Urban Railway System Route Selection in Turkey Based on Socio-Economic and Technical Criteria Using the Analytical Hierarchy Process

Abstract

With the increased importance of energy all over the world, new energy sources are needed to meet the energy demand. However, the solution to energy demand is not only finding new energy resources but also using currently consumed energy efficiently. Accordingly, decreasing the energy consumption in urban electric railway transport has become an important issue. Investment in urban railway transport is increasing continually because it is more economical, safe, and sustainable than other modes in local transportation. For new investments related to urban railway transport, decision-makers generally make an assessment that will include economic, demographic, and environmental impacts. In this study, a comprehensive top-down decision-making system for urban railway route selection was engineered by taking into account socio-economic criteria as well as technical criteria. In the context of surveys conducted with experts in the choice of railway system routing, five socio-economic criteria (passenger potential, construction cost, integration-access, zoning-ownership, travel time) and five technical criteria (vehicle energy consumption, regenerative energy, traffic operations, power supply system, efficient driving techniques) were determined. In the analysis, the analytical hierarchy process (AHP) method, one of the multi-criteria decision-making methods, was used to determine the best route among alternative routes based on all criteria.

Keywords

public transportation light rail transit catenary Analytical Hierarchy Process urban railway system route urban railway energy consumption

To meet the growing demand for energy, it is of great importance to use the consumed energy efficiently, just as much as the discovery of new energy resources. With the increase in the use of electric rail systems, which are an important part of city life all over the world, the energy efficiency and the reduction of energy consumption in these systems has become an important issue. In electric urban railway systems, energy consumption can be reduced with designs that can be tested in a feasibility study, such as optimum selection of the urban railway system routes and vehicles.

Determination of urban railway system routes depends on many criteria, such as economic, political, technical, environmental, and social issues. Criteria have superiorities and inferiorities when compared with one another depending on the alternative urban railway system routes to be decided. Considering them separately cannot provide a good solution. When engineering a new rail public transport system in light of these criteria, decisions need to be made that strongly favor sustainability. For decision-makers in Turkey, the most appropriate approach in the evaluation of rail system projects is to consider technical criteria as well as demographic and economic criteria.

With comprehensive multidisciplinary approaches, all criteria must integrate the best alternatives into the selection process, with appropriate weighting. For this reason, the analytical hierarchy process (AHP) has been used to obtain a concrete result by using many possible criteria. AHP generates a powerful decision-making calculation that is commonly used in multi-criteria decision-making.

AHP is widely used in the transportation sector, as found in the literature. Banai ( 1 ) made a general analysis of public transport decision-making, taking into account social criteria such as mobility, cost, and traffic, and identified the best alternative from three different light rail transit routes according to the criteria. Sustainable transport infrastructure planning focuses on the long-term economic, social, and environmental impacts with regard to sustainability and business strategies. In this context, the most suitable route was selected by taking the sustainability and development criteria as a priority for the Rail Baltica project ( 2 ). Energy consumption factors in urban railways have been analyzed with AHP ( 3 ). A monorail project planned for the city of Istanbul was analyzed by including engineering criteria, such as extensible, along with economic, social, and environmental criteria, such as travel time, and the most suitable project was decided using AHP ( 4 ). Unlike other AHP studies, which criteria are important in future transportation quality was determined without the level of alternatives to sustainable public transport development ( 5 ). In addition to the route determination studies, the selection of high-speed train station locations (6, 7) and the selection of transportation vehicles by municipalities responsible for public transportation ( 8 ) have been examined with multi-criteria decision-making methods. Orman et al. ( 9 ) studied public transportation scenarios (the current situation development, the highway development, the rail system, the mixed system), and AHP was used in the selection of the scenarios.

Most of the studies in the literature have been based only on social, economic, and environmental criteria. In this study, socio-economic criteria such as integration-access, zoning-ownership, and technical criteria such as energy consumption, power supply, economic driving, regenerative energy, and traffic operation have been analyzed.

In this paper, first the energy consumption points in an urban railway system and their impacts on the total consumption are examined. Energy consumption points in the urban railway systems could be considered in two main classes: as permanent loads (station, etc.) and dynamic loads (vehicle energy consumption). Emphasis is laid on the dynamic loads since they have a great impact on the costs of the operation in urban railway systems ( 10 ).

Criteria Affecting Energy Consumption in the Urban Railway System

The energy consumption of urban railway systems can be separated into two main classes. These are dynamic load, such as the traction of urban railway system vehicles, and permanent load, such as the energy consumption of passenger stations, auxiliary systems, administrative buildings, and maintenance facilities. Factors affecting the traction consumption of the vehicles are as follows:

Track profile: Factors such as the geometric structure of the track and the vertical and horizontal slopes affect vehicle energy consumption. The slope of the track is directly proportional to the energy consumption. Slope is one of the most important factors affecting energy consumption ( 10 ).

Speed limits: Urban rail transportation systems, especially, have numerous stations, curves, roundabouts or pedestrian crossings along the route. Therefore, frequent stops and starts and deceleration–acceleration situations from speed limits will increase energy consumption. Determining the optimum speed profile will lead to lower energy consumption ( 11 ).

Station locations and distances: It is necessary for the stations to be established in slope-free zones and at convenient distances by taking the passenger potential into consideration. The station spacing has an impact on energy consumption. If the station spacing is short, the vehicles will make more stop-and-go operations, which, in turn, will increase energy consumption, and the total travel time will also increase.

Vehicle characteristics: Vehicle consumption is dependent on systems such as the mechanical and electrical design of the vehicle, motor power and characteristics, vehicle load, vehicle inner consumption, and regenerative braking, and so forth. In the acceleration of the vehicle, until it reaches the operating speed, the vehicle mass and speed directly affect the vehicle’s energy consumption. Acceleration is necessary to produce kinetic energy, but too rapid acceleration causes excessive energy consumption because kinetic energy increases together with the square of the speed. Further, an increase by 10% in the train mass increases the vehicle energy consumption by the rate of 6%–8% (10, 12).

Power supply system: Energy consumption is dependent on the voltage level used (750 V DC, 1,500 V DC), the design of the power centers, the number of power centers, the structure of the cable, overhead line design, and power losses (3, 11).

Operation: The number of trips, the headway, and the number of the rolling stocks belonging to the rail system operation have an impact on the energy consumption in urban railway systems.

Efficient driving techniques: The aim of efficient driving techniques is to reduce unnecessary fuel consumption caused by driver behavior and to create awareness of efficient driving. The following examples of efficient driving techniques can be given: (a) Compliance with acceleration, cruise, coasting, and braking rules according to the determined speed profile; (b) Optimum operation of vehicle auxiliary systems (such as air conditioning) without reducing the comfort of passengers, and so forth. Energy could be saved by using efficient driving techniques on urban railway systems.

Because factors such as slopes, speed limits, station locations, and route structure affect efficient driving, determination of the urban railway system route by taking these factors into consideration has importance for energy saving in the long term. The acceleration movements, depending on the number of vehicles stopping, have a great impact on the energy demand. The train is coasting after achieving maximum speed and it continues in this manner until it reaches the next station (10, 12). Considering efficient driving techniques, energy consumption will be reduced with the changes to be made in the existing urban railway lines as well as in the project phase.

The impact of the parameters containing actual slope, actual vehicle weight, and efficient driving techniques on the existing route is shown in Tables 1 to 4, presenting the results of simulations performed with SIMUX software (10, 13, 14). The analyses were conducted on a round-trip basis.

Table 1.

Impact of Slope on Energy Consumption

Slope (%)	Energy consumption (kWh/km)
−5.0	0.62
−1.0	3.35
0.0	4.46
1.0	5.78
5.0	11.55

Table 2.

Impact of Vehicle Weight on Energy Consumption

Vehicle weight (kg)	Energy consumption (kWh/km)
31,900	2.92
38,900	3.5
45,900	4.1
49,400	4.39

Table 3.

Impact of Efficient Driving Techniques on Energy Consumption (Without Slope)

Slope (%)	Normal driving		Efficient driving		Energy gain (kWh)
Slope (%)	Consumption (kWh)	Regenerative (kWh)	Consumption (kWh)	Regenerative (kWh)	Energy gain (kWh)
0.09	4.71	2.56	2.91	1.3	25.1
0.09	4.91	2.52	2.69	0.96	27.6
0.11	6.51	3.73	2.81	1.07	37.4
0.11	5.26	3.27	2.31	1.16	42.2

Table 4.

Impact of Efficient Driving Techniques on Energy Consumption (With Slope)

Slope (%)	Normal driving		Efficient driving		Energy gain (kWh)
Slope (%)	Consumption (kWh)	Regenerative (kWh)	Consumption (kWh)	Regenerative (kWh)	Energy gain (kWh)
2.40	8.6	1.66	7.28	0.71	5.3
−2.40	2.81	4.51	1.18	3.34	27.1
3.50	11.55	2.16	9.63	0.78	5.8
−3.50	4.69	7.75	0.78	5.01	38.2

AHP

AHP is used in reaching a decision while making a selection or sequencing among many alternatives in certain and uncertain situations in which there are many decision-makers, many criteria, and many purposes. One of the most important reasons why AHP, developed by Saaty in the late 1970s, is used by decision-makers is that it takes subjective criteria into consideration in multi-criteria decisions. AHP is a technique assessing the alternatives in a hierarchical structure by combining the qualitative and quantitative factors in the assessment. The hierarchical structure is seen in Figure 1 and the steps of the method are given below (15, 16, 17).

Step 1: The first step is the stage of structuring the problem as a hierarchy. The decision-making problem is first revealed, and then the goal is determined. Secondly, the criteria for the alternatives related to the goal to be reached are defined. At the stage of defining the criteria, a list is usually prepared based on expert opinions. A hierarchical structure is created according to the determined criteria and alternatives. To achieve the goal, it is important to establish the true hierarchy and ensure the independence of components of the same level.

Step 2: The second step is the elicitation of pairwise comparison judgments and the construction of a set of pairwise comparison matrices. For making pairwise comparisons and matrices of the criteria and alternatives, the degree of importance of each of the criteria is determined. The scale developed by Saaty to determine the importance degrees, which consists of five main and four intermediate values, is given in Table 5 ( 15 ).

Table 5.

Importance Degree Scale

Importance degree	Conceptual correspondence	Explanation
1	Equal importance	Two activities contribute equally to the objective
3	Moderate importance of one over another	Experience and judgment strongly favor one activity over another
5	Essential or strong importance	Experience and judgment strongly favor one activity over another
7	Very strong importance	An activity is strongly favored and its dominance is demonstrated in practice
9	Extreme importance	The evidence favoring one activity over another is of the highest possible order of affirmation
2, 4, 6, 8	Intermediate values between the two adjacent judgments	When compromise is needed

While determining the importance degrees, decision-makers take into account the opinions of experts collected with a questionnaire or interview. When the decision-maker is a single person, it is easier to reveal their preferences, but when more than one decision-maker is involved, a consensus should be reached. It is convenient to use the geometric average of the preferences when there is more than one decision-maker.

Pairwise comparisons are conducted among the criteria by using the relative importance scale. The pairwise comparison matrix is then formed. If the number of criteria is n, the comparison matrix is the size of n × n.

The pairwise comparison matrix A with a_ij having the importance degree of the ith criterion and the jth criterion is as follows:

A = [\begin{matrix} a_{11} & a_{12} & . . . & a_{1 n} \\ a_{21} & a_{22} & . . . & a_{2 n} \\ . & . \\ . & . \\ . & . \\ a_{1 n} & a_{n 2} & . . . & a_{nn} \end{matrix}]

(1)

All the values of matrix A are positive and it is a matrix whose diagonal values are equal to 1. For example, the value of a₁₁ is equal to 1 when the first criterion is compared with the first criterion because of no superiority. Components of off-diagonal are determined by selecting the appropriate importance degree from Table 5. If the importance of the ith criterion with respect to the jth criterion is a_ij, then the importance of the jth criterion with respect to the ith criterion is 1/a_ij.

Step 3: In the third step, the weighted distribution of the criteria is determined. After matrix A has been formed as a result of the paired comparisons, it is necessary to normalize this matrix. For this purpose, the process of dividing every element into the column total is the method most widely used. Matrix C is obtained by dividing each element of matrix A into its column total.

c_{ij} = \frac{a_{ij}}{\sum_{i = 1}^{n} a_{ij}}

(2)

C = [\begin{matrix} c_{11} & c_{12} & . . . & c_{1 n} \\ c_{21} & c_{22} & . . . & c_{2 n} \\ . & . \\ . & . \\ . & . \\ c_{n 1} & c_{n 2} & . . . & c_{nn} \end{matrix}]

(3)

By using matrix C, percentage weight values of the criteria are calculated according to Equation 4 and the column vector of percentage weight values (W_i) is formed as in Equation 5.

w_{i} = \frac{\sum_{j = 1}^{n} c_{ij}}{n}

(4)

W_{i} = [\begin{matrix} w_{1} \\ w_{2} \\ . \\ . \\ . \\ w_{n} \end{matrix}]

(5)

Step 4: The fourth step is the step where consistency analysis is performed. In AHP, the success of the results will depend on the consistency of the decision-makers in the pairwise comparisons of the criteria. Therefore, consistency analysis is conducted to reveal the miscalculations of the decision-makers in the paired comparisons, providing an opportunity for error correction and avoiding exaggerative assessments. For the accuracy of the consistency analysis, the consistency rate (CR) in Equation 10 should be equal to or less than 0.10. If the consistency ratio is higher than 0.10, the decision-maker needs to revise the pairwise comparisons.

Random index (RI) is the corresponding index of consistency for judgments. The RI value is needed to evaluate the consistency and is defined for n-dimensional comparison matrices as given in Table 6 ( 18 ).

Table 6.

Random Index (RI) Values

n	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
RI	0	0	0.58	0.9	1.12	1.24	1.32	1.41	1.45	1.49	1.51	1.48	1.56	1.57	1.58

The expressions used to find CR are given in the following:

D = [\begin{matrix} a_{11} & a_{12} & . . . & a_{1 n} \\ a_{21} & a_{22} & . . . & a_{2 n} \\ . & . \\ . & . \\ . & . \\ a_{1 n} & a_{n 2} & . . . & a_{nn} \end{matrix}] [\begin{matrix} w_{1} \\ w_{2} \\ . \\ . \\ . \\ w_{n} \end{matrix}]

(6)

E_{i} = \frac{D_{i}}{W_{i}}

(7)

λ = \frac{\sum_{i = 1}^{n} E_{i}}{n}

(8)

CI = \frac{λ - n}{n - 1}

(9)

CR = \frac{CI}{RI}

(10)

where

D = the matrix product of matrices A and W

E _i = the division of the reciprocity elements of D and W.

λ = the arithmetic mean of the E_i values.

CI stands for the consistency index; CR stands for CR; RI is the RI values shown in Table 6.

Step 5: In this step, all transactions for the criteria are also made for the alternatives. As a result of these processes, the weight percent matrix of alternatives is obtained.

V_{i} = [\begin{matrix} V_{1} \\ V_{2} \\ . \\ . \\ . \\ V_{n} \end{matrix}]

(11)

In Equation 12, the relative importance value of the alternatives Z_i is found with scalar multiplication of the percentage weight matrix of the criteria W and V according to the criteria.

Z_{i} = \sum_{i}^{n} V_{i} W_{i}

(12)

After finding the relative importance values of the alternatives Z_i, the last stage is the decision phase. Within the framework of the criteria determined to achieve the goal, it is decided to select the alternative with the highest relative importance value Z_i ( 19 ).

Methodology

For a real-world application, alternative route projects prepared by Kayseri Transportation Corporation for the city of Kayseri in Talas district, Anayurt region, Turkey, have been examined. This study is a real practical field application and includes the selection of the routes designed by the Kayseri municipality with the AHP method.

The steps followed in the application of AHP to railway route selection are explained as follows:

All criteria affecting the urban railway route selection have been determined entirely by experts (general manager, department managers, engineers who work for Kayseri Transportation Company, and private sector representatives) via a questionnaire applied by interviewing. In this survey, the criteria (soil structure, proximity to commercial centers, use of energy storage systems, emissions reduction, etc.) specified by some decision-makers but not accepted as the majority have been ignored.

The criteria put forward by the experts were divided into two categories: socio-economic and technical.

All the analysis stages were made in light of the data collected officially and obtained through simulations. Real data research related to the determined criteria is made and reported to the decision-makers.

With the collaboration of experts in the context of their experiences, intuitions, knowledge, judgments, and thoughts, the pairwise comparisons of the criteria and the alternatives for each criterion are determined with reference to Table 5 and the importance degrees of the criteria and the alternatives are calculated.

Urban Railway System Route Selection With AHP

For the application, three different routes have been planned for Kayseri city, as shown in Figure 2, and the most convenient of the routes have been identified to be realized in the project ( 20 ).

In the urban railway system route selection for this study, 10 criteria classified as technical and socio-economic, given in Table 7, were determined by decision-makers. Technical criteria are those that have an impact on energy consumption. Socio-economic criteria are the factors that should be taken into consideration in the public transportation sector. Brief descriptions of these 10 criteria are below.

Table 7.

Pairwise Comparison Matrix for the Criteria

Criterion	PP	VEC	CC	TP	TO	IA	ZO	PSS	RE	ED
PP	1.000	2.000	3.000	7.000	6.000	8.000	8.000	5.000	6.000	4.000
VEC	0.500	1.000	2.000	7.000	7.000	8.000	8.000	3.000	5.000	3.000
CC	0.333	0.500	1.000	6.000	5.000	7.000	7.000	3.000	5.000	2.000
TP	0.143	0.143	0.167	1.000	0.200	3.000	2.000	0.167	0.250	0.143
TO	0.167	0.143	0.200	5.000	1.000	3.000	4.000	0.200	0.333	0.167
IA	0.125	0.125	0.143	0.333	0.333	1.000	2.000	0.200	0.333	0.143
ZO	0.125	0.125	0.143	0.500	0.250	0.500	1.000	0.167	0.333	0.143
PSS	0.200	0.333	0.333	6.000	5.000	5.000	6.000	1.000	5.000	0.333
RE	0.167	0.200	0.200	4.000	3.000	3.000	3.000	0.200	1.000	0.167
ED	0.250	0.333	0.500	7.000	6.000	7.000	7.000	3.000	6.000	1.000

Note: PP = passenger potential; VEC = vehicle energy consumption; CC = construction cost; TP = travel periods; TO = traffic operations; IA = integration-access; ZO = zoning-ownership; PSS = power supply system; RE = regenerative energy; ED = eco driving.

Passenger potential is one of the most important factors to be taken into consideration when determining routes since the urban railway system is a service sector and the customers of this sector are the passengers. A survey consisting of questions such as, “How often do you use the bus?”, “For what purpose do you usually use the bus?”, “How long is your travel time on the bus?” and so forth, was conducted by the Strategic and Institutional Development Directorate to reveal the public transportation preferences of residents of the region. In the province of Kayseri, where the study was conducted, transportation services are provided with a completely electronic ticket known as Kart38. With these electronic tickets, the number of passengers and the variety of passengers (student, elderly > 65 years, free, etc.) can be determined exactly. This survey and the electronic ticket database provided a basis for decision-makers in determining the importance degree of the passenger potential criterion.

The passenger potential in this case seems high because there are many houses and workplaces on Route 1. The student population is also high on this route and the connection of this track to the one going to the university increases the passenger potential. Routes 2 and 3 are directly connected to the university and are in a more advantageous situation for the student population. Besides, Routes 2 and 3 are shorter and faster for the passengers planning to go to downtown when compared with Route 1. The passenger potentials of these regions for bus transportation are directly related to their population size. The total population of the districts for Route 1, 2, and 3 is 90,752, 66,284, and 66,284, respectively. The population information of the region where the route will pass was taken from the Turkish Statistical Institute (TUIK) ( 21 ). The actual passenger transport data of the existing buses operating on the proposed railway routes were also obtained from the bus operation department of the Kayseri municipality as follows: the total annual number of bus passengers along Route 1, Route 2, and Route 3 is 6,580,980, 5,012,280, and 4,682,148, respectively. The importance degree (weight) of this criterion was decided by the decision-makers. In this criterion, it is assumed that all existing bus passengers on the same route will use the railway. Existing bus services will be canceled after the railway route has been decided and put into operation.

Vehicle energy consumption has the most important share in operating costs. When determining the route of the rail system, taking this criterion into consideration provides significant energy savings. The slope of the route has the highest effect on the energy consumption of the vehicles. The less the slope is, the less the energy consumption will be. The average slopes of Route 1, 2, and 3 are 2.87, 2.12, and 1.97, respectively.

Construction costs such as infrastructure, viaducts, and bridges increase costs. The decision-makers should examine the physical structures of the route. Considering the cost per kilometer of these structures according to the alternatives, the appropriate selection will reduce the construction costs. Route 1 which includes a 450 m long viaduct is 5.9 km, and it is a single-line operation. Route 2 is 8.3 km and Route 3 is 7.5 km. Both Route 2 and 3 have a double-line operation. The railway systems in question in the study cost 1.25 million euros per kilometer for a single-line operation. The construction cost information was obtained from the tender files of the railway lines under construction in similar regions. The estimated construction costs are 7.37 million euros, 10.30 million euros, and 9.37 million euros for Route 1, Route 2, and Route 3, respectively.

Travel time (or travel periods) is one of the parameters to be considered in transportation. Passengers always prefer more comfortable and shorter journeys. On Route 2 and Route 3, travel times between the university and the city center are shorter.

Traffic operation includes crossroads, pedestrian roads, signaling systems, and similar structures. Routes with heavy traffic affect energy consumption to a great extent. The deceleration–acceleration of rail system vehicles when they come to areas such as pedestrian roads and crossroads has a negative effect on the efficiency of energy consumption. The number of intersections, stations, sidewalks, and turns are 42, 39, and 23 for Route 1, Route 2, and Route 3, respectively.

Integration-access refers to the interaction between different modes of transport. When constructing a rail system line in a region, it is necessary to consider other types of transportation in that region. The integration of the newly built project with the existing bus, subway, suburban, and tramway lines in the region should be ensured. Thus, fast and comfortable transportation service is provided. The stations should be located in convenient places that passengers can easily access for the greatest convenience in transportation. In this case, all three alternative routes are integrated with bus stops. Considering integration with the existing urban railway system, Route 2 and Route 3 are more advantageous in reaching the university and the city center.

The zoning-ownership status of the routes where the rail system lines will be built should be examined. Politically, socially, and economically, the zoning-ownership status of the region has positive or negative effects on the institutions and the people of the region.

The power supply system includes the voltage level, transformer power, number of transformers, and similar components that play an important role in energy consumption. Increasing the voltage level of the system from 750 V DC to 1,500 V DC will both reduce energy losses and reduce the number of transformer centers. In this study, since the 750 V DC system will be projected as fixed for all routes, the number of transformer stations becomes important. According to the simulations made in 750 V DC systems, the transformer centers should be built at approximately 2 km intervals.

Regenerative energy is energy that is recovered during braking, that is, during the slowing down of a moving urban railway vehicle. When an urban railway vehicle is braking, if another vehicle that provides its energy from the same feeding point accelerates, the energy produced by braking is recovered. To use regenerative energy at the maximum level, the vehicles on the line should be as close to each other as possible and a route with low slope should be selected. It is more advantageous to use braking energy since Route 2 and Route 3 have double-line operations.

Economical driving methods in rail systems provide significant savings in energy consumption. The factors affecting economic driving between stations are slopes, curves, crossroads, and pedestrian roads. Fewer of these structures on the line makes economic driving more efficient.

Real Application

Determining the route of the rail system depends on many economic, political, technical, and demographic criteria. According to the alternative rail system routes to be decided, criteria have superiority or deficiency to each other. Therefore, AHP can be used to obtain powerful results in decision-making.

The pairwise comparison matrix table (i.e., matrix A in Equation 1) and importance degrees matrix table (i.e., matrix C in Equation 3) of the determined criteria used to select the urban railway system route have been formed as shown in Tables 7 and 8, respectively.

Table 8.

Importance Degrees Matrix for the Criteria

Criteria	PP	VEC	CC	TP	TO	IA	ZO	PSS	RE	ED
PP	0.332	0.408	0.39	0.16	0.178	0.176	0.167	0.314	0.205	0.361
VEC	0.166	0.204	0.26	0.16	0.207	0.176	0.167	0.188	0.171	0.27
CC	0.111	0.102	0.13	0.137	0.148	0.154	0.146	0.188	0.171	0.18
TP	0.047	0.029	0.022	0.023	0.006	0.066	0.042	0.01	0.009	0.013
TO	0.055	0.029	0.026	0.114	0.03	0.066	0.083	0.013	0.011	0.015
IA	0.042	0.025	0.019	0.008	0.01	0.022	0.042	0.013	0.011	0.013
ZO	0.042	0.025	0.019	0.011	0.007	0.011	0.021	0.01	0.011	0.013
PSS	0.066	0.068	0.043	0.137	0.148	0.11	0.125	0.063	0.171	0.03
RE	0.055	0.041	0.026	0.091	0.089	0.066	0.063	0.013	0.034	0.015
ED	0.083	0.068	0.065	0.16	0.178	0.154	0.146	0.188	0.205	0.09

Values (W, D, and E in Equations 5, 6, and 7, respectively) used to calculate the CR are shown in Table 9. Table 10 shows the fundamental parameters to be used in the consistency analysis.

Table 9.

Consistency Analysis According to the Importance Degrees of the Criteria

Criteria	n	W _i	D _i	E _i
Passenger potential	10	0.269	3.165	11.768
Vehicle energy consumption	10	0.197	2.356	11.965
Construction cost	10	0.147	1.78	12.135
Travel periods	10	0.027	0.269	10.101
Traffic operations	10	0.044	0.467	10.559
Integration-access	10	0.02	0.212	10.42
Zoning-ownership	10	0.017	0.182	10.667
Power supply system	10	0.096	1.141	11.866
Regenerative energy	10	0.049	0.556	11.291
Eco driving	10	0.134	1.638	12.254

Table 10.

Fundamental Parameters for Consistency Analysis

λ	11.303
Consistency index	0.145
Random index	1.49
Consistency rate	0.097

Note: λ = arithmetic mean of the E_i values.

The alternative routes have been assessed and scored according to the selection criteria determined with the AHP method. Firstly, the route selection was made taking into account the technical criteria that affect energy consumption, and then the selection was made by considering only the socio-economic criteria affecting energy consumption. Finally, the decision was made taking into account both technical and socio-economic criteria.

In Tables 11 to 16, the column values for Route 1, Route 2, and Route 3 show the V matrix expressed in Equation 11 and the relative importance values show the Z values expressed in Equation 12.

Table 11.

Percentage Weights of Alternatives According to Technical Criteria Only

Criteria	Alternatives			W_i
Criteria	Route 1	Route 2	Route 3	W_i
Vehicle energy consumption	0.1	0.3	0.6	0.439
Traffic operations	0.098	0.334	0.568	0.04
Power supply system	0.539	0.164	0.297	0.188
Regenerative energy	0.11	0.309	0.581	0.072
Eco driving	0.093	0.221	0.685	0.261
Relative importance value	0.182	0.256	0.563

Tables 11 and 12 represent the percentage weights of alternatives according to technical criteria only and the relative importance value for the selection based on technical criteria only, respectively. From these tables, Route 3 is the route that the decision-makers find most suitable.

Table 12.

Relative Importance Value for the Selection Based on Technical Criteria Only

Relative importance value (%)	Route
18.20	Route 1
25.60	Route 2
56.30	Route 3

Tables 13 and 14 represent the percentage weights of alternatives according to technical criteria only and relative importance value for the selection based on technical criteria only, respectively. From these tables, Route 1 is the route that the decision-makers find most suitable.

Table 13.

Percentage Weights of Alternatives According to Socio-Economic Criteria Only

Criteria	Alternatives			W _i
Criteria	Route 1	Route 2	Route 3	W _i
Passenger potential	0.665	0.231	0.104	0.507
Construction costs	0.539	0.164	0.297	0.306
Travel periods	0.089	0.324	0.587	0.088
Integration-access	0.548	0.211	0.241	0.056
Zoning-ownership	0.548	0.241	0.211	0.043
Relative importance value	0.564	0.218	0.218	NA

Note: NA = not applicable.

Table 14.

Relative Importance Value for the Selection Based on Socio-Economic Criteria Only

Relative importance value (%)	Route
56.40	Route 1
21.80	Route 2
21.80	Route 3

Tables 15 and 16 represent the percentage weights of alternatives according to both technical and socio-economic criteria, and relative importance value for the selection based on both technical and socio-economic criteria, respectively. From these tables, Route 3 is the route that the decision-makers find most suitable.

Table 15.

Percentage Weights of Alternatives According to Both Technical and Socio-Economic Criteria

Criteria	Alternatives			W _i
Criteria	Route 1	Route 2	Route 3	W _i
Passenger potential	0.665	0.231	0.104	0.269
Vehicle energy consumption	0.1	0.3	0.6	0.197
Construction costs	0.539	0.164	0.297	0.147
Travel periods	0.089	0.324	0.587	0.027
Traffic operation	0.098	0.334	0.568	0.044
Integration-access	0.548	0.211	0.241	0.02
Zoning-ownership	0.548	0.241	0.211	0.017
Power supply system	0.539	0.164	0.297	0.096
Regenerative energy	0.11	0.309	0.581	0.049
Eco driving	0.093	0.221	0.685	0.134
Relative importance value	0.375	0.238	0.388	NA

Note: NA = not applicable.

Table 16.

Relative Importance Value for the Selection Based on Both Technical and Socio-Economic Criteria

Relative importance value (%)	Route
37.50	Route 1
23.80	Route 2
38.80	Route 3

Conclusion and Suggestions

Urban railway transportation systems are among the main actors that ensure sustainability by developed and developing societies in today’s cities. After the urban railway system project becomes operational, it is important to ensure sustainability in relation to energy consumption. In planning for urban rail transportation systems, technical analysis such as estimation of the number of passengers, economic and financial suitability, operating and investment costs, weight of the vehicle, economical driving method, and regenerative energy should be done to provide a data base for the most appropriate route selection.

In this study, the most suitable route was decided with comprehensive comparisons by considering the technical criteria together with socio-economic criteria that affect energy consumption in rail transportation systems. The results coming out of the scenarios are as follows:

Route 1 was selected for the first rank with 56.4% based on social and economic criteria only.

Route 3 was selected for the first rank with 56.3% based on technical criteria only.

Route 3 was selected for the first rank with 38.6% based on both technical and social-economic criteria.

In the analysis, the most effective criterion among the technical criteria is determined to be energy consumption and the most effective criterion within the socio-economic criteria was passenger potential. As a result, Route 3 was determined to be the best alternative when the problem is considered rationally with AHP.

In rail transportation systems, investment costs are an expense that must be covered once, while operating costs continue throughout the life of the system. One of the most important items in operating costs is the cost of energy consumption. Therefore, to reduce the operating costs, energy consumption should be examined in detail under the title of technical criteria in the selection of railway routes.

All the factors that have an effect on energy consumption in railway transportation have been examined in detail and the criteria that are thought to be more dominant for decision-makers are used for the first time in this analysis with the title of technical criteria.

The most obvious potential contribution of the study of using the AHP method in railway selection is to ensure that different experts and shareholders who are competent in their fields are involved in the decision-making process and that it provides a joint evaluation of quantitative and qualitative criteria. The number of criteria and alternatives determined in practice can be increased or decreased by the decision-makers. Different values can be assigned to criteria or alternatives in pairwise comparisons. Thus, different results and different perspectives on the decision problem can be obtained. In that way, there may be more flexible and more rational inferences.

The methodology used for the first time in this study allows the decision-maker to choose a categorical perspective (social, economic, environmental, technical, etc.) or holistic (evaluating all criteria together) approach. It can be said that the technical applicability of the AHP decision-making tool in the selection of railway route is high because it uses real data in many criteria and allows the decision-maker to measure the degree of consistency of judgments (the consistency of the relationship between the pairwise comparison values of the criteria and alternatives).

In future studies, this proposed methodology can be applied effectively on the selection of urban railway routes with numerous decision-making criteria consisting of sub-criteria (environmental pollution, noise, public survey, passenger time saving, vehicle characteristics, etc.). In a more participatory-style decision-making process, professional chambers, non-governmental organizations, and academicians who have expertise in urban transportation can also be consulted to reach different opinions and ideas.

Footnotes

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: A. İnan; data collection: H. Kara; analysis and interpretation of results: A. İnan, H. Kara; draft manuscript preparation: A. İnan, H. Kara. All authors reviewed the results and approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Aslan İnan

Huzeyfe Kara

References

Banai

. Public Transportation Decision-Making: A Case Analysis of the Memphis Light Rail Corridor and Route Selection With Analytic Hierarchy Process. Journal of Public Transportation, Vol. 9, No. 2, 2006, p. 1. http://doi.org/10.5038/2375-0901.9.2.1.

Paul

Jayant

Leleur

Use of Analytic Hierarchy Process for Selection of Sustainable Transport Proposal. International Journal of Research in Mechanical Engineering & Technology, Vol. 5, No. 1, 2015, pp. 44-50.

Arıkan

Çam

Assessment of Factors of Energy Consumption in Railways With the AHP Method. The Online Journal of Science and Technology, Vol. 8, No. 1, 2018, pp. 21-28.

Hamurcu

Eren

Transportation Planning With Analytic Hierarchy Process and Goal Programming. International Advanced Researches and Engineering Journal, Vol. 2, No. 2, 2018, pp. 92-97.

Moslem

Duleba

Sustainable Urban Transport Development by Applying a Fuzzy-AHP Model: A Case Study From Mersin Turkey. MDPI Urban Science, Vol. 3, No. 2, 2019, p. 55. https://doi.org/10.3390/urbansci3020055.

Mohajeri

Amin

G. R.

Railway Station Site Selection Using Analytical Hierarchy Process and Data Envelopment Analysis. Computers & Industrial Engineering, Vol. 59, No. 1, 2010, pp. 107-114. https://doi.org/10.1016/j.cie.2010.03.006.

Dinç

Hamurcu

Eren

0-1 Target Programming Model Supported by Multi Criteria Decision Making in the Selection of Station Locations in Ankara-Sivas High Speed Train Line. Demiryolu Mühendisliği, Vol. 9, 2019, pp. 1-16.

Şengül

Ü.

Eren

Shiraz

S. E.

Public Transport Vehicle Selection of Municipalities With Fuzzy AHP. Erciyes University Journal of Faculty of Economics and Administrative Sciences, Vol. 40, 2013, pp. 143-165.

Orman

Düzkaya

Ulvi

Akdemir

Multi-Criteria Evaluation by Means of Using the Analytic Hierarchy Process in Transportation Master Plans: Scenario Selection in the Transportation Master Plan of Ankara. Journal of Science, Vol. 31, No. 2, 2018, pp. 381-397.

10.

Açıkbaş

Energy Efficient Driving Control in Multi Trains Multi Lines Rail Systems. Doctoral thesis. Istanbul Technical University, Turkey, 2008.

11.

Dalyan

Energy Optimization Oriented Studies in Railway Systems. Master’s thesis. Yıldız Technical University, Turkey, 2011.

12.

Keskin

. Railway Track Designing Effects on Energy Consumption. Master’s thesis. Istanbul Technical University, Turkey, 2013.

13.

RailSis. Rail System Simulator (SimuX). http://railsis.com/simux.asp?dil=en&. Accessed January 4, 2021.

14.

Sertsöz

Energy Efficiency of Rail Systems. Master’s thesis. Kocaeli University, Turkey, 2012.

15.

Saaty

T. L.

How to Make a Decision: The Analytic Hierarchy Process. European Journal of Operational Research, Vol. 48, No. 1, 1990, pp. 9-26. https://doi.org/10.1016/0377-2217(90)90057-I.

16.

Saaty

T. L.

Decision Making With the Analytic Hierarchy Process. International Journal of Services Sciences, Vol. 1, No. 1, 2008, pp. 83-98. https://doi.org/10.1504/IJSSCI.2008.017590.

17.

Göksu

Fuzzy Analytic Hierarchical Process and its Application in University Preference Ranking. Doctoral thesis. Süleyman Demirel University, Turkey, 2008.

18.

Saaty

T. L.

The Analytic Hierarchy Process. McGraw-Hill, New York, NY, 1980.

19.

Özden

Ü. H

. Primary School Selection With Analytical Hierarchy Method. Marmara University Journal of Economic and Administrative Sciences, Vol. 24, No. 1, 2008, pp. 299-320.

20.

Kara

Rail Route Selection With Energy Consumption Based AHP and Practice. Master's thesis. Yıldız Technical University, Turkey, 2016.

21.

TUIK. Turkish Statistical Institute. https://data.tuik.gov.tr/Kategori/GetKategori?p=Nufus-ve-Demografi-109. Accessed May 2, 2016.