Applied Soft Computing Selection of optimal ... - Semantic Scholar

Applied Soft Computing 21 (2014) 444–452

Contents lists available at ScienceDirect

Applied Soft Computing journal homepage: www.elsevier.com/locate/asoc

Selection of optimal electronic toll collection system for India: A subjective-fuzzy decision making approach Saurabh Vats a , Gaurav Vats b,∗ , Rahul Vaish b , Varun Kumar a a

Department of Electronics and Communication Engineering, Kurukshetra Institute of Technology and Management, Kurukshetra University, Kurukshetra, Haryana 136119, India School of Engineering, Indian Institute of Technology Mandi, Himachal Pradesh 175001, India

b

a r t i c l e

i n f o

Article history: Received 12 June 2013 Received in revised form 26 December 2013 Accepted 4 April 2014 Available online 18 April 2014 Keywords: Selection Electronic toll collection India Fuzzy MADM

a b s t r a c t Present study deals with the adoption of newer technologies for developing nations. Most of the developing countries due to lack of resources perform techno-socio-economic analyses on the already existing models of the developed ones. Such adopted technologies may not perform effectively because of unlike socio-economic factors. Hence, it becomes important to select new technologies based on appropriate and suitable criteria with respect to a particular country. In this paper, we have demonstrated selection of optimal electronic toll collection (ETC) system for India. In this context, we have considered thirteen crucial parameters for selection of appropriate ETC system. Cost is found to be the pivotal selection criterion in India. Further, fuzzy logic based MADM (multiple attribute decision making) approach is employed for selection of optimal ETC system for India. RFID-based (radio frequency identification) ETC is found to be the most suitable alternative among all considered ETC technologies. Our results are in strong agreement with the report of apex committee, appointed by “Government of India (Ministry of Road Transport & Highways)” for implementation of ETC in India. © 2014 Elsevier B.V. All rights reserved.

an optimal electronic toll collection system for India”. In last few years, an exponential increase in number of automobiles is noticed in India [1–5]. This can also be easily observed at Toll Plazas. Other The most influential factor towards techno-socio-economic important observations are contribution towards pollution and conditions of any country is the formulation of newer policies. An increasing road congestion. These cause time wastage, accidents inappropriate decision can adversely affect the future scenarios and unnecessary quarrels. World Bank’s report in 2009 reflects of a nation. The major part of new policy formulation of the most the seriousness of the issue [6]. They have reported a loss of “six of the countries is devoted to adoption of newer technologies. billion USD” every year in India only because of road congestion In this context, comprehensive investigation of all past, present and adverse environmental impact caused due to it. In order to and future technological, social, economical and environmental overcome this problem either we have to widen the road network aspects is mandatory. A developed country follows the same or need to adopt comparatively better tolling systems. Widening protocol. On the other hand, third world countries (due to lack of of road network is quite difficult due to constraints of space and resources) adopt the technologies which are already successfully infrastructure. Thus we are left with the only solution that is implemented in the developed ones. It is to be noted that they adaption of better electronic toll collection systems (ETC). ignore the key point of distinct resources, infrastructure, psycholThe most promising feasible ETC technologies in the world ogy of the people and many other socio-economic factors. Thus are based on (1) DSRC (dedicated short range communication) adoption as well as implementation of advanced technologies has which covers barcode and RFID (radio frequency identification) become the most frightening as well as issue of prime concern in [7–11], (2) video tolling that includes ANPR (automatic number recent time. Present study addresses such an issue – “selection of plate reader) [12–15], (3) global positioning system (GPS) or geographic information system (GIS) or vehicle positioning system (VPS) and (4) infrared short range communication (ISRC) based on ∗ Corresponding author. Tel.: +91 9467766881/1905 237921; calm active infrared [16–21]. Barcode-based ETC has a bar-coded fax: +91 1905 237945. E-mail addresses: [email protected], gaurav [email protected] sticker attached to the vehicle and is read by a laser scanner when it passes through the toll plaza. It is the simplest as well as the (G. Vats). 1. Introduction

http://dx.doi.org/10.1016/j.asoc.2014.04.006 1568-4946/© 2014 Elsevier B.V. All rights reserved.

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

oldest technology. It is widely used in various applications such as in library for managing book record, shopping plazas to take an account of sale and purchase, food industry to store food details and many more. Despite of these all it also has several drawbacks in order to be used for toll collection system such as lack of reliability (as can be easily imitated), less accuracy in bad weather, lack of flexibility, slow data read rate, less storage information and easy to be theft. Second technology is RFID-based ETC system [16,18], which has an In-vehicle unit (IVU) installed on the front windshield of the vehicle. This IVU interacts with the RFID frequency reader or antenna at toll plaza and transaction is done accordingly. It contains a cash card for payment of road tax which can either be prepaid or postpaid. It contains more information in comparison to barcode, has faster reading rate, tough to be fraudulent and also comparatively more reliable. It is also observed that sometimes it shows the problem of interference among frequency of devices (mobile phones, other IVU, walkie-talkies, FM radio or other electronic gadgets) in vicinity of the toll plaza or passing vehicles. Angle of installation and alignment plays an important role for reliability and high accuracy of these systems. Third important technology is ANPR [18,22]. It utilizes a stationary camera to record and identify the number plate of vehicles passing through toll plaza. The identified license numbers are matched in the database (connected with transport office) and toll is deducted. If the recorded number is not read properly or not found in the records, it issues an enforcement violation alarm to the alert the authorities. In this way, it simultaneously solves two objectives; identification of vehicle for deduction of toll tax and issuing/recording violation enforcement alert. The Indian government has started issuing “high security number plates”, which is tough to be falsified. Thus this technology will also be helpful to detect the stolen vehicles and vehicles with fake number plates. It also has constraints of high cost and reduced accuracy under tempestuous environment conditions. Calm active infrared [23–25] is a relatively new technology. It is similar to RFID system, the only difference is that it has an active infrared unit installed on vehicle which contains all the information. In comparison to RFID, it has a faster data reading rate, reliability, accuracy, efficiency and it works well in all environment conditions. It also comes over the problem of interference. Lack of interoperability, vendor support and high cost are the roadblocks in usage of this technology. Apart from these, it is still under research and many other aspects need to be studied yet. Fifth technology in this list is VPS. VPS-technique [8,26,27] consists of worldwide satellite navigation system incorporation with a communication mechanism. It works with the help of a global positioning system (GPS) unit installed on vehicle attached to an on board unit (OBU), which stores the coordinates of the vehicle and send the transaction information to the toll authorities via GSM (global system mobile communication). This system is highly reliable, accurate and efficient. The efficiency of this system is not affected by environmental conditions. It provides a payment option only for the distance travelled and is highly flexible in generating the corresponding payment details. It can also be used by the police petrol for highway surveillance and theft prevention of automobile. The associated shortcomings for this system are its excessively high installation, running and maintenance cost, careful handling, requirement of extra power and other accessories. It is clear that there are no clear trade-offs among the above mentioned technologies. Due to this, it becomes an important task to decide the best option among the existing ones. In such a state of ambiguity when one is not even able to choose the best among the existing alternative, there is no space for the question of adopting a hybrid technology. It also demotivates the policy makers to adopt newer advanced technologies. A single wrong decision can bring up loads of problems for coming generations with huge

445

wastage of money and time. Therefore, it becomes essential to predict the best solution in terms of best alternative for such problems using a highly subjective decision making technique. Such problems can be tackled using multiple attribute decision making (MADM) techniques. A variety of methods are reported under MADM category. These methods include simple additive weighting (SAW), analytic hierarchy process (AHP) [28], graph theory and matrix approach (GTMA) [29], VlseKriterijumska Optimisacija I Kompromisno Resenje (VIKOR) [30], technique for order preference by similarity to ideal solution (TOPSIS) [31] and many others. These have been successfully applied to various fields such as manufacturing processes [32], supply chain management [33], social science decisions [34], financial decisions [35] and engineering problems [36,37]. These methods are also used by our group in last few years and found to be efficient and effective [38–47]. MADM models are used to select best alternative from the large number of alternatives for a set of selection criteria. Moreover, these also tell about the degree of closeness in terms of rank index. The above mentioned MADM approaches work on crisp values of attributes. However, in case of selection of advanced technologies, most of the attributes/parameters depend on views of various decision makers (such as user, operators, government, distributors, technical and economy experts etc.). There are no clear boundaries among the views of these decision makers. Such selection issues can be dealt with fuzzy set theory aid with MADM approaches. The aim of present work is to select the optimal ETC technology for Indian roads under fuzzy environment using fuzzy VIKOR methodology. The present study is one of the first efforts for selection of optimal ETC.

2. Selection criteria for evaluation of ETC in India We have identified the following parameters for the selection of optimum ETC system for Indian roads. These are based on our discussion with toll authorities, operators and users (daily users as well as occasional users); and reports published by various researchers and experts [18,26,48–54] about pros and cons of various technologies. Parameter

Description

Cost (C1)

It is the prime factor for investing in any new technology. It includes cost of installation, running and maintenance. Before adopting a newer technology countrywide it is crucial to estimate about budget and this became vital in case of third world countries. So we have to make some compromises according to our financial aids. It questions the ability of system not to create any confusion regarding vehicle identity and payment options. As almost in every country different category of vehicles (e.g. two-wheelers, SUV, buses, trucks, multi-axle vehicles etc.) have to pay different amount of toll fees for the same travelled distance. So there must not be any ambiguity in vehicle detection and corresponding money deduction. It describes the extent to which the technology is safer for environment. The main aim of adopting an advanced technology in present case is to reduce the losses due to environmental pollution and road congestion. It is to be noted that sometimes apart from the main objective the same technology may also serve in several other manners. So it is always suggested to take an account of possible future prospective with a small compromise in current assets. Utilization of ETC systems for navigation, theft prevention of automobiles and traffic surveillance are well known examples of such secondary objectives.

Reliability/accuracy (C2)

Negative environmental impact (C3)

Flexibility (C4)

446

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

Parameter

Description

Congestion reduction (C5)

Our main objective behind implementation of ETC system is reduction in road congestion. It is well known that ETC has two most popular versions namely: with or without gating system depending on AVI (automated vehicle identification) technology used. Again here arise many questions like what is the congestion rate at present; how much is required presently and in future; what is the cost for different congestion levels as well as reduced congestion levels (using ETC) in terms of fuel, environmental impact and amount spent for the technology. It is quite tedious to understand, and requires knowledge of both economic and technological issues. It is the rate at which the adopted system can read, detect and covey the information of vehicles passing through the toll plaza under normal conditions. It is influenced by many factors other than the capability of adopted system such as weather condition, mist, dust, presence of metals, positioning angle of tag and receiving antenna/camera. It describes the willingness of people to accept the technology (users, operators, authorities and government). This varies man to man depending on their psychology and understanding. It judges the feasibility of implementation as per present infrastructure and economy of the country. We cannot impose huge changes in the infrastructure for just adopting a new technology thus it plays a crucial role in adoption of a newer technology. There may be cases of fraudulence with the units installed on vehicles in order to get rid of road tax as there will be no regular manual checking. Therefore, the mechanism must be rigid enough against any kind of cheating. Still, if anyhow, it is cheated than it should be smart enough to alarm for any kind of counterfeiting. One should make sure that the technology is well known for its positive as well as negative aspects. If it is not so and some shortcomings are identified later than it will be problematic for all. Simultaneously, such circumstances will lead to wastage of huge amount of time and money. Customers and operator both prefers handy technology and hence is one of the important criteria. ETC is not only a system but a network which connects a number of points including user, automated vehicle identification unit, vehicle classification unit, toll operator, transaction unit, violation enforcement, helpline and governing body. These all are interrelated. If any problem occurs in the network than it has to be resolved as soon as possible, which depends on simplicity of the system. It is observed that sometimes a small fault can affect the efficiency of complete system adversely. It defines the capability of a system (detection unit at the toll) to identify the number of vehicles at a time. It varies time to time depending on the congestion. During peak hours when congestion is high it is natural that it may not detect/miss a few vehicles due to very less spacing between consecutive vehicles or miss-alignment between transmitter and receiver. Similarly, an automobile at very high speed may be missed due to limitation of the installed devices. Thus the access range of detecting automobiles at peak hours and at high speed is termed as access rate.

Data rate (C6)

Acceptance (C7)

Implementation (C8)

Theft detection (C9)

Maturity (C10)

Ease to use (C11) System complexity (C12)

Access rate (C13)

3. Methodology used 3.1. Modified digital logic (MDL) It is a fact that all the parameters have different impact for selection of an appropriate technology for a country and hence cannot be assigned equal weights. So it becomes vital to find out the priority of each parameter as per the assets (like economic condition, infrastructure, public and government convenience etc.) of a country. MDL is one of the well known techniques to estimate

the weights under such conditions [55]. It includes expert opinion to assign initial priorities as 1, 2 and 3 for less, equally and more important parameters, respectively. Based on the expert opinion decision table is formed under pair-wise comparison. Prior to formation of the MDL table, one need to estimate the number of possible positive decisions as N = n(n − 1)/n, where n is the number of attributes/technological parameters. Further summation of all positive decisions (P) for a particular parameter on normalization leads to final weight (Wj ) as: Wj =

Pj

n

(1)

P j=1 j

3.2. VIKOR The VlseKriterijumska Optimisacija I Kompromisno Resenje (VIKOR) method is a compromise approach MADM model [30]. The analysis of VIKOR is highly accurate and provide close to real solution. It makes the use of utility weight, thus enabling the different users to apply expert opinion. The normalization norms used in VIKOR are linear. 3.3. Fuzzy logic Fuzzy approach was introduced to tackle the problems with lack of precision [56], where there are no clear boundaries between the system and surroundings. It also deals with the problems where it is tough to distinguish between member and non-member objects of a set. Belleman and Zadeh [57] employed fuzzy approach for multiple criteria decision making. It emphasizes on the possibility rather than the probability. It is based on fuzzy set theory. Bevilacqua et al. [58] defined fuzzy set as a set comprised of a membership function within the interval [0,1], which describes the extent of relevance of an element for being member of the set. In this approach, initially all comparisons are done using linguistic variables. Further, these linguistic variables are assigned fuzzy values in order to have comparable numerical values without any ambiguity using appropriate membership function. In current study, we have used trapezoidal fuzzy numbers (a1 , a2 , a3 , a4 ) for {a1 , a2 , a3 , a4 ∈ R ; a1 ≤ a2 ≤ a3 ≤ a4 }. It is one of the simplest and most commonly used kinds of division for fuzzy numbers. The membership function a (x) of trapezoidal fuzzy number is defined as ⎧ x−a 1 ⎪ , x ∈ [a1 , a2 ] ⎪ ⎪ ⎪ a2 − a1 a (x) =

⎪ ⎪ ⎨ 1,

x ∈ [a2 , a3 ]

a4 − x ⎪ ⎪ , x ∈ [a3 , a4 ] ⎪ ⎪ a4 − a3 ⎪ ⎪ ⎩ 0

(2)

Otherwise

Fig. 1 graphically illustrates trapezoidal fuzzy number. 4. Proposed subjective fuzzy–VIKOR approach In this section, we have proposed the subjective fuzzy–VIKOR approach. This technique is used in the present study for selection of electronic toll collection system for Indian roads. We have used MDL for inter-comparison among all parameters leading to their corresponding subjective weights. Further it is followed by fuzzy VIKOR to obtain the best existing technology as per Indian economy and infrastructure. It includes following steps: Step 1 Calculation of MDL weights. As discussed in Section 3 MDL weights (Wj ) are calculated for each technological parameter. This judges the

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

447

Step 4 Normalization The first basic necessity of any comparison is that all quantities being compared must be on the same scale. Therefore, our next step is normalization of aggregated fuzzy rating. Here, we can have two situations. One is properties with higher desired value (benefit criterion) and other is properties with lower desired value (cost criterion). For both the cases, normalization is done using the following Eqs.:



xij1 xij2 xij3 xij4 + , + , + , + xij1 xij2 xij3 xij4

ij =



− − − − xij2 xij3 xij4 xij1 , , , xij1 xij2 xij3 xij4

ij =

+ = max(xij4 ), where xij4

Fig. 1. Trapezoidal fuzzy number.

inter-criteria (parameter) comparison for each alternative (technology). Step 2 Define linguistic terms, relevant membership function and corresponding fuzzy numbers. A set of fuzzy rates is required in order to compare all the alternatives for each criterion. These fuzzy terms are





,

j∈J

(4)

,

j ∈ J

(5)



− j ∈ J and xij1 = min(xij1 ),

j ∈ J; J



corresponds to benefit criterion and J corresponds to cost criterion. Step 5 Defuzzification. Defuzzification is performed to obtain the crisp values for each criterion corresponding to each alternative. This provides a quantitative value for the linguistic variables and fuzzy numbers assigned based on the verbal reasoning of the decision makers. Following equations lead to the crisp values:

(x).xdx fij

= Defuzz(xij ) =



 (x).dx



xij2

=





xij3

((x − xij1 )/(xij2 − xij1 )) · xdx +



xij1

xij2

xij3

((x − xij1 )/(xij2 − xij1 ))dx + =

2

xij3

2

These crisp values, thus obtained are incorporated with MDL weightage to calculate final ranking using VIKOR approach as discussed below. Step 6 Determination of ideal and negative ideal solutions; The ideal solution f* and the negative ideal solution f− are determined as: f ∗ = {max fij } f

−

(7)

= {min fij }

Si = (3)

n  j=1

Wj

(8)

(fj∗ − fij )

(fj∗ − fj− )



Ri = Maxj Wj

Thus the obtained decision matrix (D) is shown as:

xm1

((xij4 − x)/(xij4 − xij3 ))dx

xij2

Step 7 Calculation of utility and regret measures;

k

⎢ ⎢ x21 ⎢ ⎢. D=⎢ ⎢ .. ⎢ ⎢ .. ⎣.

xij4

dx +

−xij1 − xij2 + xij3 + xij4

⎧ xij1 = min{aijk1 } ⎪ ⎪ k ⎪ ⎪ ⎪ 1 ⎪ ⎪ = aijk2 x ⎨ ij2 k 1 ⎪ ⎪ xij3 = aijk3 ⎪ ⎪ k ⎪ ⎪ ⎪ ⎩ xij4 = max{aijk4 }

x11

(6)

xij3

−xij1 xij2 + xij3 xij4 + (1/3)(xij4 − xij3 ) + (1/3)(xij2 − xij1 )

assigned by the policy/decision makers and responsible for intra criterion comparisons of the alternatives. Step 3 Construction of decision matrix. Let n be the technological parameters (criteria) and m be the technologies (alternatives). For k number of decision makers in the purposed model the aggregated fuzzy rating for Cj criterion is represented as xijk = {xijk1 , xijk2 , xijk3 , xijk4 }. For i = 1,2,. . .m; j = 1,2,. . .n and k = 1,2,. . .k, xijk is calculated as [33,59]:

⎡

((xij4 − x)/(xij4 − xij3 )) · xdx



xij2

xij1

xij4

xdx +

x12

· · · x1n

x22

· · · x2n

.. .

..

.

.. .

.. .

..

.

.. .

xm2

· · · xmn

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

;

(fj∗ − fij )

(fj∗ − fj− )

∀i

(9)

 ;

∀i

(10)

where Si and Ri represent the utility and regret measures, respectively and Wj is the relative weight assigned to the jth parameter using MDL. Step 8 Determination of VIKOR index; Qi = 

 S − S∗  i S−

− S∗

+ (1 − )

 R − R∗  i R− − R∗

; ∀i

(11)

where, Qi represents the ith alternative’s VIKOR value,  is the group utility weight, it is generally considered as 0.5 (unsupervised) and; S ∗ = Mini (Si );

(12)

448

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

Table 1 Subjective weights using MDL. Parameters/attributes

C1

C2

C3

C4

C5

C6

C7

C8

C9

C10

C11

C12

C13

Positive decisions

Weights

Ranks

Cost (C1) Reliability/accuracy (C2) Negative environmental impact (C3) Flexibility (C4) Congestion reduction (C5) Data rate (C6) Acceptance (C7) Implementation (C8) Theft detection (C9) Maturity (C10) Ease to use (C11) System complexity (C12) Access rate (C13) Sum

2 1 1 1 1 1 1 1 1 1 1 1 1

3 2 1 1 1 1 3 3 1 1 1 1 1

3 3 2 1 3 1 3 3 3 1 1 1 1

3 3 3 2 3 3 3 3 3 3 3 3 3

3 3 1 1 2 1 3 3 3 1 1 1 1

3 3 3 1 3 2 3 3 3 1 1 1 1

3 1 1 1 1 1 2 2 1 1 2 1 1

3 1 1 1 1 1 2 2 1 1 2 1 1

3 3 1 1 1 1 3 3 2 1 1 1 1

3 3 3 1 3 3 3 3 3 2 1 1 1

3 3 3 1 3 3 2 2 3 3 2 1 3

3 3 3 1 3 3 3 3 3 3 3 2 3

3 3 3 1 3 3 3 3 3 3 1 1 2

36 30 24 12 26 22 32 32 28 20 18 14 18 312

0.115385 0.096154 0.076923 0.038462 0.083333 0.070513 0.102564 0.102564 0.089744 0.064103 0.057692 0.044872 0.057692 1

1 4 7 13 6 8 2 3 5 9 11 12 10

S − = Maxi (Si );

(13)

R∗ = Mini (Ri );

(14)

R− = Maxi (Ri );

(15)

The alternative with least value of VIKOR index Qi is preferred. 5. Results and discussion In the previous sections, we have discussed the existing alternative ETC technologies and prime parameters for selection of ETC system in India. Fig. 2 demonstrates the schematic hierarchy of the formulated research problem (selection of ETC in Indian scenario). Level zero indicates our objective (selection of optimal ETC system for India) that has to be chosen from the shortlisted five important technologies namely barcode, RFID, ANPR, VPS and active infrared. Further, on the basis of discussion with decision makers (Toll users, operators, authorities and technological experts), we found that selection of the most suitable alternative among these existing technologies depends on thirteen criteria (discussed in Section 2) which are shown in level 2 (parameters) of the Fig. 2. Interdependency of technologies on these parameters shows the complexity of the problem. Moreover, it is time consuming and requires knowledge of both technological as well as economic aspects. Once the prime parameters are identified, the next question is to prioritize these parameters. It is to be noted that, different countries have different infrastructure, policies, financial support etc. and hence have different priorities. To prioritize the parameters, here we have used MDL approach. In order to assign relative weights to above mentioned parameters (Section 2), we have made pair-wise comparison and allocated 1, 2 and 3 numbers for relatively least, equal or more important parameters, respectively. The relative decision matrix is

Table 2 Linguistic variables and corresponding fuzzy numbers. Linguistic variable

Fuzzy number

Exceptionally high (EH) Very high (VH) High (H) Above average (AA) Average (A) Very low (VL) Extremely low (EL)

(0.8, 0.9, 1.0, 1.0) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.4, 0.5, 0.5, 0.6) (0.2, 0.3, 0.4, 0.5) (0.1, 0.2, 0.2, 0.3) (0.0, 0.0, 0.1, 0.2)

formed based on pair-wise comparison (MDL approach) and the calculation for weights for all the considered criteria are summarized in Table 1. Cost is found to be the most influential parameter for introducing such technologies in India. Flexibility is found to be the least contributing factor in selection of ETC system for India. Contributions of all prime factors are illustrated in Fig. 3. However, it is obvious that contributions of these parameters are subjected to fluctuation with respect to change in socio-economic aspects of any country. The next step is comparison of all alternatives for each parameter. As the comparative analysis has to be done on the basis of conclusions of the discussions of the decision makers; there arises a need of hypothetical scale to compile all these conclusions. In this context, fuzzy approach is used here. Fuzzy logic approach works well with such kind of problems. It utilizes linguistic variables for comparison among alternative technologies. These are further converted into fuzzy numbers as illustrated in Table 2 for the present study. The worst range is termed extremely low (EL) (most undesirable) and the best is termed as exceptionally high (EH). Based on our discussion with all the decision makers we filled the Tables 3 and 4. These demonstrate the linguistic decision

Table 3 Linguistic decision matrix of ETC technologies for all evaluation criteria. Evaluation criteria (parameters)

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13

ETC-Technologies (alternatives) Bar code

RFID

ANPR

VPS

Calm active infrared

VL A VL EL A A EL VL A A VL VL VL

A H A AA VH H EH EH H EH VH VL H

H H VL A VH VH A AA EH H H A VL

EH EH VL VH EH VH A A A H A H H

EH H EH VL VH EH EL VL H VL VL VH VH

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

449

Fig. 2. A schematic hierarchy for selection of the most suitable ETC technology for India.

Fig. 3. Contribution of governing parameters towards selection of ETC technology for India.

matrix and corresponding fuzzy ratings respectively. Here we have formed a single decision matrix (after discussions with all the decision makers) rather than having a separate decision matrix for each decision maker. However, it is to be noted that final decision matrix

may change depending on the requirements and circumstances. Further, fuzzy values are normalized (using Eqs. (4) and (5)) and finally converted into crisp values (Eq. (6)). Crisp values are quantitative comparative values for the subjective verbal reasoning of the

Table 4 ETC technologies (alternatives), essential parameters for evaluation and corresponding fuzzy ratings. Evaluation criteria (parameters)

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13

ETC-Technologies (alternatives) Bar code

RFID

ANPR

VPS

Calm active infrared

(0.1, 0.2, 0.2, 0.3) (0.2, 0.3, 0.4, 0.5) (0.1, 0.2, 0.2, 0.3) (0.0, 0.0, 0.1, 0.2) (0.2, 0.3, 0.4, 0.5) (0.2, 0.3, 0.4, 0.5) (0.0,0. 0, 0.1, 0.2) (0.1, 0.2, 0.2, 0.3) (0.2, 0.3, 0.4, 0.5) (0.2, 0.3, 0.4, 0.5) (0.1, 0.2, 0.2, 0.3) (0.1, 0.2, 0.2, 0.3) (0.1, 0.2, 0.2, 0.3)

(0.2, 0.3, 0.4, 0.5) (0.5, 0.6, 0.7, 0.8) (0.2, 0.3, 0.4, 0.5) (0.4, 0.5, 0.5, 0.6) (0.7, 0.8, 0.8, 0.9) (0.5, 0.6, 0.7, 0.8) (0.8, 0.9, 1.0, 1.0) (0.8, 0.9, 1.0, 1.0) (0.5, 0.6, 0.7, 0.8) (0.8, 0.9, 1.0, 1.0) (0.7, 0.8, 0.8, 0.9) (0.1, 0.2, 0.2, 0.3) (0.5, 0.6, 0.7, 0.8)

(0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) (0.1, 0.2, 0.2, 0.3) (0.2, 0.3, 0.4, 0.5) (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9) (0.2, 0.3, 0.4, 0.5) (0.4, 0.5, 0.5, 0.6) (0.4, 0.5, 0.5, 0.6) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8) (0.2, 0.3, 0.4, 0.5) (0.1, 0.2, 0.2, 0.3)

(0.8, 0.9, 1.0, 1.0) (0.8, 0.9, 1.0, 1.0) (0.1, 0.2, 0.2, 0.3) (0.7, 0.8, 0.8, 0.9) (0.8, 0.9, 1.0, 1.0) (0.7, 0.8, 0.8, 0.9) (0.2, 0.3, 0.4, 0.5) (0.2, 0.3, 0.4, 0.5) (0.2, 0.3, 0.4, 0.5) (0.5, 0.6, 0.7, 0.8) (0.2, 0.3, 0.4, 0.5) (0.5, 0.6, 0.7, 0.8) (0.5, 0.6, 0.7, 0.8)

(0.8, 0.9, 1.0, 1.0) (0.5, 0.6, 0.7, 0.8) (0.8, 0.9, 1.0, 1.0) (0.1, 0.2, 0.2, 0.3) (0.7, 0.8, 0.8, 0.9) (0.8, 0.9, 1.0, 1.0) (0.0, 0.0, 0.1, 0.2) (0.1, 0.2, 0.2, 0.3) (0.5, 0.6, 0.7, 0.8) (0.1, 0.2, 0.2, 0.3) (0.1, 0.2, 0.2, 0.3) (0.7, 0.8, 0.8, 0.9) (0.7, 0.8, 0.8, 0.9)

450

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

Table 5 Calculated crisp values for assigned fuzzy rates. Evaluation criteria (parameters)

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13

ETC-Technologies (alternatives) Bar code

RFID

ANPR

VPS

Calm active infrared

2.333 0.366 2.333 0.086 0.366 0.366 0.077 0.233 0.366 0.366 0.259 2.333 0.259

3.666 0.666 3.666 0.592 0.833 0.666 0.944 0.944 0.666 0.944 0.925 2.333 0.740

6.666 0.666 2.333 0.407 0.833 0.833 0.366 0.533 0.944 0.666 0.740 3.666 0.259

9.444 0.944 2.333 0.925 0.944 0.833 0.366 0.366 0.366 0.666 0.407 6.666 0.740

9.444 0.666 9.444 0.259 0.833 0.944 0.077 0.233 0.666 0.233 0.259 8.333 0.925

Table 6 Calculated utility measure, regret measure and corresponding VIKOR ranking. Evaluation criteria (parameters)

Utility measure Regret measure VIKOR rank index VIKOR ranks

ETC-Technologies (alternatives) Bar code

RFID

ANPR

VPS

Calm active infrared

0.750 0.102 0.907 4

0.206 0.046 0 1

0.406 0.070 0.357 2

0.488 0.115 0.759 3

0.700 0.115 0.953 5

decision makers. Table 5 illustrates the calculated crisp values obtained after normalization (not shown here) of aggregated fuzzy ratings. These quantitative numbers are used with VIKOR approach (Eqs. (7)–(15)) to obtain the rank indices of all alternatives. Table 6 shows the utility and regret measures, corresponding rank indices and ranks for the alternative technologies. Fig. 4 recapitulates the methodology used. It summarizes the proposed methodology and explicates how the views of the decision makers are quantitatively compiled by us in the present study. These started with verbal

discussions with different experts (technical, social and economical), governing bodies and common public (users) associated with various toll barriers throughout the country. Our calculations predict that RFID-based ETC is the most suitable alternative for Indian roads in current scenario. ANPR stands on second podium and is the most suitable alternative after RFID-based ETC in India. Due to high cost VPS-based ETC comes at third position despite of the fact that it is highly flexible and can act well for multitasks. Barcode-based and infrared-based systems are out of merit because of limited features and lack of maturity respectively. However, we suggest

Fig. 4. Flow chart for data collection, quantification, optimal outcome and final decision making using proposed methodology.

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

that if the decision makers can make some compromise with cost than VPS-based ETC can be proved a boon considering the future aspects. We found that our results are in good agreement with the “report by apex committee for ETC implementation”, published by government of India (GOI) “Ministry of Road Transport & Highways” in September 2011 [60]. The committee headed by “Mr. Nandan Nilekani”, stressed on user convenience, rate of acceptance and ease of implementation; recommended RFID as the most suitable ETC technology for Indian roads. Results of this committee are evidence of the effectiveness of the proposed approach. 6. Conclusions MADM methods are employed for selection of optimal electronic toll collection (ETC) system in India. Modified digital logic (MDL) method is used to calculate weightage of all influential parameters for evaluation of ETC system in India. Cost and flexibility are found to be the most and least critical parameters, respectively. Further priority order of ETC-technologies is determined using fuzzy VIKOR approach incorporation with MDL weights. RFID-based ETC system is found to be the most appropriate for Indian roads. ANPR and VPS are found to be at second and third positions respectively. The present study proposes the feasibility of fuzzy VIKOR method in policy formulation and adoption of optimal technologies as per infrastructure and economic conditions of a nation. Acknowledgments Rahul Vaish gratefully acknowledges financial support from Department of Science and Technology (DST), New Delhi and Indian National Science Academy (INSA), India under INSPIRE Faculty Award (ENG-01)-2011. References [1] K. Narayanan, Technology acquisition, de-regulation and competitiveness: a study of Indian automobile industry, Res. Policy 27 (1998) 215–228. [2] J. Pucher, N. Korattyswaropam, N. Mittal, N. Ittyerah, Urban transport crisis in India, Transp. Policy 12 (2005) 185–198. [3] R.K. Sinha, Automobile pollution in India and its human impact, Environmentalist 13 (1993) 111–115. [4] R. Ramanathan, Link between population and number of vehicles: evidence from Indian cities, Cities 17 (2000) 263–269. [5] J. Pucher, Z.R. Peng, N. Mittal, Y. Zhu, N. Korattyswaroopam, Urban transport trends and policies in China and India: impacts of rapid economic growth, Transp. Rev. 27 (2007) 379–410. [6] World-Bank, Development Dialogue; Spending on Infrastructure Drives Growth, World Bank India, New Delhi, India, 2009. [7] V. Chawla, D.S. Ha, An overview of passive RFID, IEEE Commun. Mag. 45 (2007) 11–17. [8] P. Blythe, RFID for road tolling, road-use pricing and vehicle access control, in: RFID Technology (Ref. No. 1999/123), IEE Colloquium on, IET, 1999, pp. 8/1-816. [9] M. Yu, D. Zhang, Y. Cheng, M. Wang, R.F.I.D. An, electronic tag based automatic vehicle identification system for traffic iot applications, in: Control and Decision Conference (CCDC), 2011 Chinese, IEEE, 2011, pp. 4192–4197. [10] Z. Li, Z. Zhou, C. He, X. Huang, Advances in RFID-ILA: the past, present and future of RFID-based indoor location algorithms, in: Control and Decision Conference (CCDC), 2012 24th Chinese, IEEE, 2012, pp. 3830–3835. [11] P. Huan, Z.-Y. Zhang, The electronic toll collection system of highway based on RFID [J], Commun. Transp. Syst. Eng. Inform. 2 (2004) 026. [12] S. Masada, Automatic toll collector for toll roads, Google Patents (1990). [13] R. Lotufo, A. Morgan, A. Johnson, Automatic number-plate recognition, in: Image Analysis for Transport Applications, IEE Colloquium on, IET, 1990, pp. 6/1–6/6. [14] M.L. Tam, W.H. Lam, Application of automatic vehicle identification technology for real-time journey time estimation, Inform. Fusion 12 (2011) 11–19. [15] J.S. Shieh, Method and system for two-way packet radio-based electronic toll collection, Google Patents (1995). [16] P. Blythe, Congestion charging: technical options for the delivery of future UK policy, Transp. Res. A: Policy Pract. 39 (2005) 571–587. [17] D. Kalbande, N. Deotale, P. Singhal, S. Shah, G. Thampi, An advanced technology selection model using neuro fuzzy algorithm for electronic toll collection system, Int. J. Adv. Comput. Sci. Appl. 2 (2011) 97–104.

451

[18] P.A. Sorensen, B.D. Taylor, Review and Synthesis of Road-use Metering and Charging Systems, Report Commissioned by the Committee for the Study of the Long-Term Viability of Fuel Taxes for Transportation Finance, UCLA Institute of Transportation Studies, 2005. [19] L. Vanajakshi, G. Ramadurai, A. Anand, Intelligent Transportation System, Synthesis Report on ITS Including Issues and Challenges in India, D.o.C.E. Center of Excellence in Urban Transport, Indian Institute of Technology, Madras, India, 2011. [20] A. Broaddus, C. Gertz, Tolling heavy goods vehicles: overview of European practice and lessons from German experience, Transp. Res. Record: J. Transp. Res. Board 2066 (2008) 106–113. [21] P.A. Sorensen, B.D. Taylor, Innovations in Road finance examining the growth in electronic tolling, Public Works Manage. Policy 11 (2006) 110–125. [22] M. Bibaritsch, G.O. Egeler, MAUT: enforcement: the enforcement system of the Austrian heavy goods vehicle toll, in: European Congress on Intelligent Transportation Systems and Services, 4th, 2004, Budapest, Hungary, 2004. [23] W.-Y. Shieh, W.-H. Lee, S.-L. Tung, C.-D. Ho, A novel architecture for multilanefree-flow electronic-toll-collection systems in the millimeter-wave range, IEEE Trans. Intell. Transp. Syst. 6 (2005) 294–301. [24] M. Staudinger, E. Mulka, Electronic vehicle identification using active infrared light transmission, in: At the Crossroads: Integrating Mobility Safety and Security. ITS America 2004, 14th Annual Meeting and Exposition, 2004. [25] S. Tropartz, E. Horber, K. Gruner, Experiences and results from vehicle classification using infrared overhead laser sensors at toll plazas in New York City, in: Intelligent Transportation Systems, 1999. Proceedings. 1999 IEEE/IEEJ/JSAI International Conference on, IEEE, 1999, pp. 686–691. [26] I. Catling, Road user charging using vehicle positioning systems, in: Road Transport Information and Control, 2000. Tenth International Conference on (Conf. Publ. No. 472), IET, 2000, pp. 126–130. [27] G. Charpentier, G. Fremont, The ETC system for HGV on motorways in Germany: first lessons after system opening, in: Proceedings of the European Transport Conference (ETC) 2003 held 8–10 October 2003, Strasbourg, France, 2003. [28] T.L. Saaty, How to make a decision: the analytic hierarchy process, Eur. J. Oper. Res. 48 (1990) 9–26. [29] R.V. Rao, A material selection model using graph theory and matrix approach, Mater. Sci. Eng. A 431 (2006) 248–255. [30] S. Opricovic, G.-H. Tzeng, Extended VIKOR method in comparison with outranking methods, Eur. J. Oper. Res. 178 (2007) 514–529. [31] Y.C.H. Deng, R.J. Willis, Inter-company comparison using TOPSIS with objective weights, Comput. Oper. Res. 27 (2000) 963–973. [32] R. Jeya Girubha, S. Vinodh, Application of fuzzy VIKOR and environmental impact analysis for material selection of an automotive component, Mater. Design 37 (2012) 478–486. [33] A. Shemshadi, H. Shirazi, M. Toreihi, M. Tarokh, A fuzzy VIKOR method for supplier selection based on entropy measure for objective weighting, Expert Syst. Appl. 38 (2011) 12160–12167. [34] S. Opricovic, Fuzzy VIKOR with an application to water resources planning, Expert Syst. Appl. 38 (2011) 12983–12990. [35] T. Kaya, C. Kahraman, Fuzzy multiple criteria forestry decision making based on an integrated VIKOR and AHP approach, Expert Syst. Appl. 38 (2011) 7326–7333. [36] P. Chatterjee, V.M. Athawale, S. Chakraborty, Selection of materials using compromise ranking and outranking methods, Mater. Design 30 (2009) 4043–4053. [37] K. Devi, Extension of VIKOR method in intuitionistic fuzzy environment for robot selection, Expert Syst. Appl. 38 (2011) 14163–14168. [38] A. Chauhan, R. Vaish, Hard coating material selection using multi-criteria decision making, Mater. Design 44 (2012) 240–245. [39] A. Chauhan, R. Vaish, Magnetic material selection using multiple attribute decision making approach, Mater. Design 36 (2012) 1–5. [40] A. Chauhan, R. Vaish, A comparative study on material selection for microelectromechanical systems, Mater. Design 41 (2012) 177–181. [41] A. Chauhan, R. Vaish, Pareto optimal microwave dielectric materials, Advanced Science, Eng. Med. 5 (2013) 149–155. [42] A. Chauhan, R. Vaish, Material selection for piezoelectric devices, Adv. Sci. Eng. Med. 5 (2013) 715–719. [43] A. Chauhan, R. Vaish, An Assessment of bulk metallic glasses for microelectromechanical system devices, Int. J. Appl. Glass Sci. (2013). [44] A. Chauhan, R. Vaish, C. Bowen, Piezoelectric material selection for ultrasonic transducer and actuator applications, in: Proceedings of the Institution of Mechanical Engineers, Part L: Journal of Materials Design and Applications, 2013, 1464420713493591. [45] G. Vats, R. Vaish, Selection of lead-free piezoelectric ceramics, Int. J. Appl. Ceram. Technol. (2013) 1–11. [46] G. Vats, R. Vaish, Piezoelectric material selection for transducers under fuzzy environment, J. Adv. Ceram. 2 (2013) 141–148. [47] G. Vats, R. Vaish, C.R. Bowen, Selection of Ferroelectric Ceramics for Transducers and Electrical Energy Storage Devices, Int. J. Appl. Ceram. Technol. (2013). [48] M.A. Chowdhury, A.W. Sadek, Fundamentals of Intelligent Transportation Systems Planning, Artech House Publishers, 2003. [49] J. Hellaker, Prometheus: Strategy, ITS Technology Collection on CD-ROM: SAE’s Essential Resource for ITS Vehicle Applications, 1998, 1990. [50] A. Kemeny, PROMETHEUS-Design Technics, Transportation Electronics, Vehicle Electronic in the 90, 1990. [51] J. Levine, S.E. Underwood, A multiattribute analysis of goals for intelligent transportation system planning, Transp. Res. C: Emerg. Technol. 4 (1996) 97–111.

452

S. Vats et al. / Applied Soft Computing 21 (2014) 444–452

[52] B. McQueen, I. Catling, The Development of IVHS in Europe, IVHS and Vehicle Communications, 1991. [53] T. Nakahara, N. Yumoto, ITS development and deployment in Japan, in: Intelligent Transportation System, 1997. ITSC’97. IEEE Conference on, IEEE, 1997, pp. 631–636. [54] J. Holguín-Veras, Q. Wang, Behavioral investigation on the factors that determine adoption of an electronic toll collection system: freight carriers, Transp. Res. C: Emerg. Technol. 19 (2011) 593– 605. [55] B. Dehghan-Manshadi, H. Mahmudi, A. Abedian, R. Mahmudi, A novel method for materials selection in mechanical design: combination of non-linear

[56] [57] [58] [59] [60]

normalization and a modified digital logic method, Mater. Design 28 (2007) 8–15. L.A. Zadeh, Fuzzy sets, Inform. Control 8 (1965) 338–353. R.E. Bellman, L.A. Zadeh, Decision-making in a fuzzy environment, Manage. Sci. 17 (1970), B-141–B-164. M. Bevilacqua, F. Ciarapica, G. Giacchetta, A fuzzy-QFD approach to supplier selection, J. Purch. Supply Manage. 12 (2006) 14–27. Y.Z. Mehrjerdi, Hierarchical multi-criteria risk-benefit analysis in fuzzy environment, Appl. Soft Comput. 13 (1) (2013) 590–599. GOI, Electronic Toll Collection: Report by Apex Committee for ETC Implementation in, Minister of Road Transport & Highways, 2011.