Safety which Area is (RSA), built to Figure reduce 4, airis the accident sum of damage the area [32 adjacent–36]. The to the RSA runway standard and dimensions the end havearea changed of the overrunway, the yearswhich and is built depend to onreduce the classair accident of aircraft damage using the[32–36]. runway The [ 24RSA]. The standard standard dimensionsdimensions have have increased changed historicallyover the years to and serve depend faster on and the larger class aircraftof aircraft and using to improved the runway the [24]. safety ofThe aviation standard users. dimensions The dimensions have increased of RSA varyhistorical fromly 120to serve feet wide faster by and 240 larger feet beyond aircraftthe and end to of theimproved runway tothe 500 safety feet of wide aviation by 1000 users. feet The beyond dimensions the end of RSA of the vary runway from [12037]. feet In thewide 1960s, by 240 the feet FAA beyond the end of the runway to 500 feet wide by 1000 feet beyond the end of the runway [37]. In the proposed an area that has 500 feet wide and extends 1000 feet beyond each end of the runway [24]. 1960s, the FAA proposed an area that has 500 feet wide and extends 1000 feet beyond each end of the These dimensions are used as the standard in most runways. Based on Figure4, areas (1) and (2) are runway [24]. These dimensions are used as the standard in most runways. Based on Figure 4, areas provided to reduce OR and VO accident consequences, respectively. (1) and (2) are provided to reduce OR and VO accident consequences, respectively.
Figure 4. Runway Safety Area (RSA). Figure 4. Runway Safety Area (RSA).
3.1.1.3.1.1. Accident Accident Probability Probability Modeling Modeling LogisticLogistic regression regression is is used used to to estimateestimate the probability of of accident accidentss based based on on various various factors, factors, such such asas environmental environmental factors. factors. Logistic Logistic regressionregression is similar to to linear linear regression, regression, except except that that in inthis this type type of of prediction,prediction, the the result result for for the the level level or or levelslevels ofof thethe dependent variable variable is is tw twoo solutions solutions [22]. [22 The]. The basic basic modelmodel used used to to create create the the accident accident probability probability modelmodel is as follows [22,24,31,38]: [22,24,31,38]: 1 P Accident_Occurence = (4) ( 1 ⋯ ) P Accident_Occurence = 1+e (4) { } (b0+b1X1+b2X2+b3X3+...) where P Accident_Occurence is the probability (0–100%)1 + e− of an accident type occurring given certain whereoperationalP Accident_Occurence conditions, Xi is independentis the probability variables (0–100%) (Table of 5) an including accident aircraft type occurring type, precipitation, given certain crosswind,{ visibility, ceiling,} etc., and bi is regression coefficients. The values of bi coefficients are operational conditions, Xi is independent variables (Table5) including aircraft type, precipitation, given in Table 6. After obtaining the model coefficients (bi), by placing zero (base level) or one (other crosswind, visibility, ceiling, etc., and bi is regression coefficients. The values of bi coefficients are given levels) numbers as an alternative to each of the independent variables (Xi) in the model, it is possible to in Table6. After obtaining the model coe fficients (b ), by placing zero (base level) or one (other levels) obtain the probability of different accidents. Tables 5i and 6 were adapted from ACRP Report 50 [24]. numbers as an alternative to each of the independent variables (Xi) in the model, it is possible to obtain the probability of different accidents. Tables5 and6 were adapted from ACRP Report 50 [24].
Table 5. Independent variables (Xi) in the accident probability model.
Independent Variables (Xi) Levels of Variable Commercial (C) (Base level *) Cargo (F) User Class General Aviation (GA) Taxi/Commuter (T/C)
Int. J. Environ. Res. Public Health 2020, 17, 6085 8 of 37
Table 5. Cont.
Independent Variables (Xi) Levels of Variable Large Jet (B737, A320, etc.) (C) (Base level) Heavy Jet (B777, A340, etc.) (AB) Equipment Class based on Maximum Takeoff Weight (MTOW) Large Commuter (Regional Jets, ERJ-190, CRJ-900, ATR42, etc.) (D) Medium Aircraft (Biz Jets, Embraer120, Learjet35, etc.) (E) Small Aircraft (Beech-90, Cessna Caravan, etc.) (F) Jet (Base level) Engine Type Turboprop Domestic Foreign Origin/Destination (Foreign O/D) Foreign More than 2500 ft (Base level) Less than 200 ft Ceiling Height Between 200 ft to 1000 ft Between 1000 ft to 2500 ft Visibility is more than 8 miles (Base level) Visibility is less than 2 miles Visibility Visibility is between 2 miles and 4 miles Visibility is between 4 miles and 8 miles Less than 2 knots (Base level) Between 2 knit and 5 knots Crosswind (Xwind) Between 5 knit and 12 knots More than 12 knots Less than 5 knots (Base level) Tailwind Between 5 knit and 12 knots More than 12 knots
Between 15 ◦C to 25 ◦C (Base level)
Less than 5 ◦C Air Temperature Between 5 ◦C to 15 ◦C
More than 25 ◦C Non-existence (Base level) Gust Existence Non-existence (Base level) Thunderstorm Existence Non-existence (Base level) Rain Existence Non-existence (Base level) Snow Existence Non-existence (Base level) Fog Existence Non-existence (Base level) Icing Condition Existence Non-existence (Base level) Frozen Precipitation Existence Hub Airport (Base level) Hub/Non-Hub Airport Non-Hub Airport CF < 0 (Base level) Log Criticality Factor CF > 0 It is not night (Base level) Night Condition It is night * The base levels in this table do not affect the model and are part of the ideal conditions. Int. J. Environ. Res. Public Health 2020, 17, x 9 of 36
Int. J. Environ. Res. PublicTable Health 6.2020 Values, 17, 6085of bi coefficients for the accident probability model. 9 of 37
Variable LDOR LDVO LDUS TOOR TOVO Adjusted Constant −13.065 −13.088 −15.378 −14.293 −15.612 Table 6. Values of bi coefficients for the accident probability model. User Class F - - 1.693 1.266 - VariableUser Class G LDOR1.539 LDVO1.682 1.288 LDUS - TOOR2.094 TOVO Adjusted ConstantUser Class T/C 13.065 −0.498 13.088- 0.01715.378 - 14.293 - 15.612 − − − − − User ClassAircraft F Class A/B -−1.013 -−0.770 −0.778 1.693 −1.150 1.266 −0.852 - User ClassAircraft G Class D/E/F 1.5390.935 1.682−0.252 0.138 1.288 −2.108 -−0.091 2.094 User Class T/C 0.498 - 0.017 - - Ceiling less than 200 ft− −0.019 - 0.070 0.792 - Aircraft Class A/B 1.013 0.770 0.778 1.150 0.852 Ceiling 200 to 1000 ft − −0.772 − - −−1.144 −0.114− - − Aircraft Class D/E/F 0.935 0.252 0.138 2.108 0.091 Ceiling 1000 to 2500 ft −0.345 − - −0.721 - − - − Ceiling less than 200 ft 0.019 - 0.07 0.792 - − Ceiling 200Visibility to 1000 ftless than 2 SM0.772 2.881 - 2.143 3.0961.144 1.364 0.1142.042 - − − − Ceiling 1000Visibility to 2500 from ft 2 to 4 SM0.345 1.532 - - 1.8240.721 −0.334 -0.808 - − − Visibility lessVisibility than 2 from SM 4 to 8 SM 2.881 0.200 2.143- 0.416 3.096 0.652 1.364−1.500 2.042 Visibility fromXwind 2 to from 4 SM 5 to 12 kt 1.532−0.913 -0.653 −0.295 1.824 −0.695 0.3340.102 0.808 − Visibility from 4 to 8 SM 0.2 - 0.416 0.652 1.500 Xwind from 2 to 5 kt −1.342 −0.091 −0.698 −1.045 - − Xwind from 5 to 12 kt 0.913 0.653 0.295 0.695 0.102 Xwind more than 12 kt− −0.921 2.192 −−1.166 0.219− 0.706 Xwind from 2 to 5 kt 1.342 0.091 0.698 1.045 - Tailwind from 5 to 12− kt - − 0.066 − - - − - Xwind more than 12 kt 0.921 2.192 1.166 0.219 0.706 − − Tailwind fromTailwind 5 to 12 more kt than 12 kt - 0.786 0.0660.980 ------Tailwind moreTemp than less 12than kt 5 C 0.7860.043 0.980.558 0.197 - 0.269 -0.988 - Temp lessTemp than 5 from C 5 to 15 C 0.043−0.019 0.558−0.453 −0.710 0.197 −0.544 0.269 −0.420 0.988 Temp fromTemp 5 to 15more C than 25 C 0.019 −1.067 0.4530.291 −0.4630.710 0.315 0.544−0.921 0.420 − − − − − Temp more than 25 C 1.067 0.291 0.463 0.315 0.921 Icing Condition − 2.007 2.670 2.703− 3.324 - − Icing ConditionRain 2.007- 2.67−0.126 0.991 2.703 0.355 3.324−1.541 - Rain - 0.126 0.991 0.355 1.541 Snow 0.449 − 0.548 −0.250 0.721 0.963 − Snow 0.449 0.548 0.250 0.721 0.963 − Frozen PrecipitationFrozen Precipitation - - 0.103−0.103 ------− GustsGusts - - 0.036−0.036 0.041 0.041 0.006 0.006- - − FogFog -- 1.741.740 ------ThunderstormThunderstorm 1.344−1.344 ------− Turboprop - 2.517 - 0.56 1.522 Turboprop - − −2.517 - 0.560 1.522 Foreign OD 0.929 0.334 1.354 - 0.236 Foreign OD 0.929 − −0.334 1.354 - −0.236 − Hub/Non Hub Airport 1.334 - - - 0.692 −Hub/Non−Hub Airport 1.334 - - - −0.692 − Log Criticality Factor 9.237 4.318 1.629 - 1.707 - Night ConditionLog Criticality Factor - 9.237 1.3604.318 1.629 - -1.707 - Night Condition - − −1.360 - - -
3.1.2. Accident Location Modeling 3.1.2. Accident Location Modeling The likelihood of an accident in all areas around the airport is not equal. An accident near the The likelihood of an accident in all areas around the airport is not equal. An accident near the runway is more likely to occur than the high distances from the runway. Therefore, the probability of runway is more likely to occur than the high distances from the runway. Therefore, the probability accidents depends on the location of the accident. This dependence is expressed as a model of the of accidents depends on the location of the accident. This dependence is expressed as a model of the accident location. The accident location model is based on past accident data [22,24]. According to accident location. The accident location model is based on past accident data [22,24]. According to the distribution of accidents towards the end and edges of the runway, the exponential function is the distribution of accidents towards the end and edges of the runway, the exponential function is used for modeling the accident location. The axis locations for measuring distances are viewed in used for modeling the accident location. The axis locations for measuring distances are viewed in Figures5 and6 . The nose wheel of aircraft is the reference location. These figures were adapted from Figures 5 and 6. The nose wheel of aircraft is the reference location. These figures were adapted from ACRP Report 50 [24]. ACRP Report 50 [24].
Figure 5. X-Y origin for LDOR accidents.
Int. J. Environ. Res. Public Health 2020, 17, x 10 of 36 Int. J. Environ. Res. Public Health 2020, 17, 6085 10 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 10 of 36 Figure 5. X-Y origin for LDOR accidents. Figure 5. X-Y origin for LDOR accidents.
FigureFigure 6. 6. YY origin origin for for LDVO LDVO accidents. accidents. Figure 6. Y origin for LDVO accidents. TheThe basic basic model model for the longitudinal distribution of OR accidents is [22,24]: [22,24]: The basic model for the longitudinal distribution of OR accidents is [22,24]: P Location > x ==e axn (5) P Location > x e− (5) {P Location >} x =e (5) where P Location > x is the probability the OR distance along the runway centerline beyond the where P Location > x is the probability the OR distance along the runway centerline beyond the runwaywhere endP{ Location is greater >} xthan is thex and probability x a given thedistance OR distance beyond alongthe runway the runway end. a centerlineand n are regressionbeyond the runway end is greater than x and x a given distance beyond the runway end. a and n are regression coefficients.runway end Figure is greater 7 showed than x aand typical x a given longitudinal distance location beyond distribution.the runway end. This a figure and n wasare regressionadapted coefficients. Figure7 showed a typical longitudinal location distribution. This figure was adapted fromcoefficients. ACRP Report Figure 50 7 [24].showed a typical longitudinal location distribution. This figure was adapted fromfrom ACRP ACRP Report Report 50 50 [ 24[24].].
Figure 7. OR incidents model. FigureFigure 7. 7.OR OR incidents incidents model. model. Equation (6) is used for transverse location distribution of OR and VO accidents: Equation (6) is used for transverse location distribution of OR and VO accidents: Equation (6) is used for transverse location distribution of OR and VO accidents: P Location > y =e (6) m PPLocation Location> >yy ==ee by (6) where P Location > y is the probability that accident distance− from the runway border (in VO accident)where P or Location centerline > y(in is OR the accident) probability is greater that accidentthan y and distance y is a given from distancethe runway from borderthe extended (in VO where P Location > y is the probability that accident distance from the runway border (in VO accident) runwayaccident) centerline or centerline or runway (in OR border. accident) b and is greater m are regressionthan y and coefficients.y is a given Figuredistance 8 fromshowed the a extended typical or centerline (in OR accident) is greater than y and y is a given distance from the extended runway transverserunway centerline location distribution. or runway border. This figure b and was m areadapted regression from ACRPcoefficients. Report Figure 50 [24]. 8 showed a typical centerline or runway border. b and m are regression coefficients. Figure8 showed a typical transverse transverse location distribution. This figure was adapted from ACRP Report 50 [24]. location distribution. This figure was adapted from ACRP Report 50 [24].
Int. J. Environ. Res. Public Health 2020, 17, 6085 11 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 11 of 36
Figure 8. VOVO incidents incidents model. model. 3.1.3. Accident Consequences Modeling 3.1.3. Accident Consequences Modeling The consequences of an accident are a function of several factors. In modeling the consequences The consequences of an accident are a function of several factors. In modeling the consequences of OR and VO accidents, the following variables are considered as effective factors in the accident of OR and VO accidents, the following variables are considered as effective factors in the accident severity [24]: severity [24]: Type, size, and location of the obstacle • Type, size, and location of the obstacle • Aircraft speed and wingspan • Aircraft speed and wingspan • The number of obstacles and their location distribution • The number of obstacles and their location distribution Based on the maximum speed of the aircraft, whic whichh has the least damage and casualties if if it encounters an obstacle, obstacles are divided into four groups [[24]:24]: Group one: Maximum speed is nill (such as cliff in RSA and concrete wall) Group one: Maximum speed is nill (such as cliff in RSA and concrete wall) (1) Group two: Maximum speed is 5 knots (such as brick buildings) (1) Group two: Maximum speed is 5 knots (such as brick buildings) (2) GroupGroup three: three: Maximum Maximum speed speed is is 20 20 knots knots (such (such as as ditches ditches and and fences) fences) (3) GroupGroup four: MaximumMaximum speed speed is is 40 40 knots knots (such (such as frangible as frangible structure structure and approach and approach lighting lighting system) system) The main idea in making the accident consequence model is to use models of accidents to estimate the accidentsThe main in idea which in themaking aircraft the has accident a lot of energyconsequence when itmodel encounters is to use an obstacle,models thusof accidents producing to estimatesevere consequences. the accidents This in which method the was aircraft developed has a lo int ACRPof energy Report when 3 [ 22it encounters]. For a better an understandingobstacle, thus producingof this approach, severe note consequences. Figure9. This method was developed in ACRP Report 3 [22]. For a better understanding of this approach, note Figure 9.
Int. J. Environ. Res. Public Health 2020, 17, 6085 12 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 12 of 36 Int. J. Environ. Res. Public Health 2020, 17, x 12 of 36
Figure 9. Modeling approach for OR accidents consequences. Figure 9. Modeling approach for OR accidents consequences.
In Figure 99,,D D00 isis the the distance distance to to obstacle obstacle and and d is d the is the distance distance between between the stopped the stopped aircraft aircraft and andrunwayIn runway Figure end. end.Δ 9, value D0∆ is value isthe predicted distance is predicted based to obstacle on based the and onrate thed of is decr ratethe distanceease of decrease in speed between in at speed different the stopped at di typesfferent aircraft of typessurfaces and of runway end. Δ value is predicted based on the rate of decrease in speed at different types of surfaces surfaces(paved, unpaved, (paved, unpaved, and engineered and engineered materials arrestor materials system arrestor (EMAS)), system the (EMAS)), critical thecollision critical speed, collision and (paved, unpaved, and engineered materials arrestor system (EMAS)), the critical collision speed, and speed,the size and and the type size of andobstacle type [30]. of obstacle [30]. the size and type of obstacle [30]. Based on Figure9 9,, thisthis modelingmodeling hashas three three approaches approaches [ 24[24]:]: Based on Figure 9, this modeling has three approaches [24]: (1) TheThe aircraft aircraft does does not not collide with the obstacle ( (dd
Figure 10. Modeling consequences. Figure 10. ModelingModeling consequences. consequences.
Int. J. Environ. Res. Public Health 2020, 17, 6085 13 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 13 of 36 Int. J. Environ. Res. Public Health 2020, 17, x 13 of 36
FigureFigure 11.11. The impact of obstacle lateral location inin accidentaccident consequences.consequences. Figure 11. The impact of obstacle lateral location in accident consequences. BasedBased onon FiguresFigures 1010 andand 1111,, threethree generalgeneral conditionsconditions cancan bebe expectedexpected [[24,30]:24,30]: Based on Figures 10 and 11, three general conditions can be expected [24,30]: (1)(1) Medium consequences when part of thethe obstacle is in the yellow area (1)(2)(2) MediumSevere consequences consequences whenwhen partpart ofof theththee obstacle obstacle is is in in the the orange yellow areaarea
(2)(3)(3) SevereNo consequences consequences when when the part obstacle of the is obstacle out of the is in orange the orange and yellow area areas (3) No consequences when the obstacle is out of the orange and yellow areas As can be seen in Figure 12, the W1 × L1 obstacle is located at a distance x and y of the runway As can be seen in Figure 12, the W1 L1 obstacle is located at a distance x and y of the runway 1 1 threshold.As can Itbe can seen be inassumed Figure 12,that the if the W aircraft×× L obstacle encounters is located an obstacle at a distance in the distancex and y ofbetween the runway yc and threshold. It can be assumed that if the aircraft encounters an obstacle in the distance between yc and c threshold.yf, it will haveIt can severe be assumed consequences that if the [24,30]. aircraft encounters an obstacle in the distance between y and yf, it will have severe consequences [24,30]. yf, it will have severe consequences [24,30].
Figure 12. Modeling the possibility which an aircraft collides with an obstacle with severe Figure 12. Modeling the possibility which an aircraft collides with an obstacle with severe consequences. Figureconsequences. 12. Modeling the possibility which an aircraft collides with an obstacle with severe consequences. Based on the transverse distribution model of the accident location, the probability that the aircraftBased will on be the at thetransverse mentioned distribution distance whenmodel exiting of the from accident the runway location, is the[24]: probability that the aircraft will be at the mentioned distance when exiting from the runway is [24]: e −e (7) P =e −e P = 2 (7) 2
Int. J. Environ. Res. Public Health 2020, 17, 6085 14 of 37
Based on the transverse distribution model of the accident location, the probability that the aircraft will be at the mentioned distance when exiting from the runway is [24]:
bym bym e c e− f P = − − (7) sc 2 where Psc is the probability of high consequences, b and m are regression coefficients for transverse location model, yc and yf are critical location closest to the extended runway axis and farther from the extended runway axis, respectively. By combining Equation (7), the longitudinal distribution model of accidents, and the possibility of multiple obstacles, the risk of accidents with severe consequences is determined by Equation (8) [24]:
N bym m X ci byfi (e− e− ) a(x +∆ )n Psc = − e− i i (8) 2 × i=1 where N is the number of existing obstacles, a and n are regression coefficients of longitudinal distribution model of accidents, and ∆i is the location parameter for obstacle i.
3.2. Data Gathering In this study, accident modeling is performed with the help of accident statistics in similar countries. According to the ACRP Report 3, the selection of similar countries has been done with the help of the standard of similarity of accident rates. Accordingly, the countries of the world are divided into eight separate regions [39]: Africa (AFI), Middle East and North Africa (MENA), Asia/Pacific (ASPAC), North Asia(NASIA), Commonwealth of Independent States (CIS), Europe (EUR), Latin America and the Caribbean (LATAM), and North America (NAM). Also, Table7 shows the air accident rates per one million operations performed during 2013 and 2014, as well as the average for 2009–2013. Table7 was extracted from the safety report 2014 [39].
Table 7. Air accident rates based on IATA zoning (per one million operations).
Zone Air accident Rates in 2013 Air Accident Rates in 2014 Average Air Accident Rates (2009–2013) AFI 11.18 7.12 12.45 ASPAC 2.57 2.90 2.76 CIS 2.19 3.14 5.92 EUR 1.35 2.75 2.03 LATAM 2.73 1.98 3.36 MENA 3.47 3.05 5.43 NAM 1.00 1.55 1.38 NASIA 0.95 0.53 0.82
Based on Table7 and due to the nearness of air accident rates, MENA (including Iran), CIS, LATAM, and ASPAC zones have been selected as similar countries for RE accident data collection. A total of 98 countries have been selected in these four zones, of which 19 are in MENA, 12 in CIS, 33 in LATAM, and 34 in ASPAC. All RE accidents in the mentioned countries have been carefully examined and finally, the data of 57 countries have been selected. Table8 shows the list of selected countries and the number of data extracted from each country. This table contains the data of LDOR, LDVO, TOOR, and TOVO accidents between 1990 and 2015. Int. J. Environ. Res. Public Health 2020, 17, 6085 15 of 37
Table 8. List of selected countries and the number of data usable from each country.
MENA Number of Data CIS Number of Data LATAM Number of Data ASPAC Number of Data Iran 13 Russia 65 Brazil 21 Indonesia 47 Afghanistan 4 Ukraine 8 Colombia 11 India 13 Sudan 3 Kazakhstan 5 Argentina 8 Japan 8 Morocco 2 Tajikistan 2 Bolivia 6 Malaysia 7 Tunisia 2 Uzbekistan 2 Ecuador 6 Philippines 6 UAE. 2 Armenia 1 Mexico 6 Australia 5 Yemen 2 Georgia 1 Venezuela 3 Bangladesh 4 Algeria 1 Kyrgyzstan 1 Guatemala 2 Nepal 4 Int. J. Environ.Iraq Res. Public 1 Health 2020, 17, x Honduras 2 Pakistan 4 15 of 36 Saudi 1 Nicaragua 2 Singapore 4 SaudiArabia 1 Nicaragua 2 Singapore 4 Arabia New Peru 2 2 Guinea New The Peru 2 South 2 1 Guinea 2 Bahamas Korea The Bahamas 1 South Korea 2 Belize 1 Thailand 2 Belize 1 Thailand 2 Chile 1 Vietnam 2 Chile 1 Vietnam 2 CubaCuba 11 BruneiBrunei 1 1 Dominican Dominican 1 Laos 1 Republic 1 Laos 1 Republic Guyana 1 Maldives 1 Guyana 1 Maldives 1 New Haiti 1 New 1 Haiti 1 Zealand 1 Jamaica 1 Sri LankaZealand 1 Jamaica 1 Sri Lanka 1 Panama 1 Panama 1 SUM 31 SUM 85 SUM 78 SUM 115 SUM 31 SUM 85 SUM 78 SUM 115
OnceOnce the the countries countries to to collect collect datadata havehave beenbeen identified,identified, three three initial initial forms forms have have been been prepared prepared includingincluding Basic Basic Information, Information, Aircraft Aircraft andand AirportAirport Information, and and Weather Weather Information. Information. Figure Figure 13 13 showsshows these these forms. forms.
Basic Information Aircraft & Airport Information
NO. 5 Type Fokker 100
Day 26 Jet Engine Date Month Aug Turboprop Year 2010 Commercial - C Origin Tehran Cargo - F User Class Destination Tabriz Taxi/Commuter - T/C General Aviation - G Event Location Tabriz International Airport C - Large Jet - MTOW=41k-255k lb Country Iran AB - Haevy Jet - MTOW=255k lb+ Region MENA Equipment Class D - Large Commuter - MTOW=41k-255k lb E - Medium Aircraft - MTOW=12.5k-41k lb Airline Iran Aseman Airlines F - Small Aircraft - MTOW=12.5k lb-
Incident Domestic Category Foreign OD Accident International
LDOR Hub Airport LDVO Non Hub Type LDUS TOOR Runway Distance Required (ft) 10393 TOVO Runway Distance Available (ft) 11825
X (m) 500 ICAO OITT Distance from Runway Airport Code Y (m) 0 IATA TBZ
Figure 13. Cont.
Int. J. Environ. Res. Public Health 2020, 17, x 16 of 36
Int. J. Environ. Res. Public Health 2020 17 , , 6085Weather Information 16 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 16 of 36 -200 ft
[200-1000) ft Weather Information Ceilling [1000-2500] ft +2500-200 ft ft Yes Rain [200-1000) ft Ceilling No [1000-2500] ft -2 SM +2500 ft Yes [2-4) SMRain Yes Visibility Snow No [4-8] SM No -2 SM +8 SM [2-4) SM Yes Visibility Snow [4-8] SM No -2 +8kn SM [2-5) kn Crosswind [5-12]-2 knkn Yes Gusts [2-5) kn Crosswind +12 kn No [5-12] kn Yes Gusts +12 kn No -5 kn Yes Thunderstorms Tailwind [5-12] kn No -5 kn Yes +12 kn Thunderstorms Tailwind [5-12] kn No +12 kn Yes Fog -5 CYes No Fog [5-15)-5 C No Air Temperature [15-25][5-15) C C Air Temperature +25[15-25] C C +25 C
Figure 13. Initial forms for collecting and recording accidents, aircraft, airport, and weather conditions FigureFigure 13. 13.Initial Initial forms forms for for collecting collecting and andrecording recording accident accidents,s, aircraft, airport, aircraft, and airport,weather conditions and weather information. conditionsinformation. information.
3.3. Airports 3.3. Airports Under Under Study Study A studystudyA study ofof thethe of historythehistory history ofof airofair air accidentsaccidents accidents onon on Iranian IranianIranian airportsairports shows showsshows that thatthat Mehrabad MehrabadMehrabad International InternationalInternational AirportAirportAirport inin Tehran Tehran in Tehran and and Hasheminejad and Hasheminejad Hasheminejad International Internationa Internationa Airportll Airport in Mashhad inin Mashhad Mashhad have thehave have largest the the largest share. largest share. Therefore, share. theTherefore, accidents,Therefore, the the accidents, the condition accidents, the of the condition the condition runways, of of the the and runw runw theays,ays, obstacles and thethe around obstacles obstacles the around runways around the the ofrunways these runways two of these airportsof these havetwo airports beentwo airports carefully have have been examined. been carefully carefully examined. examined. 3.3.1. Mehrabad International Airport 3.3.1. Mehrabad International Airport Mehrabad International Airport was built in 1938 in the western part of Tehran. A study of the Mehrabadpresented statistics International shows Airportthat Airport in recent was was years,built built inTehran in 1938 1938 hasin in the been the western co westernnsidered part part as of the ofTehran. main, Tehran. largest, A study A studyand of the of thepresented presentedmost importantstatistics statistics showscity showsin Iran,that thatinand recent insome recent years,cases years, as Tehran the Tehran center has ofbeen has national been considered consideredand international as the as main, the exhibitions. main, largest, largest, and andmost most Theimportant special important conditionscity cityin Iran, in of Iran, Tehran and and some have some casesa direct cases as impact asthe the center on center the ofperformance of national national and and and airinternational international traffic of the exhibitions.airport. exhibitions. TheThe specialMehrabad conditions Airport ofis currently Tehran havehave the first a direct airport impact in the oncountry the performance to receive about and 12 air million traffic traffi passengers.c of the airport. airport. MehrabadThis airport Airport has is two currently parallel therunways firstfirst airport(see Figure in the 14). country to receive about 12 million passengers. ThisThis airportairport hashas twotwo parallelparallel runwaysrunways (see(see FigureFigure 1414).).
Figure 14. Mehrabad International Airport-Runways plan. Int. J. Environ. Res. Public Health 2020, 17, x 17 of 36 Int. J. Environ. Res. Public Health 2020, 17, 6085 17 of 37 Figure 14. Mehrabad International Airport - Runways plan.
TableTable9 9was was extracted extracted from from the the International International Civil Civil Aviation Aviation Organization Organization (ICAO) (ICAO) [ 37[37]] and and shows shows thethe physical physical characteristics characteristics of of runways runways in in Mehrabad Mehrabad International International Airport. Airport.
TableTable 9. 9.Physical Physical characteristics characteristics of of Mehrabad Mehrabad International International Airport Airport runways. runways.
RunwayRunway Dimensions Dimensions (m) (m) Pavement Pavement Type Type StopwayStopway Strip Strip RESA RESA 11L11L 3989 45 3989 × 45 Asphalt Asphalt 122 122 ×45 45 - - 226 × 150 226 150 × × × 29R29R 3989 45 3989 × 45 Asphalt Asphalt 194 194 ×45 45 - - - - × × 11R11R 4030 60 4030 × 60 Asphalt Asphalt 87 87 ×45 45 - - - - × × 29L29L 4030 60 4030 × 60 Asphalt Asphalt ------× Note:Note: RESA RESA isis RunwayRunway End End Safety Safety Area. Area.
3.3.2.3.3.2. HasheminejadHasheminejad InternationalInternational AirportAirport HasheminejadHasheminejad InternationalInternational Airport,Airport, inin Mashhad,Mashhad, waswas builtbuilt inin 1951.1951. ThisThis airportairport isis thethe second second busiestbusiest airportairport inin thethe countrycountry afterafter Mehrabad Mehrabad International International Airport Airport in in Tehran. Tehran. HasheminejadHasheminejad InternationalInternational AirportAirport hashas twotwo parallelparallel runwaysrunways (See(See FigureFigure 15 15).). Also,Also, TableTable 10 10 shows shows thethe physical physical characteristicscharacteristics of of the the runways. runways. ThisThis tabletable was was extracted extracted from from ICAO ICAO [ 37[37].].
FigureFigure 15. 15.Hasheminejad Hasheminejad InternationalInternational Airport-Runways Airport-Runways plan. plan.
Table 10. Physical characteristics of Hasheminejad International Airport runways. Table 10. Physical characteristics of Hasheminejad International Airport runways.
RunwayRunway Dimensions Dimensions (m) (m) PavementPavement Type type Stopway Stopway Strip Strip RESA RESA 13L13L 3810 381045 × 45 Asphalt Asphalt 302 302 × 4545 - - - - × × 31R31R 3810 381045 × 45 Asphalt Asphalt 303 303 × 4545 - - - - × × 13R 3920 45 Asphalt 300 45 - - 13R × 3920 × 45 Asphalt 300 × ×45 - - 31L 3920 45 Asphalt 296 45 - - 31L × 3920 × 45 Asphalt 296 × ×45 - - Note:Note: RESA RESA isis RunwayRunway End End Safety Safety Area. Area.
4.4. ResultsResults andand DiscussionDiscussion TheThe collected collected data data are are from from 1990 1990 to to 2015.2015. Undoubtedly,Undoubtedly, the the number number of of RE RE accidentsaccidents and andincidents incidents thatthat occur occur during during the the period period under under review review is much is highermuch higher than the than number the ofnumber cases presented. of cases presented. However, theHowever, lack of registration the lack of registration of information of information about some ofabout the accidents some of the and accidents incidents and has incidents made it impossible has made toit useimpossible all of them to use in all this of research. them in this A total research. of 309 A events,total of including309 events, 168 including accidents 168 and accidents 141 incidents, and 141 haveincidents, been collectedhave been and collected classified. and Figureclassified. 16 shows Figure how 16 shows these eventshow these are distributed.events are distributed.
Int. J. Environ. Res. Public Health 2020, 17, 6085 18 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 18 of 36
FigureFigure 16. 16. TheThe events events distribution distribution (based (based on on the the type type of of RE RE accident). accident).
AsAs can can be be seen, seen, the the frequency frequency of of TOOR TOOR and and TOVO TOVO data data is is low. low. Therefore, Therefore, it it is is not not possible possible to to buildbuild a a model model based based on on this this type type of of accident. accident. As As a a result, result, further further research research is is based based on on LDOR LDOR and and LDVOLDVO accidents. accidents. Each Each event consistsconsists ofof18 18 variables variables (16 (16 variables variables related related to theto the probability probability model model and and2 variables 2 variables related related to the to location the location model). model). Failure Fail toure record to record some some variables variables has led has to missingled to missing data in datasome in collected some collected cases. Finally, cases. Finally, 128 and 128 143 and data 143 were data used were to construct used to construct the longitudinal the longitudinal and transverse and transverseLDOR accident LDOR model, accident respectively. model, respectively. Also, 49 data Also were, 49 used data to were construct used the to LDVOconstruct accident the LDVO model. accidentSince the model. probability Since modelthe probability and the location model and model the are location made model independently are made of independently each other, each of ofeach the other,selected each events of the is atselected least appropriate events is forat oneleast of appropriate the models. for Because one of of the limitationsmodels. Because in the numberof the limitationsof data, all in of the them number were used of data, in the all construction of them were of used the model in the and construction the validation of the was model done and only the by validationstatistical tests.was done only by statistical tests. 4.1. Accident Probability Model 4.1. Accident Probability Model Due to the lack of a suitable database for NOD flights in Iran, it is not possible to build a model for Due to the lack of a suitable database for NOD flights in Iran, it is not possible to build a model the probability of RE accidents. Therefore, the studied airports will be analyzed based on the logistics for the probability of RE accidents. Therefore, the studied airports will be analyzed based on the regression model presented in the ACRP Report 50. Details of this model are provided in Section 3.1.1. logistics regression model presented in the ACRP Report 50. Details of this model are provided in Section4.2. Accident 3.1.1. Location Model
4.2.4.2.1. Accident Longitudinal Location Location Model Model for LDOR
4.2.1. InLongitudinal this section, Location the first locationModel for model LDOR for RE accidents is presented. The goal is to find FX(x) or the probability of x (a, b) for any desired value of a and b. According to the statistical preliminary concepts: In this section,∈ the first location model for RE accidents is presented. The goal is to find FX(x) or the probability of x ∈ (a, b) for anyF desired(x) = Pvalue(X ofx) a= and1 b.P( AccordingX x) to the statistical preliminary(9) X ≤ − ≥ concepts: where FX(x) is the probability function x and P(X x) is probabilities of occurrence of X less than x. F (x) =P(X≤x)≤=1−P(X≥x) (9) Figure 17 shows the location distribution of the studied LDOR accidents and incidents (indirect X) and whereFigure F 18X(x) shows is the the probability x curve versus function P( Xx andx) .P(X≤x) is probabilities of occurrence of X less than x. ≥ Figure 17 shows the location distribution of the studied LDOR accidents and incidents (indirect X) and Figure 18 shows the x curve versus P(X ≥ x).
Int.Int. J. J. Environ. Environ. Res. Res. Public Public Health Health 20202020, ,1717, ,x 6085 19 19 of of 36 37 Int. J. Environ. Res. Public Health 2020, 17, x 19 of 36 35 35 32 32 30 30 30 30 25 25 20 20 16 16 15 13 15 13
99 10 10 88 66 4 4 5 5 33 22 2222 1 0 0 0 - 500 - 50 50 - 50 100 - 100 100 100 - 150 - 150 150 150 - 200 - 200 200 200 - 250- 250 250 250 - -300 300 300 300 -- 350350 350 -- 4040 400 400 - - 450 450 450 450 - -500 500 500 500 - 550- 550 550 550 - 600 - 600 More More than than 600600 NumberNumber ofof occurrencesoccurrences
Figure 17. LDOR distribution in the X direction (Horizontal axis shows the distance (meter) from FigureFigure 17. 17. LDORLDOR distribution distribution in in the the X X direction direction (Horiz (Horizontalontal axis axis shows shows the the distance distance (meter) (meter) from from runway end). runwayrunway end). end).
Figure 18. x curve versus P(X ≥ x) for LDOR. FigureFigure 18. 18. xx curve curve versus versus P(X P(X ≥ x)x) forfor LDOR. Given that the relationship between the two variab≥les is exponential, it is predicted that the P(XGiven ≥ x) thatfunction the relationship can be estimated between using thetwo exponential variables family is exponential, regression. it is Based predicted on Figure that the 18,P( Xthe x) Given that the relationship between the two variab les is exponential, it is predicted that ≥the functionregression can becurve estimated is considered using exponential as P(X≥x family) =e regression.+ε , Basedin which on Figure b0 and 18 , theb1 are regression unknown curve P(X ≥ x) function can be estimatedb using exponential family regression. Based on Figure 18, the b0x 1 is consideredparameters. as TablesP(X 11x )and= e12− show+ theεi, SPSS in which software b0 and output. b1 are unknown parameters. Tables 11 and 12 regression curve is ≥considered as P(X≥x) =e +ε , in which b0 and b1 are unknown show the SPSS software output. parameters. TablesTable 1111. and Estimation 12 show of unknown the SPSS parameters software ofoutput. the LDOR model in the X-direction. Table 11. Estimation of unknown parameters of the LDOR model in the X-direction. 95% Confidence Interval UnknownTable 11. EstimationEstimated of unknown Standardparameters Error of theof the LDOR Mean model in the X-direction. Lower Upper UnknownParameter Value Standard Error(SE) of 95% Confidence Interval Estimated Value 95%Bound Confidence Bound Interval UnknownParameter Estimated Standardthe Mean Error (SE) of the MeanLower Bound Upper Bound b0 0.012 0.001 0.011Lower 0.014Upper Parameter Value (SE) bb01 0.0120.804 0.001 0.011 0.011 0.781Bound 0.0140.827Bound b 0.804 0.011 0.781 0.827 b0 1 0.012 0.001 0.011 0.014 b1 0.804 0.011 0.781 0.827
Int. J. Environ. Res. Public Health 2020, 17, 6085 20 of 37
Table 12. Analysis of Variance (ANOVA) for the LDOR model in the X-direction. Int. J. Environ. Res. Public Health 2020, 17, x 20 of 36 Source of Variations Sum of Squares Degree of Freedom (df) Mean Squares RegressionTable 12. Analysis of Variance 44.340 (ANOVA) for the LDOR 2 model in the X-direction. 22.170 Residual 0.095 126 0.001 UncorrectedSource Totalof Variations Sum 44.434 of Squares Degree of 128 Freedom (df) Mean Squares - CorrectedRegression Total 10.87644.340 127 2 22.170 - Residual 0.095 126 0.001 Uncorrected Total 44.434 128 - 2 The R valueCorrected of the modelTotal is equal 10.876 to: 127 - ! The R2 value of the model Residualis equal to: Sum of Squares 0.095 R2 = 1 = 1 = 0.991 (10) − CorrectedResidual SumSum o off Squares − 10.8760.095 R =1− =1− = 0.991 (10) Corrected Sum of Squares 10.876 Based on the amount obtained for R2, it can be said that the adjusted regression can describe 99.1%Based of the on sample the amount information. obtained Based for on R2 the, it calculationscan be said provided,that the adjusted the X-direction regression model can of describe LDOR accidents99.1% of the and sample incidents information. is as follows: Based on the calculations provided, the X-direction model of LDOR accidents and incidents is as follows: 0.012x0.804 P(X x) = e− . . + εi (11) P(X≥x≥ ) =e +ε (11) FigureFigure 19 19 shows shows thethe fittedfitted curvecurve forfor thethe LDORLDOR modelmodel inin thethe X-direction.X-direction. InIn thisthis figure,figure, thethe greengreen dots,dots, thethe fittedfitted values,values, andand thethe blueblue dotsdots areare the the initial initial values. values.
FigureFigure 19.19. LDORLDOR longitudinallongitudinal regressionregression curvecurve inin thethe locationlocation accidentaccident model.model.
At this step, two conditions are considered for the adequacy of the fitted model: the variance At this step, two conditions are considered for the adequacy of the fitted model: the variance of of the errors is constant and the errors are inconsistent. For this purpose, the scattering plot of the the errors is constant and the errors are inconsistent. For this purpose, the scattering plot of the independent variable is drawn versus the residuals (Figure 20). If the model is correct, the residuals independent variable is drawn versus the residuals (Figure 20). If the model is correct, the residuals should not have a regular structure. should not have a regular structure.
Int. J. Environ. Res. Public Health 2020, 17, 6085 21 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 21 of 36
Figure 20. x scatter plot versus the residuals for the LDOR model inin thethe X-direction.X-direction. As can be seen, the model is shaky in the x > 650 range, which can be due to a lack of data As can be seen, the model is shaky in the x > 650 range, which can be due to a lack of data in this in this range, but given that almost all data is less than 650, the adequacy of the model can be range, but given that almost all data is less than 650, the adequacy of the model can be considered considered acceptable. acceptable. The probability of the plane overruns from the end of the runway, in the x-direction and during The probability of the plane overruns from the end of the runway, in the x-direction and during the landing, and locates in the range a to b from the end of the runway is as follows: the landing, and locates in the range a to b from the end of the runway is as follows:
. 0.012x0.804 . FF (x) ==1−e1 e− (12)(12) X − Consequently: Consequently: Z b . 0.196 . 0.012x0.804 . PP((aa≤x≤bx b))== 0.0097x0.0097x− e− dxdx (13) (13) ≤ ≤ a
4.2.2. Transverse Location Model for LDOR 4.2.2. Transverse Location Model for LDOR In this section, the second location model for RE accidents is presented. The goal is to find In this section, the second location model for RE accidents is presented. The goal is to find FY(y) FY(y) or the probability of y (a, b) for any desired value of a and b. According to the statistical or the probability of y ∈ (a, b) for∈ any desired value of a and b. According to the statistical preliminary preliminary concepts: concepts: F (y) = P(Y y) = 1 P(Y y) (14) Y ≤ − ≥ F (y) =P(Y ≤ y) =1−P(Y ≥ y) (14) where F (y) is the probability function y and P(Y y) is probabilities of occurrence of Y less than y. Y ≤ Figurewhere 21FY (y)shows is the the probability location distribution function y ofand the P studied(Y≤y) LDOR is probabilities accidents of and occurrence incidents of (indirect Y less than Y) and y. FigureFigure 2221 showsshows ythe curve location versus distribution P(Y y). of the studied LDOR accidents and incidents (indirect Y) ≥ and Figure 22 shows y curve versus P(Y ≥ y).
Int. J. Environ. Res. Public Health 2020, 17, 6085 22 of 37 Int.Int. J.J. Environ.Environ. Res.Res. PublicPublic HealthHealth 20202020,, 1717,, xx 2222 ofof 3636
140 140 127127 120120
100100
8080
6060
4040
2020 66 55 22 00 2211 00 00 -- 2525 25 25 -- 5050 50 50 -- 7575 75 75 -- 100100 100 100 -- 125125 125 125 -- 150150 More More thanthan 150150 NumberNumber ofof occurrencesoccurrences
Figure 21. LDOR distribution in the Y direction (Horizontal axis shows the distance (meter) from Figure 21. LDOR distribution in the Y direction (Horizontal(Horizontal axis shows the distance (meter) from runway centerline). runway centerline).
Figure 22. y curve versus PP(Y(Y ≥y y)) forforLDOR. LDOR. Figure 22. y curve versus P(Y ≥≥ y) for LDOR. Like the model presented in the previous section, the relationship between the two variables is LikeLike thethe modelmodel presentedpresented inin thethe previousprevious section,section, thethe relationshiprelationship betweenbetween thethe twotwo variablesvariables isis exponential,exponential, so it is predicted that the PP(Y(Y ≥y y)) function function cancan bebe estimatedestimated usingusing family regression. exponential, so it is predicted that the P(Y ≥≥ y) function can be estimatedb using family regression. b y 1 ( ) = 0 + ε Based on Figure 22, the regression curve is considered as PPY(Y≥yy ) =ee− +εi, in which b00and b11 BasedBased onon FigureFigure 22,22, thethe regressionregression curvecurve isis consideredconsidered asas P(Y≥y≥ ) =e +ε ,, inin whichwhich bb0 andand bb1 are unknown parameters. Tables 13 and 14 show the SPSS software output. areare unknownunknown parameters.parameters. TablesTables 1313 andand 1414 showshow thethe SPSSSPSS softwaresoftware output.output. Table 13. Estimation of unknown parameters of the LDOR model in the Y-direction. TableTable 13.13. EstimationEstimation ofof unknownunknown parametersparameters ofof thethe LDORLDOR modelmodel inin thethe Y-direction.Y-direction. Unknown Standard Error of 95% Confidence95% Confidence Interval Interval Unknown EstimatedEstimated Value Standard Error of the Mean 95% Confidence Interval UnknownParameter Estimated Standardthe Mean Error (SE) of the MeanLower BoundLower Upper BoundUpper Parameter Value (SE) Lower Upper Parameter Value (SE) Bound Bound b0 0.932 0.018 0.897Bound 0.966Bound b0 0.932 0.018 0.897 0.966 b0b 1 0.9320.275 0.005 0.018 0.264 0.897 0.2860.966 b1 0.275 0.005 0.264 0.286 b1 0.275 0.005 0.264 0.286
Int. J. Environ. Res. Public Health 2020, 17, 6085 23 of 37
Table 14. Analysis of Variance (ANOVA) for the LDOR model in the Y-direction. Int. J. Environ. Res. Public Health 2020, 17, x 23 of 36 Source of Variations Sum of Squares Degree of Freedom (df) Mean Squares RegressionTable 14. Analysis of Variance 123.142 (ANOVA) for the LDOR 2 model in the Y-direction. 61.571 Residual 0.001 141 0.000 UncorrectedSource Totalof Variations Sum 123.143 of Squares Degree of Freedom 143 (df) Mean Squares - CorrectedRegression Total 14.767123.142 2 142 61.571 - Residual 0.001 141 0.000 The R2 valueUncorrected of the model Total is equal 123.143 to: 143 - Corrected Total 14.767 142 - ! 2 Residual Sum of Squares 0.001 The R value ofR the2 = model1 is equal to: = 1 = 1.000 (15) − Corrected Sum of Squares − 14.767 Residual Sum of Squares 0.001 R =1− =1− = 1.000 (15) Based on the amount obtainedCorrected for R 2Sum, it canof Squares be said that the14.767 adjusted regression can describe 100%Based of the sampleon the amount information. obtained Based for on R the2, it calculationscan be said provided,that the adjusted the Y-direction regression model can of describe LDOR accidents100% of the and sample incidents information. is as follows: Based on the calculations provided, the Y-direction model of LDOR accidents and incidents is as follows: 0.932x0.275 P(Y y) = e− + εi (16) ≥ . . P(Y ≥ y) =e +ε (16) Figure 23 shows the fitted curve for the LDOR model in the Y-direction. In this figure, the green Figure 23 shows the fitted curve for the LDOR model in the Y-direction. In this figure, the green dots, the fitted values, and the blue dots are the initial values. dots, the fitted values, and the blue dots are the initial values.
FigureFigure 23.23. LDORLDOR transversetransverse regressionregression curvecurve inin thethelocation locationaccident accidentmodel. model.
AtAt thisthis step,step, twotwo conditions are considered for for th thee adequacy adequacy of of the the fitted fitted model: model: the the variance variance of ofthe the errors errors is isconstant constant and and the the errors errors are are inconsiste inconsistent.nt. For For this this purpose, the scattering plotplot ofof thethe independentindependent variablevariable isis drawndrawn versusversus thethe residualsresiduals (Figure(Figure 24 24).). IfIf thethe modelmodel isis correct,correct, thethe residualsresiduals shouldshould notnot havehave aa regularregular structure. structure.
Int. J. Environ. Res. Public Health 2020, 17, 6085 24 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 24 of 36
Figure 24. y scatter plot versus the residuals for the LDOR model in the Y direction.
As cancan bebe seen, seen, the the model model is is shaky shaky in thein the x > x72 > range,72 rang whiche, which can becan due be todue a lack to a of lack data of in data this in range, this butrange, given but that given almost that allalmost data isall less data than is less 72, thethan adequacy 72, the adequacy of the model of the can model be considered can be considered acceptable. acceptable.The probability of the plane overruns from the runway end, in the y-direction and during the landing,The andprobability locates inof thethe range plane a overruns to b from from the runway the runway centerline end, in is asthe follows: y-direction and during the landing, and locates in the range a to b from the runway centerline is as follows: 0.932y0.275 FY(y) = 1 e− (17) − . . F (y) =1−e (17) Consequently: Consequently: Z b 0.275 P(a y b) = 0.2563y 0.725e 0.932y dy (18) − − . P(a≤≤ y≤≤b) = a 0.2563y . e . dy (18) 4.2.3. Lateral Location Model for LDVO 4.2.3.The Lateral third Location and final Model RE accident for LDVO location model, the LDVO model, is presented in this section. The goal is to find F (y) or the probability of y (a, b) for any desired value of a and b. According to The third and finalY RE accident location model,∈ the LDVO model, is presented in this section. the statistical preliminary concepts: The goal is to find FY(y) or the probability of y ∈ (a, b) for any desired value of a and b. According to the statistical preliminary concepts: F (y) = P(Y y) = 1 P(Y y) (19) Y ≤ − ≥ F (y) =P(Y ≤ y) =1−P(Y ≥ y) (19) where FY(y) is the probability function y and P(Y y) is probabilities of occurrence of Y less than y. where FY(y) is the probability function y and P(Y≤y≤ ) is probabilities of occurrence of Y less than y. Figure 25 shows the location distribution of the studied LDVO accidents and incidents (indirect Y) and Figure 25 shows the location distribution of the studied LDVO accidents and incidents (indirect Y) Figure 26 shows y curve versus P(Y y). and Figure 26 shows y curve versus ≥P(Y ≥ y).
Int. J. Environ. Res. Public Health 2020, 17, 6085 25 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 25 of 36 Int. J. Environ. Res. Public Health 2020, 17, x 25 of 36
35 35 29 30 29 30 25 25 20 20 15 15 10 7 10 6 5 6 7 5 5 5 11 0 11 0 0 0 0 - 25 25 - 50 50 -75 75 - 100 100 - 125 125 150 More than 0 - 25 25 - 50 50 -75 75 - 100 100 - 125 125 150 More150 than 150 Number of occurrences Number of occurrences
Figure 25. Location distribution of LDVO accidents (Horizontal axis shows the distance (meter) from FigureFigure 25.25. Location distributiondistribution ofof LDVOLDVO accidentsaccidents (Horizontal(Horizontal axisaxis showsshows thethe distancedistance (meter)(meter) fromfrom runway edge). runwayrunway edge).edge).
Figure 26. y curve versus P(Y ≥ y) for LDVO. Figure 26. y curve versus P(Y y) for LDVO. Figure 26. y curve versus P(Y≥ y) for LDVO. Like the previous models, the relationship between the two variables is exponential, so it is LikeLike thethe previousprevious models,models, thethe relationshiprelationship betweenbetween thethe twotwo variablesvariables isis exponential,exponential, soso itit isis predictedpredicted that the PP(Y(Y ≥y y)) function function cancan bebe estimatedestimated usingusing familyfamily regression.regression. Based on Figure 26,26, predicted that the P(Y ≥ y) function can be estimated using family regression. Based on Figure 26, ≥ b the regression curve is considered as P(Y≥y) =e b0y 1 +ε , in which b0 and b1 are unknown the regression curve is considered as P(Y y) = e − + εi, in which b0 and b1 are unknown the regression curve is considered as P(Y≥y) =e +ε , in which b0 and b1 are unknown parameters. Tables 15 and 16 show the SPSS≥ software output. parameters.parameters. TablesTables 1515 andand 16 16 show show thethe SPSSSPSS softwaresoftware output.output.
Table 15.15. Estimation ofof unknownunknown parametersparameters ofof thethe LDVOLDVO model.model. Table 15. Estimation of unknown parameters of the LDVO model. 95% Confidence Interval UnknownUnknown Estimated Standard ErrorError ofof the Mean 95% Confidence95% Confidence Interval Interval Unknown EstimatedEstimated Value Standard Error of the Mean Lower Upper ParameterParameter Value the Mean (SE)(SE) Lower BoundLower Upper BoundUpper Parameter Value (SE) Bound Bound b 0.048 0.006 0.036Bound 0.061Bound b0 0 0.048 0.006 0.036 0.061 bb0 1 0.0480.821 0.039 0.006 0.743 0.036 0.898 0.061 b1 0.821 0.039 0.743 0.898 b1 0.821 0.039 0.743 0.898
Int. J. Environ. Res. Public Health 2020, 17, 6085 26 of 37
Int. J. Environ. Res. Public HealthTable 2020 16., 17Analysis, x of Variance (ANOVA) for the LDVO model. 26 of 36
Source of VariationsTable 16. Analysis Sum of Squaresof Variance (ANO DegreeVA) offor Freedom the LDVO (df) model. Mean Squares Regression 17.461 2 8.731 ResidualSource of Variations Sum 0.118 of Squares Degree of Freedom 47 (df) Mean Squa 0.003res UncorrectedRegression Total 17.57917.461 2 49 8.731 - Corrected TotalResidual 4.0780.118 47 48 0.003 - Uncorrected Total 17.579 49 - Corrected Total 4.078 48 - The R2 value of the model is equal to: 2 The R value of the model is equal to: ! Residual Sum of Squares 0.118 R2 = 1 Residual Sum of Squares = 1 0.118 = 0.971 (20) R =1− − Corrected Sum of Squares =1−− 4.078 = 0.971 (20) Corrected Sum of Squares 4.078 2 BasedBased on on the the amount amount obtained obtained for for R R2, ,it it can can be be said said that that the the adjusted adjusted regression regression can can describe describe 97.1%97.1% of of the the sample sample information. information. Based Based on on the the ca calculationslculations provided, provided, the the model model of of LDVO LDVO accidents accidents andand incidents incidents is is as as follows: follows: 0.821 P(Y y) = e 0.048y + ε (21) . − . i P(Y ≥≥ y) =e +ε (21) Figure 27 shows the fitted curve for the LDVO model. In this figure, the green dots, the fitted Figure 27 shows the fitted curve for the LDVO model. In this figure, the green dots, the fitted values, and the blue dots are the initial values. values, and the blue dots are the initial values.
FigureFigure 27. 27. LDVOLDVO regression regression curve curve in in the the location location accident accident model. model.
AtAt this this step, step, two two conditions conditions are are considered considered for for th thee adequacy adequacy of ofthe the fitted fitted model: model: the the variance variance of theof theerrors errors is constant is constant and and the the errors errors are are inconsiste inconsistent.nt. For For this this purpose, purpose, the the scattering scattering plot plot of of the the independentindependent variable variable is is drawn drawn versus versus the the residuals residuals (Figure (Figure 28).28). If If the the model model is is correct, correct, the the residuals residuals shouldshould not not have have a a regular regular structure. structure.
Int. J. Environ. Res. Public Health 2020, 17, 6085 27 of 37 Int. J. Environ. Res. Public Health 2020, 17, x 27 of 36
FigureFigure 28.28. yy scatterscatter plotplot versusversus thethe residualsresiduals forfor thethe LDVOLDVO model.model.
FigureFigure 28 28 shows shows no no significant significant violations violations of the of defaults.the defaults. Therefore, Therefore, the selected the selected regression regression model ismodel quite is appropriate. quite appropriate. TheThe probabilityprobability ofof thethe planeplane veersveers ooffff fromfrom thethe runwayrunway edge,edge, inin thethe y-directiony-direction andand duringduring thethe landing,landing, andand locateslocates inin thethe rangerange aa toto bb fromfrom thetheedge edge of of the the runway runway is is as as follows: follows:
. ( ) . 0.048y0.821 (22) FF (yy) ==1−e1 e− (22) Y − Consequently: Consequently: Z b . ( ) . 0.179 . 0.048y0.821 PP(aa≤yy ≤bb) == 0.03940.0394yy− ee− dydy (23) (23) ≤ ≤ a TableTable 17 17 summarizes summarizes the the models models presented presented in this in section.this section. It should It should be noted be that noted in the that proposed in the models,proposed the models, variables the must variables be entered must be in theentered meter. in the meter.
TableTable 17.17. SummarizeSummarize ofof accidentaccident locationlocation model.model.
AccidentAccident Type Type Point Point Model Model Interval Interval Model Model RR22 0.804 R b 0.804 LDOR (X) P(X x) = e 0.012x . . ( ) = 0.196 . 0.012x . . 0.991 LDOR (X) P(X≥x) =e− P(a≤x≤ba x b ) = a 0.0097x 0.0097x− e−e dx dx 0.991 ≥ ≤ ≤ 0.275 R 0.275 ( ) 0.932y b 0.725 0.932y LDOR (Y) P Y y = e− . P(a y b) = 0.2563y− e− . dy 1.00 LDOR (Y) P(Y≥≥y) =e . P(a≤y≤b≤ ≤ ) = a 0.2563y . e . dy 1.00 0.821 R b 0.821 LDVO P(Y y) = e 0.048y P(a y b) = 0.0394y 0.179e 0.048y dy 0.971 − a − − ≥ . ≤ ≤ . LDVO P(Y ≥y) =e . P(a≤y≤b) = 0.0394y . e . dy 0.971 4.3. Accident Consequence Model 4.3. AccidentAs described Consequence in Section Model 3.1.3 , the accident consequence model for both LDOR and LDVO accidentsAs described is constructed in Section according 3.1.3, to the coeaccidentfficients consequence obtained in model the accident for both location LDOR model. and LDVO Based onaccidentsEquations is constructed (7) and (8), according Table 18 shows to the the coefficients accident consequenceobtained in the model accident for both location LDOR model. and LDVO. Based Allon parametersEquations (7) of and Table (8), 18 Table are fully 18 shows introduced the accident in Section consequence 3.1.3. model for both LDOR and LDVO. All parameters of Table 18 are fully introduced in Section 3.1.3.
Int. J. Environ. Res. Public Health 2020, 17, 6085 28 of 37
Int. J. Environ. Res. Public Health 2020, 17, x 28 of 36 Table 18. Accident consequence model. Table 18. Accident consequence model. Accident Type Interval Model Accident Type Interval Model 0.275 0.275 0.932y 0.932y 0.804 P (e− . c e− f . ) LDOR . . 0.012(xi+∆i) Psc =(e −e−2 ) e− . LDOR P = × ×e . ( ∆ ) 0.821 2 0.048y0.821 0.048y LDVO = e− c e− f Psc . −2 . e . −e . LDVO P = 2 4.4. Practical Application of Proposed Models 4.4. MathematicalPractical Application modeling of Proposed of various Models phenomena is encountered with the success of society when it can beMathematical proved by its modeling practical of and various intuitive phenomena application. is encountered In this regard, with this the studysuccess also of society examines when the practicalit can be applicationproved by its of practical the proposed and intuitive models. application. Mehrabad In International this regard, this Airport study and also Hasheminejad examines the Internationalpractical application Airport haveof the been proposed selected models. as case Me studieshrabad to reviewInternational the proposed Airport models.and Hasheminejad The general stepsInternational for using Airport models have are: been selected as case studies to review the proposed models. The general (1)stepsSelection for using of models the airport are: under review (2)(1) SelectSelection the of runway the airport to be under examined review and determine the obstacles around it (3)(2) CollectionSelect the runway of NOD to data be relatedexamined to theand desired determine airports the obstacles in at least around one year, it to update the accident (3) probabilityCollection of model NOD data related to the desired airports in at least one year, to update the accident (4) Investigateprobability dimodelfferent scenarios about bopping with existing obstacles (5)(4) SelectInvestigate the flight different to be scenarios studied and abou determinet bopping the with variables existing required obstacles in the models (5) Select the flight to be studied and determine the variables required in the models (6) Calculate the probability of a RE accident (6) Calculate the probability of a RE accident (7) Calculate the probability of a collision with an obstacle (accident location model) (7) Calculate the probability of a collision with an obstacle (accident location model) (8) Calculate the consequence probability of an accident (8) Calculate the consequence probability of an accident AfterAfter inspectinginspecting bothboth airports,airports, the Can river at Mehrabad Airport Airport and and a a Radio Radio antenna antenna at at HashemiehnejadHashemiehnejad AirportAirport werewere selectedselected asas obstacles.obstacles. 4.4.1. Risk Assessment of LDOR at Mehrabad International Airport because of Can River 4.4.1. Risk Assessment of LDOR at Mehrabad International Airport because of Can River TheThe runways runways 11L 11L and and 11R 11R of of Mehrabad Mehrabad AirportAirport overlapoverlap withwith thethe CanCan riverriver whichwhich isis crossedcrossed byby a bridge.a bridge. Figure Figure 29 29 shows shows this this area area of of Mehrabad Mehrabad Airport.Airport. In this study, study, the the runway runway 11R 11R is is considered. considered. TheThe exact exact specifications specifications ofof thethe runwayrunway 11R11R and the Can river are shown in in Figure Figure 30.30. Based Based on on this this figure,figure, in in distance distance between between 155 155 toto 205205 m,m, if the airc aircraftraft deviates 35 35 m m from from the the centerline centerline of of the the runway runway toto the the right, right, it it will will fallfall intointo thethe CanCan river.river.
FigureFigure 29.29. InterferenceInterference of Can river with Mehrabad Mehrabad Airport Airport runways. runways.
Int. J. Environ. Res. Public Health 2020, 17, 6085 29 of 37
Int. J. Environ. Res. Public Health 2020, 17, x 29 of 36
FigureFigure 30. 30.The The exact exact location location ofof thethe runway 11R 11R and and the the river river can can interference. interference.
ThenThen we havewe have to chooseto choose a flighta flight to to Mehrabad Mehrabad airport and and determine determine its itsspecifications. specifications. Table Table 19 19 includes the introduction of the selected flight. After selecting the desired flight, the variables related includes the introduction of the selected flight. After selecting the desired flight, the variables related to the probability model of accidents should be determined. Meteorological Aviation Routine to the probability model of accidents should be determined. Meteorological Aviation Routine Weather Weather Report (METAR) related to Iran’s daily flights can be obtained from www.irimo.ir. Table 20 Reportshows (METAR) the variables related for to flight Iran’s 2807. daily flights can be obtained from www.irimo.ir. Table 20 shows the variables for flight 2807. Table 19. Selected flight specifications at Mehrabad Airport. Table 19. Selected flight specifications at Mehrabad Airport. Origin Mashhad OriginDestination Tehran Mashhad Date 02.12.2016 LandingDestination time in the destination 18:58 Tehran FlightDate number 2807 02.12.2016 Landing time inAirline the destination Meraj 18:58 FlightAircraft number type Airbus 2807320 Airline Meraj Table 20. Variables relatedAircraft to the typeprobability model of RE Airbus accidents 320 for flight 2807.
Variable Value or Type Description Table 20. VariablesUser Class related to the probability C model of RE accidents Commercial for flight 2807. Maximum Takeoff Weight (MTOW) C Large Jet (B737, A320, etc.) VariableEquipment Class Value or Type Jet Description - ForeignUser Class Origin/Destination C Domestic Commercial - Maximum TakeoCeilingff WeightHeight +2500 More than 2500 ft C Large Jet (B737, A320, etc.) (MTOW)Visibility +8 More than 8 miles EquipmentCrosswind Class Between Jet 2 and 5 3 knots - Foreign Origin/TailwindDestination Domestic Less than 5 3 knots - CeilingAir Height Temperature Between+2500 5 and 15 More 6 than ºC 2500 ft VisibilityGust + Do8 not exist More than - 8 miles CrosswindThunderstorm Between Do 2 not and exist 5 3 knots- TailwindRain Less Do than not 5 exist 3 knots- Air TemperatureSnow Between Do 5 not and exist 15 6- ºC GustFog Do not Do existnot exist - - ThunderstormIcing Condition Do not Do existnot exist - - FrozenRain Precipitation Do not Do existnot exist - - Hub/Non-HubSnow Airport Do not exist Hub - - LogFog Criticality Factor Do not CFexist > 0 Needed length - is 6049 ft Icing ConditionNight Condition Do not It existis night - - Frozen Precipitation Do not exist - Hub/Non-Hub Airport Hub - Log Criticality Factor CF > 0 Needed length is 6049 ft Night Condition It is night - Int. J. Environ. Res. Public Health 2020, 17, 6085 30 of 37
Based on the logistics regression model presented in the Equation (4), the general form of the accident probability model is as follows:
1 1 P Accident_Occurence = = (24) (b +b X +b X +b X +...) Z { } 1 + e− 0 1 1 2 2 3 3 1 + e− According to Tables5 and6, the value of Z for LDOR accident is calculated as follows:
Z = 13.065+ 1.539(G) 0.498(T/C) 1.013(A/B) + 0.935(D/E/F) − − − 0.019(Ceiling < 200ft) 0.772(Ceiling 200 to 1000 ft) − − 0.345(Ceiling 1000 to 2500 ft) + 2.881(Visibility < 2SM) − +1.532(Visibility 2 to 4 SM) + 0.200(Visibility 4 to 8SM) 0.913(Xwind 5 to 12 kt) 1.342(Xwind 2 to 5kt) − − (25) 0.921(Xwind 12 kt) + 0.786(Tailwind 12kt) − ≥ ≥ +0.043(Temp < 5 C) 0.019(Temp 5 to15C) − 1.067(Temp 25 C) + 2.007(Icing Condition) + 0.449(Snow) − ≥ 1.344(Thunderstorm) + 0.929(Foreign OD) − +1.334( Hub/Non Hub Airport) + 9.237(Log Criticality Factor) − Now, by replacing the model variables with values of 0 and 1 of the selected flight (according to Table 20), the probability of an LDOR accident at Mehrabad Airport resulting from the flight number 2807 of Meraj airlines is obtained.
Z = 13.065+ 1.539(0) 0.498(0) 1.013(0) + 0.935(0) 0.019(0) 0.772(0) − − − − − 0.345(0) + 2.881(0) + 1.532(0) + 0.200(0) 0.913(0) − − 1.342(1) 0.921(0) + 0.786(0) + 0.043(0) 0.019(1) − − − (26) 1.067(0) + 2.007(0) + 0.449(0) 1.344(0) + 0.929(0) − − +1.334(0) + 9.237(1) = 5.189 − ( ) = 1 = 1 = 3 = P LDOR + Z ( 5.189) 5.547 10− 0.5547% 1 e− 1+e− − × As a result, the probability of overrun from runway end for the Airbus 320 of Meraj airline is 0.55%. If the aircraft overruns from the runway end, the next step is to calculate the probability of falling into the Can river. According to Table 17, the probability of falling in the Can river is equal to:
R 0.804 ( ) = 255 0.196 0.012x = = P 155 x 205 155 0.0097x− e− dx 0.0805 8.05% ≤ ≤ 0.275 (27) P(Y 35) = e 0.932(35) = 0.0839 = 8.39% ≥ − The probability of the aircraft overrun from the runway end and falling into the Can river is equal to: 5 P(CS) = P(LDOR) P(155 x 205) P(Y 35) = 3.7464 10− (28) × ≤ ≤ × ≥ × Since the type of obstacle, in this case, is the river, so the consequences of the accident are expressed as a fall and not a collision. The fall will occur only when the aircraft deviates 35 m from the runway centerline. On the other hand, to calculate the accident with severe consequences, according to Table 18, we need two critical values: yc and yf. Therefore, it is not possible to use the accident consequences model in this case.
4.4.2. Risk Assessment of LDVO at Hasheminejad International Airport because of Radio Antenna In this section, the possibility of colliding aircraft landing on the runways 31L and 31R with the radio antenna between these two runways is examined. Figures 31 and 32 show the location of this antenna relative to the desired runway. An input flight to Hasheminejad airport has been selected and Int. J. Environ. Res. Public Health 2020, 17, 6085 31 of 37 its specifications are given in Table 21. Also, Table 22 shows the variables related to the probability model of accidents on this flight. Int.Int. J. J. Environ. Environ. Res. Res. Public Public Health Health 20202020,, 1717,, xx 3131 of of 36 36
FigureFigure 31.31. RadioRadio antenna antenna between runways 31L and 31R 31R in in Hasheminejad Hasheminejad airport. airport.
Figure 32. The exact location of the Radio antenna relative to runway 31L. Figure 32. The exact location of the Radio antenna relative to runway 31L. Table 21. Selected flight specifications at Hasheminejad Airport. Table 21. Selected flightflight specificationsspecifications at Hasheminejad Airport.Airport. Origin Tehran OriginOrigin Tehran Destination Mashhad DestinationDestinationDate 02.13.2016 Mashhad
Landing timeDateDate in the destination 02.13.2016 20:03 LandingLanding time Flighttime in in thenumber the destination destination 20:03 6254 FlightFlightAirline number number Taban 6254 Airline Taban AircraftAirline type McDonnellTaban Douglas MD-88 AircraftAircraft type type McDonnell DouglasDouglas MD-88MD-88
Int. J. Environ. Res. Public Health 2020, 17, 6085 32 of 37
Table 22. Variables related to the probability model of RE accidents for flight 6254.
Variable Value or Type Description User Class C Commercial Maximum Takeoff Weight (MTOW) C Large Jet (B737, A320, etc.) Equipment Class Jet - Foreign Origin/Destination Domestic - Ceiling Height +2500 More than 2500 ft Visibility Between 5 and 15 3.72 miles Crosswind Less than 2 1 knot Tailwind Less than 5 4 knots Air Temperature Less than 5 2 C − ◦ Gust Do not exist - Thunderstorm Do not exist - Rain Do not exist - Snow Do not exist - Fog Do not exist - Icing Condition Do not exist - Frozen Precipitation Do not exist - Hub/Non-Hub Airport Hub - Log Criticality Factor CF > 0 Needed length is 5813 ft Night Condition It is night -
Based on Equation (4), the general form of the accident probability model is as follows:
1 1 P Accident_Occurence = = (29) (b +b X +b X +b X +...) Z { } 1 + e− 0 1 1 2 2 3 3 1 + e− According to Tables5 and6, the value of Z for LDVO accident is calculated as follows:
Z = 13.088+ 1.682(G) 0.770(A/B) 0.252(D/E/F) + 2.413(Visibility < 2SM) − − − +0.653(Xwind 5 to 12 kt) 0.091(Xwind 2 to 5 kt) − +2.192(Xwind 12 kt) + 0.066(Tailwind5 to 12kt) ≥ +0.980(Tailwind 12 kt) + 0.558(Temp < 5 C) ≥ 0.453(Temp 5 to 15C) + 0.291(Temp 25 C) (30) − ≥ +2.670(Icing Condition) 0.126(Gusts) + 0.548(Snow) − 0.103(Frozen Precipitation) 0.036(Gusts) + 1.74(Fog) − − 2.517(Turboprop) 0.334(Foreign OD) − − +4.318(Log Criticality Factor) 1.360(Night Condition) − Now, by replacing the model variables with values of 0 and 1 of the selected flight (according to Table 22), the probability of an LDVO accident at Mehrabad Airport resulting from the flight number 6254 of Taban airlines is obtained.
Z = 13.088+ 1.682(0) 0.770(0) 0.252(0) + 2.143(0) + 0.653(0) 0.091(0) − − − − +2.192(0) + 0.066(0) + 0.980(0) + 0.558(1) 0.453(0) − +0.291(0) + 2.670(0) 0.126(0) + 0.548(0) 0.103(0) (31) − − 0.036(0) + 1.74(0) 2.517(0) 0.334(0) + 4.318(1) 1.360(1) − − − − = 9.572 −
1 1 5 P(LDVO) = = = 6.965 10− = 0.006965% (32) Z ( 9.572) 1 + e− 1 + e− − × Int. J. Environ. Res. Public Health 2020, 17, 6085 33 of 37
As a result, the probability of deviation from the runway edge for flight 6254 of Taban airline is 0.01 percent. According to Table 17, if aircraft deviates from the runway, the probability of collision it with the radio antenna is:
0.048(50)0.821 P(Y 50) = e− = 0.3038 = 30.38% (33) ≥ Int. J. Environ.Int. J. Environ. Res. Public Res. PublicHealth Health2020, 17 2020, x , 17, x 33 of 3633 of 36 Therefore: 3 P(CS) = P(LDVO) P(Y 50) = 2.115 10− (34) P(CS)P=P(CS)(LDVO=P(LDVO) ×P()Y≥50×P× (Y≥50) ≥= 2.115) = 2.115 × 10 ×× 10 (34) (34) Therefore, it can be said that the probability of Taban airlines flight 6254 deviates from the edge of Therefore,Therefore, it can it be can said be thatsaid the that probability the probability of Taban of Taban airlines airlines flight flight 6254 deviates6254 deviates from fromthe edge the edge the runway and collides with the radio antenna of Hasheminejad airport is equal to 0.002115. of theof runway the runway and collides and collides with withthe radio the radio antenn antenna of Hasheminejada of Hasheminejad airport airport is equal is equal to 0.002115. to 0.002115. About the probability of an accident with severe consequences, critical transverse distances (yc AboutAbout the probability the probability of an ofaccident an accident with withsevere severe consequences, consequences, critical critical transverse transverse distances distances (yc (yc and yf) must first be obtained. These distances are determined with the help of aircraft dimensions and yandf) must yf) mustfirst befirst obtained. be obtained. These These distances distances are determined are determined with withthe help the helpof aircraft of aircraft dimensions dimensions (See Figure 33). (See Figure(See Figure 33). 33).
FigureFigure 33. Dimensions 33. Dimensions of aircraft of aircraft MD-88. MD-88.
AccordingAccording to Figure to Figure 33, the 3333, ,wingspan the wingspan of the of airc theraft aircraftairc MD-88raft MD-88 is 32.9 is m. 32.9 According m. According According to Figure to to Figure Figure 12, 12,12, the consequencesthe consequences of colling of colling with withan obstacle an obstacle is seve isre severeseve whenre when the obstacle the obstacle is located is located in the in orange the orange area area (middle(middle 1/3 wingspan). 11/3/3 wingspan). Figure Figure 34 shows 34 shows the critical the critical positions positions of the of obstacle the obstacle over overthe MD-88 the MD-88 aircraft. aircraft.
FigureFigure 34. Determination 34.34. Determination of critical of critical aircraft aircraftaircraft location locationlocation for consequences forfor consequencesconsequences model. model.model. Note: Note:Note:Distance Distance from from radio radioantenna antenna to runway to runway axis is axis 83 mis (See 83 m Figure (See Figure 32). Also, 3232)).. Also,when Also, when whenthe obstacle the the obstacle is at positions is is at at positions positions 1 and 1 12, and and 2, 2, distancedistance from nose from wheel nose wheel to runway to runway axis names axisaxis namesnames yf and yy fyf andc, respectively. yc,, respectively. respectively.
As shownAs shown in Figure in Figure 34, y c34, is 77.525yc is 77.525 m. With m. Witha similar a similar argument, argument, yf is equivalentyf is equivalent to 88.475 to 88.475 m. m. BasedBased on the on equati the equations inons Table in Table 19: 19:
. . . . . . . ( . ) . . . ( . ) . . e e . −e −e . e e . ( . )−e −e . ( . ) P =P = = = = 0.0161= 0.0161 (35) (35) 2 2 2 2 If theIf aircraft the aircraft collides collides with withthe radio the radio antenna, antenna, 1.61% 1. of61% accidents of accidents will likelywill likely occur occur with withsevere severe consequences.consequences. As a Asresult, a result, the prtheobability probability of an of accident an accident with withsevere se vereconsequences consequences due todue the to the departuredeparture of aircraft of aircraft No. 6254 No. from6254 fromthe side the ofside th eof runway the runway 31L of 31L Hasheminejad of Hasheminejad airport airport and collision and collision with withthe radio the radio antenna antenna is equal is equal to: to:
P(LDVOP(LDVO Severe Severe) =P)(LDVO=P(LDVO) ×P()Y≥50×P(Y≥50) ×P ) ×P= 3.40515 = 3.40515 × 10 × 10 (36) (36)
Int. J. Environ. Res. Public Health 2020, 17, 6085 34 of 37
As shown in Figure 34, yc is 77.525 m. With a similar argument, yf is equivalent to 88.475 m. Based on the equations in Table 19:
0.048y0.821 0.048y0.821 0.048(77.525)0.821 0.048(88.475)0.821 e c e− f e e P = − − = − − − = 0.0161 (35) sc 2 2 If the aircraft collides with the radio antenna, 1.61% of accidents will likely occur with severe consequences. As a result, the probability of an accident with severe consequences due to the departure of aircraft No. 6254 from the side of the runway 31L of Hasheminejad airport and collision with the radio antenna is equal to:
5 P(LDVO Severe) = P(LDVO) P(Y 50) Psc = 3.40515 10− (36) × ≥ × × The evaluation of Mehrabad and Hasheminejad airports verified that presented models can be used well in Iran. After setting the risk threshold by the airport officials, it is possible to make the necessary decisions and budget for the next steps. For example, the acceptable level of risk at Mehrabad Airport may be such that a decision is made to cover the Can river. But the risk threshold at Hasheminejad Airport may be such that there is no need to move the radio antenna. Therefore, the models presented in this paper can be a good criterion for evaluating the obstacles around the runway to prevent unprofessional, unsafe, and uneconomical decisions. In general, these accident assessment models will improve the performance of the air transportation system, both in terms of safety and economics. In other words, these models help engineers make the most optimal decisions and always be in the position that has the safest and most economical conditions possible. Besides, the models presented in this paper help to eliminate the permanent lack of a proper risk assessment system for runways at Iranian airports.
5. Conclusions In today’s world, the development and advancement of the air transportation industry are evident. Airports are one of the most important and vital centers. Runways are also one of the main physical parts of airports. Accidents on runways cause them to be blocked. Blocking the runway also significantly reduces airport capacity (At airports with one runway, capacity drops to zero). Also, these accidents may danger human lives and environmental elements. Therefore designing safer runways is of utmost importance. The primary goal of this study is to achieve a risk assessment model of RSA. A review of studies and research in this field showed that only the United States has been able to achieve a comprehensive and complete model for assessing the risk of accidents and incidents resulting from the departure of aircraft from runways. Therefore, an attempt was made to calibrate the model for Iranian airports. Focusing on RE accidents and carefully examining predictive models of the probability of occurrence, location, and severity of consequences when smashing with obstacles have been the mainstays of this study. Therefore, after reviewing much research conducted in this field, an attempt was made to compile a well-written and complete process to collect data from modeling RE accidents in Iran. In this way, the lack of systematic documentation and the lack of access to the codes of NOD flights led to the inability to build a model of the probability of accidents. Subsequently, the accuracy of the constructed models was confirmed by statistical tests. In the end, Mehrabad International Airport in Tehran and Hasheminejad International Airport in Mashhad were examined by the models. Compared to the comprehensive and famous model provided in the United States, the probability model in this paper is the same because there was no suitable database for NOD flights in Iran. In location and consequences models, the general form of the models was chosen similar to the American model, but the model coefficients presented in this paper are different from the American model coefficients. In fact, the model presented by the authors has been calibrated for Iran and allows the runway risk assessment to be done more accurately. Now, using the presented models, we can assess the risk of Int. J. Environ. Res. Public Health 2020, 17, 6085 35 of 37 obstacles around the runways of airports and decide about them based on the level of risk and the requirements of the conditions. New airports can also be located in such a way that we have the least risk in the event of aircraft accidents. This decision is very important in times of crisis because it can have a significant impact on the continuity of airport services. In the long-term, these models will help increase public health and protect the environment. The most important limitation of this study was the lack of access to NOD flight data. Therefore, for future studies, the authors suggest designing and setting up an accurate system for recording all the necessary information. In this case, the model of the probability of accidents can be calibrated for Iran. The authors also believe that adding new metrics such as obstacle type and dimensions, obstacles density, etc. in future studies can improve models and make them more comprehensive. Another idea that the authors have in mind for future studies is to develop their modeling so that other types of accidents, including LDUS, TOOR, and TOVO, can be evaluated.
Author Contributions: Conceptualization, Y.Y. and N.K.; methodology, Y.Y., N.K. and D.M.; software, A.M. and D.M.S.; validation, Y.Y. and A.M.; formal analysis, Y.Y.; investigation, N.K.; resources, D.M.S. and A.S.Z.; data curation, N.K., D.M.S. and A.S.Z.; writing—original draft preparation, N.K.; writing—review and editing, A.M., D.M. and D.M.S.; visualization, D.M.S.; supervision, D.M.; project administration, A.M.; funding acquisition, A.M. All authors have read and agreed to the published version of the manuscript. Funding: We acknowledge the financial support of this work by the Hungarian State and the European Union under the EFOP-3.6.1-16-2016-00010 project and the 2017-1.3.1-VKE-2017-00025 project. This research has been supported by the Project: “Support of research and development activities of the J. Selye University in the field of Digital Slovakia and creative industry” of the Research and Innovation Operational Programme (ITMS code: NFP313010T504) co-funded by the European Regional Development Fund. This research has been in part supported by the Alexander von Humboldt Foundation. Conflicts of Interest: The authors declare no conflict of interest.
References
1. Bai, M.; Yang, W.; Song, D.; Kosuda, M.; Szabo, S.; Lipovsky, P.; Kasaei, A. Research on Energy Management of Hybrid Unmanned Aerial Vehicles to Improve Energy-Saving and Emission Reduction Performance. Int. J. Environ. Res. Public Health 2020, 17, 2917. [CrossRef][PubMed] 2. Košˇcák, P.; Berežný, Š.; Vajdová, I.; Koblen, I.; Ojciec, M.; Matisková, D.; Puškáš, T. Reducing the Negative Environmental Impact of Winter Airport Maintenance through Its Model Design and Simulation. Int. J. Environ. Res. Public Health 2020, 17, 1296. [CrossRef][PubMed] 3. Yousefi, Y. Runway Safety and Methods for Assessment of Accident Risk, Case Study: Some Airports in Iran. Master’s Thesis, Department of Civil Engineering, Iran University of Science and Technology, Tehran, Iran, 2016. 4. Henriksen, L.F.; Ponte, S. Public orchestration, social networks, and transnational environmental governance: Lessons from the aviation industry. Regul. Gov. 2018, 12, 23–45. [CrossRef] 5. Džunda, M.; Dzurovˇcin,P.; Koblen, I.; Szabo, S., Jr.; Jenˇcová, E.; Cekan,ˇ P.; Korba, P.; F˝oz˝o,L.; Melníková, L.; Tobisová, A.; et al. Selected Aspects of Navigation System Synthesis for Increased Flight Safety, Protection of Human Lives, and Health. Int. J. Environ. Res. Public Health 2020, 17, 1550. [CrossRef][PubMed] 6. Kumar, A.; Aswin, A.; Gupta, H. Evaluating green performance of the airports using hybrid BWM and VIKOR methodology. Tour. Manag. 2020, 76, 103941. [CrossRef] 7. Arora, D.; Ravi, S. Issues and Challenges of Indian Aviation Industry: A Case of Jet Airways. Rev. Manag. 2019, 9, 19–23. 8. Van Eekeren, R.; Wright, S.; Cokorilo,ˇ O. Early Cost Safety Analysis of Runway Events. Int. J. Traffic Transp. Eng. 2018, 8, 261–270. 9. Akinyemi, O.O.; Adebiyi, K.A. Modelling uncertainty in runway safety intervention performance evaluation. Int. J. Reliab. Saf. 2016, 10, 158–173. [CrossRef] 10. Dambier, M.; Hinkelbein, J. Analysis of 2004 German general aviation aircraft accidents according to the HFACS model. Air Med. J. 2006, 25, 265–269. [CrossRef][PubMed] 11. Cokorilo,ˇ O.; De Luca, M.; Dell’Acqua, G. Aircraft safety analysis using clustering algorithms. J. Risk Res. 2014, 17, 1325–1340. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 6085 36 of 37
12. Green, L.L. Development of A Bayesian Belief Network Runway Incursion and Excursion Model. In Proceedings of the American Society for Engineering Management 2014 International Annual Conference, Virginia Beach, VA, USA, 15–18 October 2014. 13. Ravenscroft, D.L.; Verbeke, J.P.; Pschierer, K.C. System and Method for Detecting and Preventing Runway Incursion, Excursion and Confusion. U.S. Patent 8,401,774, 19 March 2013. 14. Akinyemi, O.; Adebiyi, K. Development of system dynamics based simulation models for runway safety planning. Int. J. Ind. Eng. Prod. Res. 2019, 30, 381–403. 15. Bhargava, D.; Marais, K. Understanding Current Ways of Reporting Runway Incursion Incidents at Towered Airports. In Proceedings of the 20th International Symposium on Aviation Psychology, Dayton, OH, USA, 7–10 May 2019; p. 1. 16. Chen, Z.H.; Juang, J.C. Mitigation of Runway Incursions by Using a Convolutional Neural Network to Detect and Identify Airport Signs and Markings. Sens. Mater. 2019, 31, 3947–3958. [CrossRef] 17. Zhang, J.; Liu, H.; Yang, Q. Design and Implementation of Runway Incursion Automatic Warning Simulation System. In Proceedings of the 2019 5th International Conference on Transportation Information and Safety (ICTIS), Liverpool, UK, 14–17 July 2019; pp. 611–613. 18. Suroso, H.C.; Revadi, C.E. The Main Factors that Affect Pilot Attention and Decision Making During Landing Operation Leading to Runway Incursion. In Proceedings of the 2019 1st International Conference on Engineering and Management in Industrial System (ICOEMIS 2019), Malang, Indonesia, 8–9 August 2019. 19. Chang, Y.H.; Yang, H.H.; Hsiao, Y.J. Human risk factors associated with pilots in runway excursions. Accid. Anal. Prev. 2016, 94, 227–237. [CrossRef][PubMed] 20. Vorobyeva, O.; Bartok, J.; Šišan, P.; Nechaj, P.; Gera, M.; Kelemen, M.; Polishchuk, V.; Gaál, L. Assessing the Contribution of Data Mining Methods to Avoid Aircraft Run-Off from the Runway to Increase the Safety and Reduce the Negative Environmental Impacts. Int. J. Environ. Res. Public Health 2020, 17, 796. [CrossRef] 21. Gandhewar, P.; Sonkusare, H.G. Runway Excursion: A Problem. IOSR J. Mech. Civ. Eng. 2014, 11, 75–78. [CrossRef] 22. National Academies of Sciences, Engineering, and Medicine. Analysis of Aircraft Overruns and Undershoots for Runway Safety Areas (ACRP Report 3); The National Academies Press: Washington, DC, USA, 2008. 23. Moretti, L.; Di Mascio, P.; Nichele, S.; Cokorilo, O. Runway veer-off accidents: Quantitative risk assessment and risk reduction measures. Saf. Sci. 2018, 104, 157–163. [CrossRef] 24. National Academies of Sciences, Engineering, and Medicine. Improved Models for Risk Assessment of Runway Safety Areas (ACRP Report 50); The National Academies Press: Washington, DC, USA, 2011. 25. Eddowes, M.; Hancox, J.; MacInnes, A. Final Report on the Risk Analysis in Support of Aerodrome Design Rules; Report for the Norwegian Civil Aviation Authority; AEA Technologies plc: Warrington, UK, 2001. 26. Moretti, L.; Cantisani, G.; Di Mascio, P.; Nichele, S.; Caro, S. A runway veer-off risk assessment based on frequency model: Part I. Probability analysis. In Proceedings of the International Congress on Transport Infrastructure and Systems TIS, Rome, Italy, 10–12 April 2017; p. 12. 27. Moretti, L.; Cantisani, G.; Caro, S. Airport veer-off risk assessment: An Italian case study. ARPN J. Eng. Appl. Sci. 2017, 12, 900–912. 28. Kirkland, I.D.L.; Caves, R.E.; Humphreys, I.M.; Pitfield, D.E. An improved methodology for assessing risk in aircraft operations at airports, applied to runway overruns. Saf. Sci. 2004, 42, 891–905. [CrossRef] 29. Wong, D.K.Y.; Pitfield, D.E.; Caves, R.E.; Appleyard, A.J. The development of a more risk-sensitive and flexible airport safety area strategy: Part I. The development of an improved accident frequency model. Saf. Sci. 2009, 47, 903–912. [CrossRef] 30. Ayres Jr, M.; Shirazi, H.; Carvalho, R.; Hall, J.; Speir, R.; Arambula, E.; David, R.; Gadzinski, J.; Caves, R.; Wong, D. Modelling the location and consequences of aircraft accidents. Saf. Sci. 2013, 51, 178–186. [CrossRef] 31. Jeon, J.; Song, J.; Kim, H.; Song, B. A Research on the Advanced Risk Assessment of Runway Safety Areas with Enhanced Algorithm. Int. J. Sci. Eng. Technol. 2016, 5, 67–70. 32. Shin, D.W.; Kim, W.Y. The Study on the Runway Safety Area for the Light Sport Aircraft. J. Korean Soc. Aviat. Aeronaut. 2013, 21, 41–45. [CrossRef] 33. Štumper, M.; Kraus, J.; Vittek, P.; Lališ, A. Runway safety areas. MAD Mag. Aviat. Dev. 2015, 3, 5–8. [CrossRef] 34. Szabo, S.; Vittek, P.; Kraus, J.; Plos, V.; Lališ, A.; Štumper, M.; Vajdova, I. Probabilistic model for airport runway safety areas. Transp. Probl. 2017, 12, 89–97. [CrossRef] Int. J. Environ. Res. Public Health 2020, 17, 6085 37 of 37
35. Potente, C.; Ragnoli, A.; Tamasi, G.; Vergari, R.; Di Mascio, P. Quantitative Risk Assessment of Temporary Hazards and Maintenance Worksites in the Airport Safety Areas: A case study. Transp. Res. Procedia 2018, 35, 166–175. [CrossRef] 36. White, J.; Subbotin, N. Development of Engineered Materials Arresting Systems from 1994 through 2003; U.S. Department of Transportation, Federal Aviation Administration: Washington, DC, USA, 2018. 37. ICAO. Aeronautical Information Publication. Available online: www.icao.int (accessed on 10 April 2015). 38. Hong, S.B.; Dilshod, T.; Kim, D.H. An Application of the Risk Assessment Model for Runway End Safety Areas to a Specific Airport in Korea. Int. J. Control. Autom. 2016, 9, 287–298. [CrossRef] 39. IATA. Safety Report 2014. 2015. Available online: http://www.aviatian-accidents.net/report-download.php (accessed on 10 April 2015).
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).