There are large amount of literature discussing the subjects of sustainability/sustainable transportation, performance evaluation and the methods of applying sustainability for transportation agencies.
1.1. Sustainability and Sustainable Development
Sustainability, explored in terms of the theories of sustainable development, can be described by many expressions, in the majority of them the three dimensions (social, economical and environmental) appear clearly. A clear relation links the dimensions together i.e. Economic systems are contained within a social framework; similarly, society exists within the natural environment. These should be taken into account when following termed as a triple bottom line approach to sustainability emerged. Indeed, a lot of sustainability definitions deal with these three dimensions e.g. “sustainability is not about threat analysis, it is about system analysis specifically, how environment, economic and social systems interest to their mutual advantages and disadvantages at various space based scales of operations” (Transportation Research Board 1997).
Sustainability role shine in the compromise between conflicting set of goals (reducing use of resources, increasing economy growth, achieving social equity) and several demands desirable from different elements of society, each according to his interest.
That defined sustainability role as the solution that meets challenges through integrated solutions rather than through fragmented approaches that meet one of those goals at the expense of the others. And it takes a long-term perspective―one that’s focused on the present and future―Institute for Sustainable Communities (ISC 1997)  .
The terms of sustainability and sustainable development emerge in various events, conferences and can be traced by several publications   .
However, the majority of work covering sustainability unavoidably point out at 1987 report for the United Nations World Commission on Environment and Development (the Brundtland Commission report)  . As examined by Jones et al.  , the Brundtland definition of sustainability is so popular due to the fact that it presents a broad agenda that even entities with conflicting interests or goals can agree upon. On the other hand, the Brundtland work has been criticized for focusing too much on human development and needs. Other approaches involve the Natural Step Approach framework, suggested by Robert  , and the concept of Natural Capitalism  , which sees the natural environment as the primary interest of sustainability.
1.2. Sustainability and Sustainable Transportation
Transportation systems have been described as the “lifeblood” of cities in recognition this critical role (V. R, 1999)  . Transport is linked to all aspects of our life. Our natural environment, economic, prosperity, and social wellbeing all depend on transportation systems that are safe, clean, efficient, and equitable. Conversely, the social, economic, and environmental impacts of transport are framed as critical issues that can challenge the sustainability of cities and regions.
A sustainable development cannot be achieved without a sustainable transport system. The sustainability within transportation is described by a series of characteristics according to the center of sustainability (CST 2005)  :
1) Allows the basic access needs of individuals and societies to be met safely and in a manner consistent with human and ecosystem health, and with equity within and between generations.
2) Is affordable, safe, operates efficiently, offers choice of transport mode, and supports a vibrant economy.
3) Limits emissions and waste, minimizes consumption of non-renewable resources, limits consumption of renewable resources to the sustainable yield level, reuses and recycles its components, and minimizes the use of land and the production of noise.
Another description of sustainable transportation was addressed by Banister who portrayed sustainable mobility paradigm involving four primary elements (Technology, demand management, integrated land use and transportation planning, and public awareness and acceptance)  .
A review of the literature marks certain common boundaries among various sustainable transportation initiatives and definitions; these widely involve concerns about environmental impacts, assurance on safety, affordability, and accessibility of transportation services, etc.
1.3. Frameworks, Indicators, and Performance Measures for Sustainability in Transportation
Performance measures (or indicators) are measurable criteria aiming to monitor progress toward achieving targets. A considerable amount of knowledge exist at the international scientific community concerning individual factors (indicators) affecting sustainability (Munier, 2005)  . In addition, Rand ET al 2004, Litman 2009 have set a list of several indicators covering the three sustainability aspects (Economical-Social-Environmental)  . That led the majority of researches to follow a convenient way to measure the level of sustainability by using indicators (ECTS 2001, Hakkinen 2007)   .
When introducing an integral methodology of sustainability performance measurement to be utilized by a transportation agency, it is helpful to observe how performance indicators are mixed into frameworks and applied.
The commonly-usable performance measurement process scan be characterized as having this procedure  : 1) define objectives; 2) determine targets; 3) measure performance; 4) evaluate performance versus targets; and 5) assess and review process.
From a transportation viewpoint, the authors debate the requirements of robust sustainability frameworks involving flexibility, understanding of trade-offs, maintaining linkages with agency goals and objectives, addressing needs of all stakeholders.
Internationally, to convince any of transport agencies into modifying their system to another more sustainable, major studies and researches have been executed to find scientific tools and methods to measure sustainability performance. Researchers seeks suitable ways to transfer this knowledge to transport managers, decisions makers and municipalities to start correcting their mistaken vision of profit which should not only based on increasing operating economic growth but should be seen from the sustainability triple bottom line perspective.
While literature provides several examples of sustainability indicators and guidance of their selection, a very scarce documented examples move through the framework application process involving defining sustainability and applying indicators measurement.
In this paper, we present a composite indicator framework to monitor the sustainability cost performance of Transport modes in Lebanon and similar countries cities. The framework follows the performance measurement process, applies indicators measurement and covers 6 indicators of the triple bottom line of sustainability (economy, society, and environment): operating cost, energy consumption, noise exposure, congestion delay, emissions, and safety. The cost measurement is executed through an evaluation toolkit emanating from these indicators.
2. Lebanon Case
Lebanon similar to the majority of third world countries lack immensely regarding the field of transportation sustainability in comparison to the modern world as all of his land transport system is mostly based on private gasoline ICE cars (85% of land transport) and on petroleum resources. Even, its public transport system, characterized by poor management, consists of old diesel buses and vans in poor condition that doesn’t provide reliability or security for passengers’ trips. Road traffic accidents―the leading cause of death by injury and the 10th leading cause of all deaths globally―make up a surprisingly significant portion of the worldwide burden of ill-health. Car accidents reached 4205 in Lebanon in 2012. In addition, 25% of CO2 emissions come from the land transport sector which comes only second to electricity as the most national excessive energy consumers. That led Lebanon to be in the lead of countries that are still in the birth stage regarding the sustainability knowledge and the ways of applying its principles, especially in transport sector which is drowning in complete mess causing negative consequences on all vital partitions.
In order to challenge the consequences resulting from the huge unorganized investment in Lebanon transport sector reliable on non renewable resources and on traditional expanding roads solution, new plans and programs need to be followed by government and organizations to adjust the negative direction of the index of the sustainability of transport system. The rectification process initiates by identification of the problem which is obviously the auto dependence ICE mobility in these countries. The second step will be to offer solutions, propose logical plans and strategies to the faced problem. Utterly, the principal work is to formulate valid methods to monitor the development of transport sustainability in the city regarding the suggested solutions.
At Lebanon level, little has been done by both researchers and government in the transportation sustainability field, to provide methods and techniques to measure the cost performance of unsustainable mobility systems within its cities. The studies previously done care most for environmental cases regarding transport sector as Greenhouse gases emitted (National Greenhouse Gas Inventory Report For Transport Sector 2015)  and influence on climate change (Lebanon TNA report for Climate Change 2012)  . Some researches investigate the Lebanese passenger transport energy consumption (Electris et al., 2009)  , while others focus on driving conditions and market share to estimate operating costs of the Lebanese system (Mansour, C. and Zgheib, E., 2012)  . But, the topic considering the annual economic loss resulting from Lebanon strictly unsustainable mobility system is rarely considered a major case for government and effectives parties.
3. Sustainability Cost Assessment Framework
The need to measure sustainability appeared at the same time as the concept of sustainability. Figure 1 illustrates the flowchart of the proposed sustainable cost assessment framework using a cost evaluation toolkit that covers indicators of the triple bottom line of sustainability.
The framework presented is valid for Lebanon and similar nations facing the same identified problem which is the total ICE auto dependence mobility and working towards achieving the sustainability in transportation sector.
The framework include principal relevant aspects important to the transport system sustainability, there are 6 aspects grouped according to category:
1) Economic dimension (operating cost, congestion cost)
2) Environmental dimension (noise exposure, energy consumption, émissions)
3) Social dimension (safety and crashes cost)
3.1. Cost Evaluation Toolkit
Figure 2 illustrates the selected sustainability indicators used in toolkit formulation and their respective components.
For all transport modes, a distance based model may be applied to calculate total operating cost, noise exposure, energy consumption and CO2 emissions costs in a desired duration of time.
A distance based model is formed by multiplying (total distance travelled by mode during a time t) × specific factor depending on vehicle and fuel type.
This toolkit will focus on the distance based model to calculate the outcome of the 4 mentioned indicators and manual formulated approaches for others.
Figure 1. Framework for sustainable transportation planning process in Lebanon.
Figure 2. Cost evaluation toolkit sustainable indicators components.
3.1.1. Distance Base Model
The degree of difficulty in calculating transportation emissions depends largely on which gases are included in the analysis. In most cases, CO2 emissions are relatively straightforward to estimate, since they are primarily dependent on only two factors: the type and quantity of fuel burned. N2O and CH4 emissions, on the other hand, depend largely on the emissions control equipment used (e.g., type of catalytic converter). Since N2O and CH4 emissions comprise a relatively small proportion of overall transportation emissions, only CO2 emissions estimates are included in this tool.
(Emission/distance unit), (energy consumption/distance unit) and (operating cost/distance unit) factors are used to calculate the outcome by multiplying it by aggregated distance data. The emission factor represents the fraction of carbon component in fuel delivered into atmosphere after the chemical oxidation take place.
Distance data units can be in several forms: vehicle kilometers travelled, freight ton-kilometers, passenger-kilometers traveled. Distance data and (emission, operating cost, noise cost and energy consumption) factors varies according to vehicular engine category. So the following equation outlines the approach to calculating the cost based on distance data traveled:
CO2 Emissions/mode = (1)
Energy consumption/mode = (3)
Di: Distance data travelled according to vehicular engine of type i
Ei: (Emission/distance unit) factor for vehicular engine of type i
Oci: (Operating cost/distance unit) factor for vehicular engine of type i
Eci: (Energy/distance unit) factor for vehicular engine of type i in MJ
∑Dv = total distance traveled by vehicle mode in VKM.
Nc = marginal noise costs per distance depending on mode type.
N = number of vehicle of transport mode.
The SC-CO2 is a measure, in dollars, of the long-term damage done by a ton of carbon dioxide (CO2) emissions in a given year. The value of 41 USD/t CO2 is adopted (Van Den Bergh and Botzen, 2014)  .
1 MJ = 0.0076 gal & 1 US gallon price = 3$ → 1 MJ cost = 0.0228$.
3.1.2. Time Framed Model for Safety Indicator Cost
The outcome will be based on collecting transport accidents data for a near previous time n (years), and assigning it into three components: fatalities, injuries and property damages.
The following equations outlines the model to calculating the average of each of these components based on time framed approach
Mean of fatalities per mode MF = (7)
Mean of injuries per mode MI = (8)
Ra = average rate of accident/1 vehicular unit of mode =
Fa = average rate of fatality/1 accident of mode = .
Ia = average injuries rate/1 accident of mode = .
F = number of mode fatalities per year
I = number of mode injuries per year.
A = number of mode accidents per year
N = number of mode vehicular units for year i
I = year number 1, 2, ∙∙∙, n,
3.1.3. Congestion Cost Approach
Congestion cost approach interested basically in roadway transport wheeled mode because rail mode transport by principle uses separated exclusive lanes for travel and usually should not face congestion issues.
Calculating congestion cost is calculating the cost of 2 main consequences delivered from it:
− Congested roadway vehicle running cost
− Loss due to value of time for in vehicle travel (stopping + slow movement)
In order to measure the roadway vehicle running cost, a road congestion status need to be assigned to one of the categories according to various parameters as shown in Table 1  .
Secondly, based on the roadway congestion category, an estimated value can be deducted for the loss per vehicle mile from Table 2 according to specified parameters as nature of road and vehicle type  .
The following equation is used to find the first cost component:
Table 1. Roadway congestion categories.
Source: Transportation Cost and Benefit Analysis II?Congestion Costs Victoria Transport Policy Institute.
Table 2. Estimated highway congestion costs (cents per vehicle mile).
Source: Transportation Cost and Benefit Analysis II?Congestion Costs Victoria Transport Policy Institute.
Di: distance data traveled by vehicle type i
Lci: congestion loss/distance factor by vehicle type i
Value of time proposed methodology is divided into 2 divisions based on trip purpose model (work or non work trip travel time delays) and both rely on the average hourly income salary of households.
Average hourly income salary H (11)
For the work purpose trip, the following equation can be used:
Delay ∆T = (Tf − Ti) = actual trip time-free flow trip time in hours
L1: Loss due to delays in travel time per capita trip.
H: Average hourly income salary
The Toolkit for the Economic Evaluation of transport projects(World bank 2005) suggest a coefficient of 0.33H to mirror additional costs related with work trip of an employee and income rate such as overheads, paid holidays, sickness, etc  .
For the non work purpose trip
There is no proper basis matching between the value of the non-working hour and salary rate theoretically, however its value is associated to income, though any salary change should be expressed in a variation in the value of non-working hour. Due to lack in local data, its amount is taken as 30 percent of the hourly salary rate as recommended by the World Bank Toolkit  .
Delay ∆T = (Tf − Ti) = actual trip time-free flow trip time in hours.
L2: Loss due to delays in travel time per capita trip.
H: Average hourly income salary.
The total value of time loss can be estimated using the equation:
∑Zi represent the passengers facing delay during work trip
∑Yi represent the passengers facing delay during a non work trip.
Congestion cost CC = Rc + T in $ (15)
To summarize all costs:
Total cost per mode = CC + OC + ECC + ENC + SSC + SCC (16)
4. Case Study: The Cost Assessment of ICE Passenger Cars Mode in GBA and Their LRT (Light Rail Transit) Alternative Using the Framework and the Proposed Evaluation Toolkit
The Greater Beirut Area (GBA) location addressed in Figure 3 extends from Nahr-el-Damour south to Nahr-el-Kalb north, enclosing more than 40% of the population estimated to be 5,102,830 in 2011, including foreign workers and Palestinian refugees (Central Administration Statistics CAS 2013).
Figure 3. Location of the greater Beirut area.
Transport systems in GBA can be described as strictly unsustainable. Traffic conditions are mostly congested as shown in Figure 4 especially at the entrances to Beirut, with daily passenger trips amounting to around 2.8 million daily automobile passenger trips in 2007, which are expected to increase to 5 million in 2015 (Ministry of Public Works and Transport, 2007). The average speeds during the day along the major axes in the GBA range between 15 and 30 km/hr, dropping to 10 km/hr and less in the commercial districts within the city at peak times (MOE, 2012 i.e. ministry of environment)  .
4.1. Future LRT Plan & Assessment
Installing a light railway system for Lebanon aims to account for multiple objectives:
− Access needs of individuals
− Environmental protection
− Economic vitality
Shifting to rail transport will be stepping into an environmentally-friendly mass transport option. The particularity of rail is its high transport capacity per trip compared to other modes of transportation (Golinska et al., 2012). In this case, the energy requirements per passenger-trip dramatically decrease and savings are not only in terms of energy and emissions, but in space as well since one train transporting two hundred passengers would take less space than two hundred cars, for example.
As a first step, a light rail line must be implemented from Jounieh to Beirut airport. Such line is considered as a congestion relief for the whole area of Beirut, especially on Dawra highways and jell el dib which are a nightmare for Lebanese citizens since the activity of vehicles along this line is considered very high active and an implementation of such system will find its success easily.
From Jounieh to Beirut is a highly active area, daily congestion is always expected along this line and the peak hour spread all day. The light rail line passes
Figure 4. GBA congestion status (Council for development and reconstruction lebanon 2015).
inside Beirut through its central district which adds a sense of civilization and modernization to the city and help moving inside the city easier. The last destination would be Beirut airport and this is essential because of the précising dates of travelers and the importance of this facility (Figure 5).
Reasons for choosing this location
1) Highest population density in Lebanon is concentrated along the chosen direction of light rail line. This will form a good launching for the project.
2) Most businesses and resorts are distributed along the path and this will improve mobility and economy as well.
3) Most congested roads and highways are along this area
4) High percentages of accidents that takes place on daily basis due to poor maintenance of roads and absence of safety precautions.
Light Rail Transit Demand and Analysis
Trip end model
Greater Beirut Area is located in the central coastal area of Lebanon. It covers an area of 233.2 km2, which corresponds to 2.2% of the Lebanese territory. The average population density in GBA is approximately 6200 inhabitants km2  and the average number of cars per household was 1.8  . To estimate the expected percentage of transit mode split, urban travel factor UTF is calculated using trip end model curve as indicated in Figure 6:
=>The expected percentage of (Lrt) transit mode split is approximately = 43%.
4.2. Data Collection
o The 2012 Lebanese fleet database shows a total of 1.58 million registered vehicles, of which 85% consisting of ICE passenger car = 1,343,000 PC (Lebanon Lebanon Technology Needs Assessment report 2012)  .
Figure 5. Proposed light rail line layout (27 km).
Figure 6. Transit mode split versus urban travel factor.
o The final number of passenger cars in circulation was 1,108,328 vehicles (MOIM, 2012 i.e. Lebanese Ministry of Interior and Municipalities).
Due to lack of recent studies by Lebanon ministries, to estimate the 2016 fleet size, the 2012 passenger fleet is increased because the sector of car sale has experienced a positive trend in growth over the period 2012-2016 by growing at an average annual rate of 3.3% (Bank med analysis of Lebanon transport sector)  as shown in Table 3.
The 2016 fleet size estimated is close to the 1,232,229 PC used in the article “Influence of boundary conditions and anthropogenic emission inventories on simulated O3 and PM 2.5 concentrations over Lebanon (2016)”  .
o 40% of Lebanese passenger cars circulate within Greater Beirut Area (MOIM, 2016 i.e. Lebanese Ministry of Interior and Municipalities) => the number of passenger cars circulating in GBA for 2016 is taken 505,051.
o 75% of GBA trips are work based trip and 25% are non work based trip, and the average mobility rate = 3 trip/passenger/day (Team international 2013)  .
Passenger cars PC are distributed according to engine size as shown in Table 4.
o The available data on the average annual traveled distance per vehicle class is considered 15,000 km (MOE/UNDP/GEF, 2015)  .
Operating cost of driving PC
The operating cost will be divided on its 3 principal costs depending on distance:
Table 3. New car sales (2010-2016).
Table 4. CO2 emissions and fuel consumption regarding distribution of passenger car fleet according to engine size (MOE 2015).
• Fuel consumption (L/100 km) & 1 L = 0.9$.
• Engine oil change recommended each 3000 km of 32 - 46$ according to size (car dealership surveys in Lebanon).
• Average (maintenance-tire-tax-insurance-salvage value) mixed cost of 21 cents/veh.km (Mansour, C. and Zgheib, E., (2012).
For different engine size, the overall operating cost is ranging between 29 and 31.7 cents per veh.km which is close to the value of 33 cents estimated by AAA organization study (Your Driving Costs’ study, ORLANDO, 2016).
o Occupancy rate of a vehicle is taken as 1.2 passengers per car (MOE/UNDP/GEF, 2015).
o The road transport sector in Lebanon considered having high passenger transport energy intensity in 2007, estimated at 2.45 - 3.08 MJ/passenger-kilometer (Electris et al., 2009).
o The accidents data used are taken from kunhadi organization and Beirut traffic center database; records shown in Table 5 & Table 6 are extracted from Lebanese ISF (internal security forces) reports and are used to calculate the values of Ra, Fa and Ia.
Based on Accident data for the 2010 year (last study executed), we get approximately:
Cost of accident fatality = 360,145$.
Cost of accident injury = 46,990$.
Cost of property damage/accident = 27,567$/accident.
Table 5. The stats of accidents, injuries, and fatalities.
Table 6. Accidental costs (USD Million).
Source: Traffic control center for Lebanon.
• Ra = average rate of accident/1 PC = 0.00389
• Fa = average rate of fatality/1 accident of PC = 0.128373
• Ia = average injuries rate/1 accident of PC = 1.348031
4.3. Calculation and Discussion
Passengers trips/day = Number of vehicles/day × occupancy × mobility rate (18)
To monitor the annual cost performance of LRT, a comparison should be done between LRT trips and same trips number by PC using the framework assessment toolkit.
From an operational perspective, a service with operating period 18 hours, 100 seconds headway and LRT vehicles with a capacity of 400 persons would be adequate for accommodating ridership.
Daily Demand = 260,606 passengers.
− Productivity = operate period/headway x length = 17,496 veh.km/day.
− Annual LRT productivity = 6,386,040 veh.km.
Transporting 260,606 passengers making an annually average of 15,000 km travel is equal:
Pkm = Passengers number × average annual travelled distance
= 260,606 × 15,000 = 3,909,090,000 km = 2,443,181,250 miles.
1) Environmental Noise Cost
ENC for PC mode = 34,400,045$ & ENC for LRT mode = 2,095,132 $
Marginal noise cost is estimated at 8.8 ?or a passenger car and 273.4 ?for passenger train per 1000 vkm (Values from CE Delft et al. 2011)  .
2) For safety indicator cost
LRT excels in safety and has a record clearly superior to that of automobile travel. Among transit modes, LRT is the safest. Data submitted to FTA document that LRT’s safety record is 0.7 accidents per billion miles of vehicle travel (BTS, National Transportation Statistics, 2003)  .
The accident rate of LRT is approximately touching 0 considering annual mileage of millions instead of billions miles →A = 0 → Ra = 0 → MF = MI = 0 → SCC = 0$.
3) Congestion delay indicator cost
Average hourly income salary H =
In Lebanon, the Labor Law indicates a maximum of 48 working hours/week. As well, the working hours/week in private sector (48 hours) are not the same for the public sector (near 40 hours).
In this study, an average of 44 working hours is adopted considering the average of 40 and 48 weekly working hours. Delays at the majority of intersections in GBA are ranging from 5 to 30 minutes per trip (MOE/URC/GEF, 2012).
For the purpose of this research, it is assumed that the delay per passenger trip is 17 min based on the average minimum 5 min of delay per passenger trip and the maximum 30 min of delay per passenger trip.
Average Monthly Salary (After Tax) = 1032 $  .
− Value of working hour = 1.33H = 7.19$
− Value of non working hour = 0.3H = 1.62$
➢ For LRT mode:
The separate exclusive lanes and priority right of way for LRT mode allow it to avoid this problem → CC for LRT=0$.
For PC mode
The average speeds during the day along the major axes in the GBA range between 15 and 30 km/hr so Beirut falls under the extreme roadway congestion condition.
A high congestion cost for all highways to be taken as 0.1317$/vehicle mile = 0.0823$/vehicle km (Litman et al. 2011)  .
=>CC = Rc + T = 620,076,474$
4) Operating cost indicator for PC mode
Take into consideration the same distribution of engine size for the volume of 217,172 PC:
5) Social Carbon cost SCC for PC mode
6) Energy consumption cost ECC for PC mode
As for LRT mode, distance data units are passengers miles ,and their respective factors are adopted as follow:
➢ Average Operating Cost ($/pass. mile) = 0.69 
➢ GHG emission (lbCO2/pass. mile) = 0.41 
➢ Energy consumption (MJ/pass. mile) = 0.56 
• A summary of the above calcuation is given by the table (Table 7) and the figure below (Figure 7) showingthe performance cost measures of the 2 analysed modes, concluding with a net saving coming out of adopting the LRT model instead of current ICE mode of transport.
Figure 7. LRT, ICE and ICE to LRT modification cost analysis results.
Table 7. Case study summary.
5. Conclusion and Future Work
Starting from sustainable indicators (social, economical and environmental) to formulate a cost toolkit to measure the performance of ICE transport system and evaluate their modification is the main interest that this paper tries to deliver. That will help ICE auto dependent countries suffering from sustainability lateness to begin considering this vital issue and propose solutions that can be assessed in front and achieve the maximum benefit.
The case study assessed and the LRT solution in GBA show clearly an environmental amelioration with a reduction of 4,651,817,100 MJ of energy and 181,936 ton of CO2 emissions in addition to a saving of 255,643,035$ annually.
This notable improvement come only through a modification involving less than 20% of the total volume of gasoline car running in Lebanon and considering only GBA as the main congestion zone. Altering transport mode from private gasoline car to mass transit in Lebanon who is paying approximately a bill of millions dollars annually due to his unsustainable mode of transport proved to be efficient and beneficial to the country in all sustainable dimensions.
From framework methodology viewpoint, there remain some interesting topics to explore as further environmental indicators involving other GHG emissions (NO and NOX emissions) can be included. As well, more social indicators as equity and accessibility can be added, other economic values as employment and affordability could also take part in the performance measurement process. In fact, more researches covering a bigger set of indicators will make the framework more accurate and give a more comprehensive form.
The authors wish to thank Beirut Arab University Dean and civil engineering department for their support; they provided insight and expertise that greatly assisted the research, although they may not agree with all of the interpretations/conclusions of this paper. We also thank Traffic center of Beirut, the ministry of environment and the ministry of interior and municipalities who provided us with the data required to complete our work.
Conflicts of Interest
The authors declare no conflicts of interest regarding the publication of this paper.
 Faour, G. and Mhawej, M. (2014) Mapping Urban Transitions in the Greater Beirut Area Using Different Space Platforms. National Center for Remote Sensing, National Council for Scientific Research (CNRS), Beirut.
 Chalak, A., Al-Naghi, H., Irani, A. and Abou-Zeid, M. (2015) Commuters’ Behavior towards Upgraded Bus Services in Greater Beirut: Implications for Greenhouse Gas Emissions, Social Welfare and Transport Policy. The Issam Fares Institute for Public Policy and International Affairs, American University of Beirut, Beirut.
 Korzhenevych, A. and Dehnen, N. (2014) Update of the Handbook on External Costs of Transport.