Transportation Research Part A 121 (2019) 37–55

Contents lists available at ScienceDirect

Transportation Research Part A

journal homepage: www.elsevier.com/locate/tra

Optimal budget allocation for risk mitigation strategy in trucking

T
industry: An integrated approach
⁎
Krishna Kumar Dadsenaa, S.P. Sarmahb, , V.N.A. Naikana, Sarat Kumar Jenac
a
Subir Chowdhury School of Quality and Reliability, IIT Kharagpur, 721302, India
b
Department of Industrial and Systems Engineering, IIT Kharagpur, 721302, India
c
Goa Institute of Management, Poriem, Sattari, Sanquelim, Goa 403505, India

A R T IC LE I N F O ABS TRA CT

Keywords: Trucking industry plays a major role in the transportation of goods across different geographical
Risk-mitigation strategy locations. The operational complexities of the trucking industry lead to various risks. This study
Integrated approach focuses on effective design and implementation of risk-mitigation strategies for the trucking
Possibility theory industry with consideration for budget restrictions. In this paper, both subjective and objective
Budget allocation
attributes are considered in the mathematical modeling and thereby tries to capture realism in
Trucking industry
the strategic decision-making process. The results of the study provide novel insights that relate
the impact of risk on the cost of mitigation. Further, the effect of three characteristics, targeted
risk level (TRL), implementation cost (IC) of strategy, and the probability of risk occurrence (PO)
is shown in designing and developing risk-mitigation strategies. The experimental analysis not
only augments theoretical knowledge related to risk management decision-making processes but
also contributes to designing and developing a risk-mitigation strategy under economic con-
straints. From the managerial perspective, the study demonstrates how decision-makers can
benefit from an integrated approach to develop a more holistic understanding of risk-manage-
ment processes. This study also provides guidelines in policy selection considering higher return
on investment (ROI). The paper concludes by highlighting the key findings and discussing op-
portunities for future research.

1. Introduction

In many countries, trucking sector is the lifeline for the road freight transportation system, and it serves almost all types of
industries. As per the statistics, approximately 70% of the US community is served by the trucking industry (FHWA 2015), which
generated the revenue of $676.2 billion in 2016 (American Trucking Association, 2017). Similarly, 65% of the Indian community is
served by the trucking industry (Raghuram, 2015), and the industry is expected to be worth of $300 billion by 2020 (Sangwan, 2017).
However, in the current business environment, the trucking industry is subjected to various disruptions in its operations and they are
considered as significant risk. Risk is defined here as the likelihood of occurrence of the problems and their negative impact (Young
and Tippins, 2000). Now a days, the likelihood of occurrence of man-made factors has also became a crucial issue for the transport
managers (Klemick et al., 2015). Man-made factors are human actions which may lead to failure or disruption in the operation of the
industry (Rodrigue et al., 2009). In the operation of the trucking industry, man-made factors are such as changes in regulations,
political unrest, strike, and changing levels of economic development. In India, waiting at tollgates, stoppage delays on highways and

⁎
Corresponding author.
E-mail addresses: [email protected] (S.P. Sarmah), [email protected] (V.N.A. Naikan), [email protected] (S.K. Jena).

https://doi.org/10.1016/j.tra.2019.01.007
Received 13 June 2018; Received in revised form 25 September 2018; Accepted 3 January 2019
0965-8564/ © 2019 Elsevier Ltd. All rights reserved.
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

check posts leads to the loss of approximately $1.48 to $2.77 billion of fuel, which indirectly affects the operation of the Indian
trucking industry (Mitra, 2009). Industry incurs an annual loss of $14.7 billion due to the increasing fuel consumption costs and $6.6
billion because of transportation delays (Mitra, 2009). Further, the growing shortage of the truck drivers severely affects the trucking
business and leads to operational failures and economic losses. The problem of driver shortage is found not only in India but also in
developed countries including the United States of America and the countries of the European Union (Min and Lambert, 2002).
Currently, around 27% of India’s entire truck fleet remains idle every day due to the shortage of drivers (Numadic, 2017). Another
example of disruption in the operation of the Indian trucking industry is due to the unstructured freight rate. The Government of India
(GoI) incurred a revenue loss of around $11.5 million due to the strike of sixteen trucks and tipper owners for nine days to protest
against the freight rate (Bansal, 2015). Similarly, according to an Indian news portal report, 970 carriers with a substantial number of
trucks were forced to close their businesses in 2013 because of being unable to handle the risks associated with their business
(Magoci, 2016). For better understanding of the aforementioned risks and their losses, one must understand the day-to-day opera-
tions of the industry (Biyani, 2017).
From the above discussion, one can recognize the importance of giving attention to the control of risks in the decision-making
process of the trucking industry. Kersten (2012) mentioned the incompetence in the traditionally designed operation of the trucking
industry. Therefore, for managing those risks and their assessment, selection of risk mitigation strategies become a critical element in
the process. Proper handling of risks requires industry to identify the unexpected losses arising from a sequence of failures and/or
causal events (Lewis, 2003). Organizations must understand the potential of such events that give rise to the problems and their
negative impacts on efficient operations of the industry (Mattsson and Jenelius, 2015). Management needs to rethink about the
decision-making process to accommodate a better understanding of risk assessment and investment required to mitigate and control
the risks (e.g. recognizing risk vs. feasibility of implementation on a given budget, Vasco and Morabito, 2016; Zeng and Yen, 2017).
In the present study, we have considered three objectives, namely, feasibility of mitigation strategies, economic cost reduction
(ECR) and targeted level of risks under a given budget simultaneously and thereby this work is different from the earlier work. To
address this, an integrated approach is proposed in this paper to derive expressive results for managerial decision-making. This study
makes three major contributions. First, it enhances the understanding of the effect of interconnection between risks, mitigation
strategies, and their economic consequences related to the trucking industry. Second, this study has considered the subjective as well
as objective uncertainties for design and development of risk mitigation strategies. Third, this study examines and identifies the
optimal range of budgetary investment in the selection of a risk mitigation strategy to maximize the return on investment (ROI).
The rest of the paper is organized as follows. In Section 2, the related literature is discussed. The methodology and formulation of
the mathematical model, its calculation is shown in Section 3. The solution and results are discussed in Section 4. Sensitivity analysis
and managerial implications are discussed in Section 5. The validation of the proposed model is discussed in Section 6. Finally, in
Section 7, the study is concluded and future directions are presented.

2. Literature review

Current statistics show that road freight transportation has become one of the more vulnerable elements in supply chain man-
agement. Abkowitz (2002) argues risk management in the transportation industry and recommends that if these risks are monitored,
evaluated, and controlled as a single integrated function then better overall risk-management strategies will begin to emerge. Dillon
et al. (2003) identified the importance of considering resource restrictions during the risk analysis and suggested that quantitative
risk analysis under budget constraints adds the robustness in risk-mitigation decision-making processes. Wyman (2011) finds that a
systematic risk-assessment process helps in the establishment of transportation and logistics strategy, which can result in an annual
cost savings of 5–15%. Kersten (2012) demonstrated the different risk factors faced by the trucking industry and recommended the
risk mitigation strategy to control them. However, the execution of these mitigation strategies and their economic implications have
not been considered. A recent report by Blecker et al. (2015) summarized the risks related to the road freight transportation services.
Their study emphasizes that identification of risk factors and their interconnections are important steps in decision-making during
designing and implementation of mitigation strategies. To improve the performance of the transportation system amid evolving
vulnerabilities, an effective risk-management process must consider risk identification as an integral part of the risk-management
process (Crainic and Laporte, 2016).
The early research on road freight transportation risk-management was mainly concerned with individual risks (driver’s shortage,
uncertain load information, etc.). To overcome these challenges, researchers have proposed different approaches which are sum-
marized in Section 4.1. There are various studies that use either qualitative or quantitative technique for the design and selection of
risk mitigation strategy. While Diabat et al., (2012) and Shafiee, (2015) proposed a fuzzy based qualitative approach for the selection
of risk mitigation strategy. Aqlan and Lam, (2015a), Zhang and Zuo, (2016) have developed a quantitative technique for the same.
However, Zavadskas et al., (2016) identified that individual approach (either qualitative or quantitative) fails to account for the
interconnections between the risks, mitigation strategies, and their economic consequences. Therefore, this study proposes an in-
tegrated multi-objective optimization approach aided with a qualitative method for selecting a most appropriate set of mitigation
strategy for the trucking industry under a given budget.
An integrated approach allows researchers to combine unique elements in the transport policy making, which seeks to achieve
positive outcomes against multiple policy goals (Hickman et al., 2012; Aqlan and Lam, 2015b). This approach draws more useful
conclusions in the risk management of road freight transport industry in comparison to the other qualitative analysis approaches. An
integrated approach focuses on the necessity of taking action within the financial constraints in the dynamic environment.
Baykasoğlu and Subulan (2016) developed an integrated model using advanced operation research (OR) and an artificial intelligence

38
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

(AI) based approach to capture the realistic scenario in the current dynamic supply chain environment. Under the traditional model,
the transporters are less attentive towards eliminating the environmental risks and deviation from their expected level of perfor-
mance. Development of risk identification and mitigation strategies must move beyond a simplified approach as it seeks to consider
the impact of designing risk-mitigation strategies for the overall disruption of the supply chain (Basole and Bellamy, 2014; Sherali
et al., 2011).
The risks involved in the decision-making process directly affect the feasibility of the implementation of contingency plans for
supply chain disruptions (Kull et al., 2014). This connection serves as a motivation for decision-makers to start focusing on a wider
variety of factors when considering mitigation strategies such as load information, driver shortage, fragmentation, congestion, and
delay. There are very few formal studies where the probability of occurrence and the severity of these risks have been quantitatively
measured to support risk-mitigation of decision-making processes for logistics companies (Sangwan and Liangrokapart, 2015).
The focus of our work is on analyzing the interconnection between risk and mitigation strategies in decision-making under budget
constraints by considering the degree of necessity (defined as the requirement potential of the mitigation strategy for a risk factor)
and degree of feasibility (defined as the cost of implementing an organization policy/strategy) of mitigation strategies which is absent
in the earlier study and thereby, this study tries to bridge this gap in the literature. Hue-Williams (2016) highlighted that taking an
efficient and cost-effective strategic approach to risk can unlock opportunities to gain a competitive advantage.
In summary, two issues are particularly important in designing and implementing risk-mitigation strategies: understanding risks
and their probability of occurrence with economic consequences and understanding the effect of design characteristics (i.e. targeted
level of risk (TRL), probability of occurrence of risk (PO), and implementation cost (IC)) on the implementation and investment stage
of decision-making processes.

3. Methodology

3.1. Model development

Here, the mathematical model is developed incorporating economic cost reduction (ECR), risk reduction level (RRL) as well as the
feasibility of the risk mitigation strategy so that the appropriate set of strategy can be identified. Consideration of all these objectives
together in a single model provides a decisive and helpful insight to risk managers. In other words, the proposed approach can be
used for operational as well as strategic planning to cover the short and long-term risk. The proposed integrated multi-objective
optimization model considers the degree of feasibility and the degree of necessity of the mitigation strategy. Fig. 1 illustrates the
methodology of the proposed approach.
To formulate the mathematical model, a set of indices, parameters and decision variables have been considered, and their no-
tations are represented in Table 1.
Mathematical Model:
I J Q
Max economic cost reduction(Z1) = ∑ ∑ ∑ Xj ∗ (Ci − (Cq + Cj ))
i=1 j=1 q=1 (1)

I J
Rij' ∗ Xj
Max total risk reduction level(Z2) = ∑∑ Ri
i=1 j=1 (2)
I J
Max feasibility(Z3) = ∑ ∑ μijF ∗ Xj
i=1 j=1 (3)

Subject to:
I J
Ri − ∑ ∑ Rij' ∗ Xj ≤ Ri∗
i=1 j=1 (4)
I J
∑ ∑ Xj ∗ μijF > 0
i=1 j=1 (5)
J
∑ Cj ∗Xj ≤ Δ  δ
j=1 (6)
I J
∑ ∑ Xj ∗ μijNec ≥ Ri∗
i=1 j=1 (7)
J
∑ Xj ≥ 1
j=1 (8)

39
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Start Step 1

Field survey, discussion and literature review

Input Model parameters

Formulate the Model

Step 2

Calculate the degree of feasibility, degree of Subjective Criteria

Experts Preference necessity, risk reduction level

Calculate solution for each objective Step 3

Generate possible combinations of mitigations strategy

No
Check for all combinations
with their constraints Reject
individually for all three
objectives

Yes

Calculate the optimal solution with respect to all feasible combinations of

mitigations strategy

Sensitivity analysis and validation Step4

Fig. 1. Methodology of the proposed approach.

Xj ∈ [0, 1]

Objective (1) seeks to maximize the economic cost reduction (ECR) with a more economical and implementable mitigation
strategy. ECR focuses on identifying and reducing business expenses to increase profits, and it starts with the budgeting process. The
manager compares actual results to the budget expectations, and if actual costs are higher than estimated, management takes action
to reduce the trade-off. Objective (2) maximizes the reduction in risk level (RRL), which is standardized by dividing the RRL by the
original value of the risk, and objective (3) is to maximize the feasibility of mitigation strategies for the risks. Constraint (4) ensures
that the risk reduction after implementation of the mitigation strategy (MS) must achieve a minimum targeted level of risks. Con-
straint (5) ensures the feasibility of the mitigation strategy must be a positive non-zero number. Constraint (6) ensures acceptability

40
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Table 1
Notations of the sets, parameters and decision variable.
Set of indices

i Risk set index (i = 1,2,…..I)

j Mitigation strategy set index (j = 1, 2 …J)
q Set of expected loss due to risks (q = 1, 2, 3…Q)
Parameters
Ri Current level of risk i before implementation of the strategy
Rij' Amount of reduction in risk i after implementation of strategy j
Cj Evaluated cost of implementation for mitigation strategy j
Cq Expected loss after implementation of mitigation strategy
Ci Loss due to the occurrence of risk i
Ri∗ Target level of risk i
Δ Dedicated budget for the mitigation strategy
δ Percentage of budget to be spent on mitigation strategy
μijNec Degree of necessity of mitigation strategy j for the risk i

μijF Degree of feasibility of mitigation strategy j for the risk i

Decision variable
Xj
⎧1, if response strategy is selected
⎨
⎩ 0, otherwise

of risk must be based on budgetary and risk preference criteria. Constraint (7) indicates that the degree of necessity for the mitigation
strategy must be greater than the acceptable level of risk. Constraint (8) ensures that summation of the combination of the mitigation
strategies must be greater than or equal to 1.

3.2. Calculation of fuzzy parameters (μijF , μijNec andRij' )

To capture uncertainty in the modeling process, the fuzzy approach has been used to calculate some of the parameter values,
namely, degree of feasibility (μijF ), degree of necessity ( μijNec ), and risk-mitigation matrix (Rij' ). The stepwise procedure for calculation
is given below.

3.2.1. Stepwise procedure

3.2.1.1. Define scale and collect the expert’s opinion. According to Hsu and Chen, (1996), a heterogeneous group of experts and their
subjective opinions provides more accuracy in the decision-making process as weightage for different terms varies with different
experts. These issues have been considered by the authors while aggregating the expert’s opinions to calculate the index of consensus.
This helps to get consensus information and construct the fuzzy judgement matrix for multi-criteria decision making and it is also
supported by Ölçer and Odabaşi (2005). In the present study, data collection covers a heterogeneous group of experts from the
trucking industry. Therefore, we have also adopted same aggregation approach as it can effectively deal with the problem of this
nature. The quality score of an experts has been calculated by considering the following factors: professional position, job tenure,
education and age (Miri Lavasani et al., 2011; Omidvari et al., 2014; Yazdi, et al., 2017). The score ratings of the experts are recorded
in Table 2.
For this purpose, four experts have been interviewed, and based on their score ratings, the weight of each expert’s opinion (Total
individual expert score divided by total score) has been calculated and presented in Table 3.
In order to elicit the expert's opinion on feasibility and necessity of mitigation strategy, the triangular fuzzy number is used and
presented in Table 4.

3.2.1.2. Aggregation of expert opinion into a single judgment. Aggregation of expert’s opinion for feasibility, necessity, and RMM have
been done by using the following equations.

∼ ∼
1. Computing the degree of similarity Suv (Tu, Tv ) is opinion between each pair of expertsEu andEv . According to the consideration
∼ ∼ ∼ ∼
Suv (Tu, Tv ) when A = (a1, a2 , a3) and B = (b1, b2 , b3) are two triangular fuzzy number, then degree of agreement is defining as:
∼ ∼ 1 K
Suv (Tu, Tv ) = 1 −
K
∑k =1 |ak − bk| (9)

2. Next computing the average agreement (AA) degree

1 K ∼ ∼
AA (Eu ) =
K−1
∑u ≠ v Suv (Tu, Tv ))
v=1 (9a)

3. Computing the relative agreement (RA) degree RA (Ek ) of the experts

41
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Table 2
Score rating according to Experts traits.
Item Categorize Score

Professional position Director of operations, transportation 5

Transport Manager 4
Transport Supervisor 3
Technical 2
Worker 1

Job tenure More than 30 years 5

20–29 4
10–19 3
6–9 2
≤5 1

Education PhD 5
Master 4
Bachelor 3
Higher national diploma 2
School level 1

Age More than 60 years 5

50–60 4
40–49 3
30–39 2
Less than 30 1

Table 3
Experts weight for group decision making.
Expert Position Education Job tenure Age Weight Score

Designation (Score) Qualification (Score) Year(Score) Year (Score)

Expert 1 Director of operations, transportation (5) Master (4) 7 (2) 35 (2) 0.26
Expert 2 Chief Executive Officer (4) Master (4) 12 (3) 45 (3) 0.28
Expert 3 Transport Manager (4) Bachelor (3) 6 (2) 32 (2) 0.22
Expert 4 Transport Supervisor (3) Graduation (3) 15 (3) 42 (3) 0.24

Table 4
Scale for the mitigation strategy selection.
Feasibility Necessity

Ranking Description Definition Description Definition

(0.9, 1.0, 1.0) Highly feasible Practically very much possible VH: Dire needed Needed to a great extent
(0.7, 0.8, 0.9) Feasible Capable of being perceived by the senses H: Much needed Required because it is essential
(0.5, 0.6, 0.7) Moderately feasible It can be done or achieved M: Obligatory Required as a part of routing or course of action
(0.3, 0.4, 0.5) Executable Practically very difficult to bring in to the action L: Ancillary Not much needed
(0.1, 0.1, 0.2) Nearly infeasible Impracticable/not being doable VL: Not necessary Very less importance

AA (Eu )
Eu (u = 1, 2, ⋯..K )asRA (Eu ) = K
∑u = 1 AA (Eu ) (9b)

4. Estimate the consensus coefficient (CC) degree, CC (Eu ) of the experts

CC (Eu ) = β . W (Eu ) + (1 − β ). RA (Eu ) (9c)
where W (Eu ) is the weight of each expert and β (0 ≤ β ≤ 1) is the weight of each term,
Finally, the aggregated result of expert’s judgment
∼ ∼ ∼ ∼
RAG = CC (E1) ⊗ R1 ⊕ CC (E1) ⊗ R1⋯⋯..⊕CC (Em) ⊗ RM (9d)
∼
Then, defuzzification of triangular fuzzy number A = (a1, a2 , a3) is performed by the following method:
1
DDefuzzification = (a1 + a2 + a3)
3 (9e)
Step. 3 Calculation of degree of feasibility, necessity, and development of RMM as follows. Fig. 2 represents the ordinate of the

42
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Fig. 2. Intersection between M1 and M2 .

highest intersection points between fuzzy membership function.

∼ ∼
V (M2 ≥ M1) = SUP
x ≥ y [min(μ M1 (x ), μ M2 (y )]

⎧ 1, ifm2 ≥ m1
⎪ 0, ifl2 ≥ u2
ˇ2 ≥ M
V (M ˇ 1) =
⎨ l2 − u2
⎪ (m2 − u2) − (m1 − l1 ) , otherwise
⎩ (10)

4. Solution and results

The challenge in solving multi-objective optimization problems is the increase in complexity with an increase in number of risks
and strategies where several criteria need to be identified, filtered and minimized. Therefore, we have converted the multi-objective
optimization problem into a single objective problem using global criterion method. This method is appropriate for minimizing the
total relative deviation by providing weights (wk, wherek ∈ 1, 2, 3) to these objectives as presented in Eq. (11).

Zp∗ − Zo ⎞
Min(Z) = ∑ wk ⎛⎜ ∗ ⎟
p ∈ 1,2,3 ⎝ Zp ⎠
k ∈ 1,2,3
o ∈ 1,2,3 (11)
In the proposed model if the number of risks or/and strategies increase, the traditional solution techniques become difficult and
complex (Nooraie and Parast, 2015). For such problems, development of model-based algorithm is more feasible, which provides a
near optimal solution with feasible points (Narenji et al., 2011). A model-based algorithm is better for solving such problems as it
helps in capturing the complexities of such cases through black-box evaluation (Hale and Zhou, 2015). Therefore, we have proposed a
model-based algorithm, (the detail steps are presented in Appendix A.1) to solve the proposed model. The mathematical model is
solved using MATLAB R2017 on a 64-bit Intel i5-4570 computer with 3.20 GHz CPU and 4G RAM.

4.1. Computational results

The proposed approach is validated by applying in the Indian trucking industry as it is vulnerable to the several types of risks
mentioned earlier in this paper. The experts were provided with several risks identified from the literature survey to assist them while
responding. We interviewed a qualified panel of experts and practitioners in order to obtain the data of risk factors and its mitigation
strategy which is provided in Tables 5 and 6 respectively. The probability of occurrence and the estimated economic impact in terms
of monetary value (INR) was given by the panelist based on their experience and past historical data. Finally, the expected economic

Table 5
List of risks estimated probability of occurrence and impacts and expected economic loss.
ID Risk name Probability of occurrence (PO) Impact (INR) Expected economic loss (INR) (PO × Impact)

R1 Bribery 0.416 642,857 267428.5

R2 Shortage of drivers 0.750 1,735,336 1,272,001
R3 Underutilization (Empty running, overloading) 0.683 53,571 41624.67
R4 Fragmentation 0.777 300,000 225,000
R5 Tax evasion, regulatory constraint 0.815 32,517 25460.81
R6 Detention of vehicles 0.733 900,000 733,500
R7 Unfavourable rise in fuel costs 0.783 77915.19 63500.88
R8 Uncertainty in load information 0.733 175,968 128984.5
R9 Delay 0.783 1,040,000 814,320
R10 Congestion (Increase in Number of vehicles) 0.815 254,821 174042.7

43
 K.K. Dadsena et al.

Table 6
Risk mitigation strategies.
ID Proposed risk mitigation strategy Risk Aggregated Impact (INR) Expected Cost of Main objective Benefits under normal condition
Covered response effects implementation (INR)

A1 Training and Skill development program Cantor R2, R9 0.602 267428.5 160991.96 To improve the driving skill Controlling the GHG emission
et al. (2010), De Vries et al. (2017), Ho et al. (2015)
A2 Collaboration among owners Caris et al. (2013), R4, R3 0.594 225,000 133,830 To meet customer requirement Collaboration among the carriers may lead to
Chen et al. (2013), Ergun et al. (2007), Rajapakshe by collaboration among low operational risk and increase the benefit of
et al. (2014) different category of owners the owners
A3 Use of advanced technology (ICT) Dillon et al. R6, R10 0.597 174042.7 104042.75 Improvement of demand Smart and connected transport networks and it
(2003), Perego et al. (2011), Yoshimoto and Nemoto management capability also have a great potential GHG emission
(2005) reduction by improving the operational
efficiency
A4 Dynamic assortment planning Tang (2006) R9 0.655 41624.67 27285.53 Improves the fleet management To reduce fragmentation in the industry
A5 Development of a database Gunasekaran et al. R6, R8 733,500 331510.28 Can help in minimizing empty Helps in scheduling of trucks based on best
(2017), Zolfagharinia and Haughton (2014) running possible way considering return loads

44
A6 Hedge against the volatility of increase in operating R7 0.606 128984.5 78267.82 To avoid the loss or sharp profit Policy against fluctuating fuel price
cost Parikh and Khedkar (2013), Srivastava (2006) margin
A7 Reducing the variability of lead time by strategic R8 0.547 814,320 445494.66 Improves the fleet management To reduce fragmentation in the industry
Stocking Dullaert and Zamparini (2013), Jula et al.
(2005), Üster and Kewcharoenwong (2011)
A8 Family-friendly Scheduling for drivers (Improving R2, 0.400 1,272,001 508800.5 To solve the problem of driver Minimum number of idle truck due to driver
the working conditions and maintaining the staff shortage shortage
requirement) Goel and Irnich (2016), Üster and
Kewcharoenwong (2011)
A9 Improving the milestone management rules and R1, R5 0.600 25460.81 15276.49 To control the illegal money Can reduce CO2 emissions by 4% and also helps
regulations Dadsena et al. (2016), Parikh and collection, maintain, and in improving the operator productivity
Khedkar (2013) analyze roadway assets
A10 Modularity Piao et al. (2010) R10, R9 0.652 63500.88 41435.18 To handle this increasingly Apart from the monetary benefits, the routing
complex technology system ensures that the wastage of fuel on
repeated and redundant runs is curbed to
decrease emission and time
Transportation Research Part A 121 (2019) 37–55
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Table 7
The optimal solution for selection of risk mitigation strategy under varying budget investment.
Budget in INR (×104) Functional Values different objectives Set of strategy

Economic cost reduction in INR (x10^4) Risk reduction level Feasibility Multi-objective

40 24.413 40.9005 25.835 0.0000 0 1 1 0 0 1 0 0 1 1

50 49.929 45.6439 32.01 0.1126 0 0 1 0 1 1 0 0 1 1
60 56.932 49.1995 31.799 0.1252 0 0 0 1 1 1 0 0 1 1
70 88.391 55.5066 37.42 0.2043 0 1 1 1 1 1 0 0 1 0
80 94.487 55.5066 38.05 0.1265 1 0 1 1 1 1 0 0 1 1
90 102.1703 62.3517 42.121 0.0998 1 1 1 1 1 1 0 0 1 1
100 124.82 62.3517 42.121 0.1290 1 0 1 1 0 1 0 1 1 1

cost is calculated by multiplying the probability of occurrence and the corresponding impact.
Moreover, experts were also asked to provide appropriate risk-mitigation strategies along with their estimated implementation
costs which are presented in Table 6 and the values were validated through the published report (Mitra, 2009). Similar to the risk
data, the expected cost of implementation is the product of aggregated response of the experts on the mitigation strategies and their
monetary impact. The detail calculation of aggregated response of the experts is presented in Appendix B.

4.2. Impact of budget allocation on objectives values

We observed that one of the major challenges faced by the current logistic managers is to find the optimal budget allocation for
selecting the most suitable set of mitigation strategies. The response of objectives value for all individual as well as value for the
multi-objective is calculated with the varying budget as shown in Table 7. While solving the model, feasibility of the solution was
checked with an initial budget investment of INR 1 lakh. It has been observed that the solution is infeasible up to INR 3 lakhs while
considering at an incremental step of INR 1 lakh. However, it is found that from INR 4 lakhs onwards, the optimal solution of each
objective becomes feasible. The graphical representation of functional values for all three objectives is shown in Fig. 3.
In this figure, budget investment is plotted on the x-axis with functional values on the y-axis. All three objectives have different
budget levels between INR 4 to 10 lakhs within which the optimal solution of each objective becomes feasible. Fig. 3 also shows that
the functional values for risk reduction and feasibility have optimal value at INR 9 lakhs with mitigation strategy set
[1 1 1 1 1 1 0 0 1 1]. However, for ECR, INR 9 lakhs has an optimal value with mitigation strategy set [1 1 0 0 0 1 0 1 1 0], which
includes an ROI of 13.52% where ROI = (Economic cost reduction − Budget allocated)/Budget allocated. Further, our analysis
provides experimental results based on the TRL, PO, and IC considering the same criteria, which varies depending on the situation
and impact of losses.
From Fig. 3, the minimum budget level required for a feasible set of mitigation strategies considering all objectives to sustain the
operations can be observed. Clearly, these budget values represent the minimum budget investment required to obtain a feasible set
of solutions.

4.3. Optimal range of budget investment considering ROI

The variation of budget investments with their ROI is shown in Fig. 4.

The figure shows that for a given budget of INR 7 lakhs, the highest ROI is 27% with the strategy set [0 1 1 1 1 1 0 0 1 0]. The
analysis of the results also shows several interesting and counter-intuitive results such as at INR 9 lakhs budget investment, one

Economic cost Reduction in INR (x104) Risk Reduction Level Feasibility

200
175
FUNCTION VALUE

150
125
100
75
50
25
0
4 5 6 7 8 9 10
BUDGET IN LAKHS (INR)
Fig. 3. Impact of budget allocation on functional value.

45
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

30

20

10

ROI (%) 0
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
-10
BUDGET IN LAKHS (INR)
-20

-30

-40

-50
Fig. 4. The impact of budget investment on ROI in strategic decision-making.

obtains an ROI of 14% with strategy set [1 1 1 1 1 1 0 0 1 1]; however, at INR 10 lakhs, the ROI is 25% with strategy set
[1 0 1 1 0 1 0 1 1 1]. These insights will help managers to decide which outcome would satisfy their targeted requirement. As for
example, if strategy eight (A8) is essential, then he must go for 10 lakhs of investment. Similarly, the results can help the manager to
choose the appropriate outcome considering the available budget for risk-mitigation strategies. This analysis is very useful for the
decision-maker to select the most suitable investment policy with an eye on achieving the higher ROI. Investing more than the
sufficient level of budget does not improve the ROI.

5. Sensitivity analysis and managerial implication

Here, sensitivity analysis is conducted for certain key parameters relevant for the formulation of an optimal set of strategies,
including PO, IC as well as changes in TRL. Sensitivity analyses allow decision makers in better understanding of responses to changes
in environmental conditions (Tokar, 2010). Giannakis and Papadopoulos (2016) suggested that controlled (experimental) case
studies can be conducted to ensure the internal validity and reliability of risk management process. In this paper, the response of all
the three objectives (ECR, RRL, and feasibility) is analyzed with a small deviation from the value shown in Table 7. It is worth noting
that the variation in TRL, PO, IC, and budget allocation impacts the design of decisions and strategic planning. The experimental
design is developed and performed to analyze the robustness of the proposed study. The individual parameters are varied by ± 10%
with budget allocation keeping the other parameters constant. The ROI is calculated with respect to individual budget allocation.
Problem instances are obtained by combining:

(i) Seven possible scenarios for budget allocation, which varies between INR 4 to 10 lakhs.
(ii) Three possible scenarios for the TRL and one for ROI.
(iii) Three possible scenarios for IC and one for ROI.
(iv) Three possible scenarios for PO of risk values and one for ROI.

A total of 448 experiments were performed in this computational setup. The main objective of the numerical analysis and
experiment is to present useful data to managers for effective, strategic, and operational decision-making in the design and im-
plementation phase. The optimization model is solved for different scenarios which we named as Scenario I (Actual obtained values),
Scenario II (10% decrease from Actual obtained values), and Scenario III (10% increase from Actual obtained values).

5.1. How does functional values change with budget for varying occurrence probability of risk while TRL and IC are fixed?

Identifying the probability of occurrence of a risk is the most crucial step in its mitigation. Once a risk is identified, the managers
can think ways to mitigate it and allocate budget for this purpose. However, the PO can change depending on various operational
factors. For example, as shown in Table 5, the driver shortage is one of the risks identified, whose probability of occurrence can vary
with market demand. Similarly, the occurrence of risk due to under-utilization of trucks will vary with an imbalance in demand and
supply. So, it becomes imperative for the managers to be prepared for these variations in PO of any risk and spend money judiciously
on its mitigation. To address this situation, we analyze the effect of the varying occurrence probability of risk on budget investment
policy, which is shown in Fig. 5. It can be observed that the functional values of RRL and feasibility show almost similar values for all
scenarios with budget investment at INR 9 lakhs.
Fig. 5 provides the manager necessary information to decide what amount he should invest based on the indispensability of the
mitigation strategy. For scenario III, the limit is INR 10 lakhs. However, scenarios I and II show an almost linear increase: INR 7 lakhs
with strategy sets [0 1 1 1 1 1 0 0 1 0] and [1 1 1 1 0 1 0 0 1 1] with ROI 27% and 57% respectively and 10 lakhs with strategy sets
[1 0 1 1 0 1 0 1 1 1] and [0 1 0 1 0 1 0 1 1 1] with 25% and 54% of ROI respectively. These analyses reveal how the occurrence
probability affects the cost of implementation of the budget which gives very useful insights to the managers.

46
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

(SCENARIO I) (SCENARIO II)

Economic Cost Reduction in INR (x104) Economic Cost Reduction in INR (x104)
Risk Reduction Level Risk Reduction Level
Feasibility Feasibility
200 200
FUNCTIONAL VALUE

FUNCTIONAL VALUE
175 175
150 150
125 125
100 100
75 75
50 50
25 25
0 0
4 5 6 7 8 9 10 4 5 6 7 8 9 10
BUDGET IN LAKHS (INR) BUDGET IN LAKHS (INR)

(SCENARIO III) Effect of occurence probability

Economic Cost Reduction in INR (x104) of risk on ROI
Scenario I Scenario II Scenario III
Risk Reduction Level
Feasibility 80
60
FUNCTIONAL VALUE

200
175
ROI (%)

40
150
125 20
100
75 0
50 4 5 6 7 8 9 10
-20
25
0 -40
4 5 6 7 8 9 10
-60
BUDGET IN LAKHS (INR) BUDGET IN LAKHS (INR)
Fig. 5. Behavior of ECR, RRL, and feasibility of mitigation strategy with varying budget allocation and (PO). Note: Scenario I (Obtained values for
PO as shown in Table 5), Scenario II (when PO decreases 10% from obtained values), and Scenario III (when PO increases 10% from obtained
values).

5.2. How does objective value change with budget for different target levels of risk while PO and IC are fixed?

The managers usually face a dilemma about the amount of money to invest for mitigation of a risk. Without proper economic
analysis of the mitigation strategy, they might spend more money than required. After identifying the probability of occurrence of any
risk it will be beneficial for the managers if they are able to vary the TRL and check if they are spending the optimum amount of
money on its strategy. Therefore, in this section, we have analyzed the functional values for varying levels of target risk keeping PO
and IC constant for different budget values as shown in Fig. 6.
The budget can be seen to decrease in the same manner with a decrease in the targeted level of risk. The scenario I and scenario II
show similar trends of functional values with an optimal ROI of 26% at INR 9. However, the feasibility and risk level reduction is
almost same for scenario III and shows higher ROI of 33% at INR 9 lakhs. So, from the above analysis, the managers can make smooth
decisions according to their prerequisite strategies and allocated budget.

5.3. How does functional values change with a budget for varying cost of implementation while PO and TRL are fixed?

Once a manager has identified the PO of any risk and selected its TRL, the most crucial decision which he/she must make is that of
the amount of money to be spent on its mitigation. As he has a limited budget to tackle all the risks, spending the optimum amount of
money for all the risks is essential for him. Therefore, it becomes important for the manager to check if he/she is spending more
money than required for any risk. To analyze this situation, we studied the effect of varying IC keeping PO and TRL constant on all
functional values for different budget investments. The graph of these variations is shown in Fig. 7. It is easy to observe from the

47
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

(SCENARIO I) (SCENARIO II)

Economic Cost Reduction in INR (x104) Economic Cost Reduction in INR (x104)

Risk Reduction Level Risk Reduction Level

Feasibility Feasibility
FUNCTIONAL VALUE

200 200

FUNCTION VALUE
175 175
150 150
125 125
100 100
75 75
50 50
25 25
0 0
4 5 6 7 8 9 10 4 5 6 7 8 9 10

BUDGET IN LAKHS (INR) BUDGET IN LAKHS (INR)

(SCENARIO III) Effect of targeted risk level on

Economic Cost Reduction (x104) ROI
Scenario I Scenario II
Risk Reduction Level
Scenario III
Feasibility
40
FUNCTIONAL VALUE

200 30
175 20
150
125 10
100 0
ROI

75 -10 4 5 6 7 8 9 10
50 -20
25 -30
0
4 5 6 7 8 9 10
-40
-50
BUDGET (INR) BUDGET IN LAKHS (INR)
Fig. 6. Behavior of ECR, risk reduction level (RRL) and feasibility of mitigation strategy with varying budget allocation and target risk level (TRL).
Note: Scenario I (Obtained values for TRL as shown in Table 5), Scenario II (when TRL decreases 10% from obtained values), and Scenario III (when
TRL increases 10% from obtained values).

figure that the feasibility and RRL show slight variation for all the scenarios and that the optimal budget investment value is INR 9
lakhs. However, the ECR shows a noticeable increase from budget INR 9 to 10 lakhs. From the ROI point of view, scenario II is not
acceptable to the decision-maker.
However, for the scenario I and scenario III, the optimal ECR is INR 10 lakhs. It is obvious that scenario III gives the higher ROI
because the cost of strategy implementation is less: INR 8 lakhs, 89% of ROI with strategy set [1 1 1 1 1 1 0 0 1 0]. For scenario I, the
ideal cost was INR 7 lakhs with ROI of 27% with strategy set [0 1 1 1 1 1 1 0 0 1 0]. Interestingly, an investment of INR 10 lakhs for
scenario III shows ROI 71% with strategy set [1 1 1 1 0 1 0 1 1 1]. Therefore, from this data, the decision-maker can easily decide
which individual scenario will be the best budget investment policy. This result can also be used for reverse analysis of what can be
the optimal implementation cost for the organization with respect to the total budget investment policy.

6. The model analysis and validation

Several risk managers have attempted to design risk-mitigation approaches to generate higher ROI. However, in the trucking
industry, it is very difficult to change the risk-mitigation process by considering factors like TRL, PO, and IC, which all influence the
selection of a risk-mitigation strategy with maximum ROI.
To test the proposed model, a demanding experiment was conducted to direct the likelihood of selecting strategies based on a set
of risk-mitigation strategies and risk of the operational process. This method best isolated the effects of how the risk manager
combines risk-mitigation strategies. To validate the proposed model, the experiment is designed using the most effective trends and
removed some of the informational redundancies which may occur in the natural environment. To better understand how these
model parameters support design decisions in complex decision-making problem a statistical method design of experiment (DOE) can
be used (MacCalman et al., 2015). DOE permit systems engineers to efficiently explore the design space in order to recognize the most
significant design drivers, their impact, the interactions between them and their increasing or decreasing rates of change (Padhi et al.,
2016; Padhi et al., 2013; Pennington and Tuttle, 2007). Designing risk-mitigation approach parameters involve economic and

48
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

(SCENARIO I)
(SCENARIO II)
Economic Cost Reduction in INR (x104) Economic Cost Reduction in INR (x104)
Risk Reduction Level Risk Reduction Level
Feasibility Feasibility

FUNCTIONAL VALUE
200 200
FUNCTIONAL VALUE

175 175
150 150
125 125
100 100
75 75
50 50
25 25
0 0
4 5 6 7 8 9 10 4 5 6 7 8 9 10

BUDGET IN LAKHS (INR) BUDGET IN LAKHS (INR)

(SCENARIO III) Effect of implementation cost

on ROI
Economic Cost Reduction in INR (x104)
Scenario I Scenario II
Risk Reduction Level
Scenario III
Feasibility
100
FUNCTIONAL VALUE

200
175
150 50
125
ROI (%)

100
75 0
50 4 5 6 7 8 9 10
25 -50
0
4 5 6 7 8 9 10
-100
BUDGET IN LAKHS (INR) BUDGET IN LAKHS (INR)
Fig. 7. Behavior of ECR, RRL, and feasibility of mitigation strategy with varying budget allocation and implementation cost (IC). Note: Scenario I
(Obtained values for IC as shown in Table 6), Scenario II (when IC decreases 10% from obtained values), and Scenario III (when IC increases 10%
from obtained values).

strategic factors. In such an environment, the risk manager wishes to modify the values of parameters that influence the ROI and
provide the best combination of optimal RRL and maximum feasibility of the strategy. Furthermore, we tested and validated the
proposed model using DOE.
For instance, to analyze and validate the quantitative model, we changed different model parameter values to assess the para-
meter sensitivity, using three different scenarios with diverse levels, where for both Scenario I (TRL) and scenario II (IC) we have
considered only two (low and actual) levels, and for scenario III (OP) three (low, actual, and high) levels. The model parameter values
were selected based on optimization results. The experiment consisted of a mixed design with three factors that were manipulated in
a 2 × 2 × 3 factorial design as shown in Table 8.
An analysis of ANOVA for ROI (ECR) was performed, and the effects of individual factors as well as interaction are shown in
Table 9. The test statistic shows a model F value of 15.5953 less than 0.00001, at the 5% level of significance (α of 0.05).
Since the test statistic is much lower than the critical value, we accept the hypothesis and conclude that our model is statistically
significant. Analysis of the factors indicates that the loss due to varying PO (Model A) and IC of the mitigation strategies (Model C) are
significant at the 5% level of significance.

7. Conclusion and future scope

Managing risk successfully has always been a difficult task for almost all successful businesses. The need for strategic risk-
management methods that consider economic consequences has been mentioned in recent reports, trade magazines, and research
articles. Nevertheless, there have been no comprehensive models aimed at understanding the selection of risk-mitigation strategies
for the trucking industry that consider both subjective and objective parameters. This study focuses on the design, development, and
implementation of risk-mitigation strategies under budget constraint in the operational process of the trucking industry using an
integrated approach. Our study has confirmed the interconnection between risk intensity and its costly nature in the mitigation

49
 K.K. Dadsena et al.

Table 8
Influence of factors and parameter values.
Factors Parameter values

Scenario I (Coded value: 0) Scenario II (Coded value: +1) Scenario III (Coded value: −1)

50
A Loss due to Varying PO [267428.5, 225000, 174042.7, 41624.67, 733500, 128984.5, [331714, 255000, 199525, 46981.8, 823500, 146581, [203142, 195000, 148561, 36267.6, 643500, 111388, 710320,
(INR) 814320, 1272001, 25460.81, 63500.88] 918320, 1445535, 28712.5, 71292.4] 1098648, 22209.1, 71292.4]
B TRL [0.20 0.18 0.15 0.18 0.24 0.21 0.20 0.24 0.24 0.20] [0.10 0.08 0.05 0.08 0.14 0.11 0.10 0.14 0.14 0.1]
C IC (INR) [160992.05, 133830, 104042.8, 27285.53, 331510.3, 78267.82, [134249.1, 111150, 86499.22, 23101.69, 257458.5, 5266.16,
445494.7, 508800.5, 15276.49, 41435.18] 364001.1, 381600, 12730.30, 35052.50 ]
Transportation Research Part A 121 (2019) 37–55
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Table 9
The result of ANOVA of TRL, PO, and IC on the ROI.
Sum of square Degree of freedom Mean Square F0 Prob > F

Model 5.610E+12 11 5.100E+11 15.5953 < 0.00001 The result is significant at p < .05.
A 2.648E+12 2 1.324E+12 40.4923 < 0.00001 The result is significant at p < .05
B 3.294E+10 1 3.294E+10 1.0050 0.3228 Not Significant
C 2.888E+12 1 2.888E+12 88.3306 < 0.00001 The result is significant at p < .05
AB 1.483E+08 2 0.741 E+08 0.0022 0.99780 Not Significant
AC 3.970E+10 2 1.986E+10 0.6064 0.5507 Not Significant
BC 3.475E+08 1 3.469 E+08 0.0106 0.9185 Not Significant
ABC 0.271 E+08 2 0.135 E+08 0.0004 0.9996 Not Significant
SSerror 1.177E+12 36 3.27E+10
SSTotal 6.787E+12 47

process.
The model-based algorithm helps to examine and evaluate a multi-objective model when complexity increases with the addition
of variables. The model helps in better understanding of the budget allocation considering risk impact in designing and development
of risk management process. The results of the study provide a specific call to risk managers to recognize the roles of risk reduction
and economic factors in the implementation phase. In addition, the optimal range of budget investment based on the maximum ROI is
found out for the requisite set of strategy. This study gives direction to road freight transporters for efficient and effective operational
design of their risk-management activities. The validation of the proposed model has been illustrated by a case study of the Indian
trucking industry.
Results show that the probability of risk occurrence and implementation cost of mitigation strategy has a significant impact on
design and development during the decision-making process. Results show theoretical as well as computational insights in under-
standing risk intensity and its costly nature in the risk-mitigation process. An experimental analysis has demonstrated that an in-
terdisciplinary, an integrated technique provides holistic data that strengthens the proposed model’s claim to represent a more
realistic scenario.
This study is subject to a few limitations, each offering opportunities for future research. First, despite our efforts to reduce
heterogeneity of expert’s opinion, we acknowledge that a large number of experts may help to leverage preference diversity and
perceiving individual’s heterogeneity in the decision-making process. Second, the performance of the proposed algorithm can be
tested if the number of risks and strategies increase. Third, current study considers the degree of necessity and feasibility calculated
from subjective opinion. Additionally, one can explore the degree of acceptability, sustainability, and reliability issues in the future
work. Further, the present model is developed and solved considering risk mitigation strategies only and contingency planning which
is a post occurrence response of an event can be considered as a future scope of the study.

Appendix A.1

Algorithm. The solution steps of the developed algorithm are described below:

Step 1. Initialize coefficient matrix; Rij' , μijNec , μijF , Ri , Cj, Cq, Ci, Ri∗
Declare objective functionZ1, Z2, Z3, Z;
Declare constraints; Constraint 1 to Constraint 10;
Declare comb_Z1, comb_Z2 , comb_Z3 , comb_Z;
Step 2. For j = 1 to 1024
For Constraint 1 to Constraint 10
If all constraints are satisfied for Z1
Store the combinations at variable comb_Z1;
End If
If all constraints are satisfied for Z2
Store the combinations at variable comb_Z2 ;
End If
If all constraints are satisfied for Z3
Store the combinations at variable comb_Z3 ;
End If
End loop
End loop
Step3:
For n = 1 to n = length (comb_ Z1)
If value of Z1is maximum
Store the value at Z1;
End If
End loop
For n = 1 to n = length (comb_ Z2 )
If value of Z2 is maximum

51
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Step 1. Initialize coefficient matrix; Rij' , μijNec , μijF , Ri , Cj, Cq, Ci, Ri∗
Store the value at Z2 ;
End If
End loop
For n = 1 to n = length (comb_ Z3 )
If value of Z3 is maximum
Store the value atZ3 ;
End If
End loop
Step 4:
∗
Z − Z1 ∗
Z − Z2 3 Z∗ − Z
Setz = (w1) ∗ ⎛ 1 ∗ ⎞ + (w 2) ∗ ⎛ 2 ∗ ⎞ + (w3) ∗ ⎛ 3 ∗ ⎞ ; ⎜ ⎟

⎝ Z1 ⎠ ⎝ Z2 ⎠ ⎝ Z3 ⎠
For j = 1 to 1024
For Constraint 1 to Constraint 10
If all constraints are satisfied for Z
Store the combinations at variable comb_Z ;
End If
End loop
End loop
Step 5
For n = 1 to n = length (comb_Z)
If value of Z is minimum
Store the value at Z;
End If
End loop

Repeat the above steps for different percentage of budget.

Appendix B

(see Tables B1 and B2).

Table B1
Linguistic Assessment for calculation of RMM, Feasibility, and Necessity.
Experts Weight-age (R1) (R2) (R3) (R4) (R5) (R6) (R7) (R8) (R9) (R10)

Risk factors E1 0.26 M VH H VH M H H VH H M

E2 0.28 H H H VH H H H H VH H
E3 0.22 M H VH H H H M H H M
E4 0.24 H H VH VH M M H VH H H

Aggregated (0.665 (0.662 (0.742 (0.501 (0.600 (0.730 (0.663 (0.506 (0.505 (0.669
value 0.765 0.763 0.843 0.634 0.700 0.830 0.763 0.637 0.637 0.769
0.865) 0.863) 0.943) 0.767) 0.800) 0.930) 0.823) 0.768) 0.768) 0.869)

Exper-ts Weight-age (M1) (M2) (M3) (M4) (M5) (M6) (M7) (M8) (M9) (M10)

Strategy E1 0.26 M M H H M H M M H H
E2 0.28 H M M H L H M L M H
E3 0.22 M H H M M M M L M H
E4 0.24 H H M H M M H M H M

Aggregated (0.495 (0.600 (0.669 (0.504 (0.662 (0.496 (0.665 (0.505 (0.729 (0.734
value 0.630 0.700 0.769 0.636 0.762 0.631 0.765 0.636 0.829 0.834
0.765) 0.800) 0.869) 0.768) 0.862) 0.765) 0.865) 0.768) 0.929) 0.934)
Feasibility E1 0.26 M VH VH H VH H H H VH VH
E2 0.28 H H H M H H M M H VH
E3 0.22 M M H H VH VH M H H H
E4 0.24 M H H M H VH M M M H

Aggregated (0.50, (0.606, (0.650, (0.496, (0.696, (0.693, (0.450, (0.497, (0.603, (0.706,
value 0.601, 0.705, 0.750, 0.596, 0.796, 0.792, 0.550, 0.596, 0.703, 0.806,
0.702) 0.806) 0.850) 0.696) 0.896) 0.892) 0.650) 0.697) 0.801) 0.906)
Necessity E1 0.26 VH VH VH VH VH VH H H H VH
E2 0.28 H VH H H VH VH M H H VH
E3 0.22 H H VH H VH H M H VH H
E4 0.24 H H H VH H VH M M H H

Aggregated (0.65, (0.707, (0.695, (0.700, (0.752, (0.755, (0.450, (0.552, (0.644, (0.706,
value 0.751, 0.805, 0.796, 0.800, 0.852, 0.855, 0.550, 0.652, 0.744, 0.806,
0.852) 0.904) 0.985) 0.900) 0.952) 0.955) 0.650) 0.752) 0.844) 0.906)

52
 K.K. Dadsena et al.

Table B2
Degree of feasibility, degree of necessity and RMM.
Feasibility Coefficient Degree of Necessity Coefficient RMM

R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R1 R2 R3 R4 R5 R6 R7 R8 R9 R10

(M1) [[0.350 1.000 1.000 1.000 0.550 0.550 0.580 0.200 0.250 0.320] [0.500 0.500 0.456 0.380 0.500 0.500 0.500 0.500 0.500 0.800] [0.330 1.000 0.230 0.600 0.600 0.575 0.580 0.100 1.000 0.350]
(M2) [0.000 0.500 0.321 0.600 0.550 0.000 0.600 0.500 0.500 0.500] [0.500 0.700 0.500 1.000 0.380 0.550 0.380 1.000 0.900 0.500] [0.310 1.000 1.000 1.000 0.610 0.570 0.350 0.580 1.000 0.300]
(M3) [0.500 0.000 0.480 0.430 0.500 0.500 0.500 0.500 1.000 1.000] [0.500 0.300 0.540 1.000 0.400 0.585 0.420 0.700 1.000 0.500] [0.330 0.800 1.000 1.000 0.620 0.570 0.580 1.000 1.000 0.320]

53
(M4) [0.350 0.110 1.000 1.000 0.600 0.575 0.580 1.000 0.250 0.320] [0.500 0.487 0.500 1.000 0.380 0.575 0.400 1.000 0.500 0.500] [0.500 0.350 1.000 0.510 0.280 0.500 0.350 1.000 0.380 0.480]
(M5) [0.500 0.500 0.500 0.310 0.410 0.580 0.400 0.501 0.500 0.500] [0.500 0.500 0.120 0.250 1.000 0.380 0.275 0.500 0.500 0.480] [1.000 0.000 0.000 0.480 1.000 1.000 0.280 0.000 0.000 1.000]
(M6) [0.500 0.900 0.500 0.280 0.397 0.578 0.400 0.480 0.320 0.500] [0.480 0.500 0.260 0.200 0.275 0.350 1.000 0.500 0.500 0.475] [0.350 0.000 1.000 1.000 0.600 0.550 0.580 1.000 0.000 0.350]
(M7) [0.250 0.125 1.000 0.500 0.500 0.380 0.500 1.000 0.250 0.250] [0.250 0.280 1.000 0.500 0.500 0.380 0.500 1.000 1.000 0.250] [0.300 0.000 1.000 0.500 0.480 0.380 0.250 1.000 1.000 0.250]
(M8) [0.325 1.000 1.000 1.000 0.597 0.570 0.600 0.980 0.250 0.320] [0.500 0.350 0.250 0.512 1.000 0.500 1.000 0.350 0.300 0.478] [0.000 1.000 0.000 0.250 1.000 0.000 1.000 0.000 1.000 0.330]
(M9) [0.500 0.500 0.321 0.600 0.600 1.000 0.600 0.500 0.500 0.500] [0.500 0.500 0.450 0.475 0.550 0.500 0.550 0.500 0.500 1.000] [0.350 1.000 1.000 1.000 0.600 0.580 0.580 0.210 0.000 0.380]
(M10) [0.500 0.500 0.500 1.000 0.400 0.580 0.398 0.500 1.000 0.500] [0.500 0.800 0.500 0.220 0.380 0.540 0.380 0.500 1.000 0.500] [0.500 0.350 0.250 0.500 0.250 0.500 1.000 0.300 0.350 0.480]

Note: Targeted risk level (Ri∗) = [0.2 0.18 0.15 0.18 0.24 0.21 0.20 0.24 0.24 0.20].
Transportation Research Part A 121 (2019) 37–55
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Appendix C. Supplementary material

Supplementary data to this article can be found online at https://doi.org/10.1016/j.tra.2019.01.007.

References

American Trucking Association, Inc. available at < http://www.trucking.org/article/Trucking-Industry-Revenues-Were-$676.2-Billion-in-2016 > , (accessed date
December 5, 2017).
Abkowitz, M.D., 2002. Transportation risk management: a new paradigm. Security Papers (Knoxville: Southeastern Transportation Center, University of Tennessee)
6(4), 93–103.
Aqlan, F., Lam, S.S., 2015a. Supply chain risk modelling and mitigation. Int. J. Product. Res. 53 (18), 5640–5656.
Aqlan, F., Lam, S.S., 2015b. A fuzzy-based integrated framework for supply chain risk assessment. Int. J. Product. Econ. 161, 54–63.
Bansal, K., 2015. Truckers’ Strike Enters 9th Day.
Basole, R.C., Bellamy, M.A., 2014. Supply network structure, visibility, and risk diffusion: a computational approach. Decision Sci. 45 (4), 753–789.
Baykasoğlu, A., Subulan, K., 2016. A multi-objective sustainable load planning model for intermodal transportation networks with a real-life application. Transport.
Res. Part E: Logi. Transport. Rev. 95, 207–247.
Biyani, V., 2017. Decoding the Indian transportation industry and the digital opportunities.
Blecker, T., Kersten, W., Ringle, C.M., 2015. Operational excellence in logistics and supply chains. Proceedings of the Hamburg International Conference of Logistics
(HICL). epubli.
Cantor, D.E., Corsi, T.M., Grimm, C.M., Özpolat, K., 2010. A driver focused truck crash prediction model. Transport. Res. Part E: Logis. Transport. Rev. 46 (5),
683–692.
Caris, A., Macharis, C., Janssens, G.K., 2013. Decision support in intermodal transport: a new research agenda. Comput. Indus. 64 (2), 105–112.
Chen, J., Sohal, A.S., Prajogo, D.I., 2013. Supply chain operational risk mitigation: a collaborative approach. Int. J. Product. Res. 51 (7), 2186–2199.
Crainic, T.G., Laporte, G., 2016. Transportation in supply chain management: recent advances and research prospects. Int. J. Product. Res. 54 (2), 403–404.
Dadsena, K.K., Naikan, V., Sarmah, S., 2016. Ranking of factors by using IT2 Fuzzy FMEA for a trucking industry, POMS 27th Annual Conference.
De Vries, J., de Koster, R., Rijsdijk, S., Roy, D., 2017. Determinants of safe and productive truck driving: empirical evidence from long-haul cargo transport. Transport.
Res. Part E: Logis. Transport. Rev. 97, 113–131.
Diabat, A., Govindan, K., Panicker, V.V., 2012. Supply chain risk management and its mitigation in a food industry. Int. J. Product. Res. 50 (11), 3039–3050.
Dillon, R.L., Paté-Cornell, M.E., Guikema, S.D., 2003. Programmatic risk analysis for critical engineering systems under tight resource constraints. Operations Res. 51
(3), 354–370.
Dullaert, W., Zamparini, L., 2013. The impact of lead time reliability in freight transport: a logistics assessment of transport economics findings. Transport. Res. Part E:
Logis. Transport. Rev. 49 (1), 190–200.
Ergun, O., Kuyzu, G., Savelsbergh, M., 2007. Reducing truckload transportation costs through collaboration. Transport. Sci. 41 (2), 206–221.
Giannakis, M., Papadopoulos, T., 2016. Supply chain sustainability: a risk management approach. Int. J. Product. Econ. 171, 455–470.
Goel, A., Irnich, S., 2016. An exact method for vehicle routing and truck driver scheduling problems. Transport. Sci. 51 (2), 737–754.
Gunasekaran, A., Subramanian, N., Papadopoulos, T., 2017. Information technology for competitive advantage within logistics and supply chains: a review. Transport.
Res. Part E: Logis. Transport. Rev. 99, 14–33.
Hale, J.Q., Zhou, E., 2015. A model-based approach to multi-objective optimization. In: Proceedings of the 2015 Winter Simulation Conference. IEEE Press, pp.
3599–3609.
Hickman, R., Saxena, S., Banister, D., Ashiru, O., 2012. Examining transport futures with scenario analysis and MCA. Transport. Res. Part A: Policy Practice 46 (3),
560–575.
Ho, W., Zheng, T., Yildiz, H., Talluri, S., 2015. Supply chain risk management: a literature review. Int. J. Product. Res. 53 (16), 5031–5069.
Hsu, H.-M., Chen, C.-T., 1996. Aggregation of fuzzy opinions under group decision making. Fuzzy sets and systems 79 (3), 279–285.
Hue-Williams, M., 2016. Navigating risk in the transportation sector, Willis Towers Watson, Transportation Risk Index. < https://www.willistowerswatson.com//
media/WTW/PDF/ Insights/2016/09/WTW_ Transportation_Risk.pdf > . (accessed December 5, 2017).
Jula, H., Dessouky, M., Ioannou, P., Chassiakos, A., 2005. Container movement by trucks in metropolitan networks: modeling and optimization. Transport. Res. Part E:
Logis. Transport. Rev. 41 (3), 235–259.
Kersten, W., 2012. Risk management in logisitics: empirical results from the Baltic Sea region from 2010 until 2012. CASH.
Klemick, H., Kopits, E., Wolverton, A., Sargent, K., 2015. Heavy-duty trucking and the energy efficiency paradox: evidence from focus groups and interviews.
Transport. Res. Part A: Policy Practice 77, 154–166.
Kull, T.J., Oke, A., Dooley, K.J., 2014. Supplier selection behavior under uncertainty: contextual and cognitive effects on risk perception and choice. Decision Sci. 45
(3), 467–505.
Lewis, M.A., 2003. Cause, consequence and control: towards a theoretical and practical model of operational risk. J. Operations Manage. 21 (2), 205–224.
Mattsson, L.-G., Jenelius, E., 2015. Vulnerability and resilience of transport systems–a discussion of recent research. Transport. Res. Part A: Policy Practice 81, 16–34.
MacCalman, A., Kwak, H., McDonald, M., Upton, S., 2015. Capturing experimental design insights in support of the model-based system engineering approach.
Procedia Computer Science 44, 315–324.
Miri Lavasani, M.R., Wang, J., Yang, Z., Finlay, J., 2011. Application of fuzzy fault tree analysis on oil and gas offshore pipelines. Int. J. MArine Sci. Eng. 1 (1), 29–42.
Magoci, J., 2016. Reveal How Profitable The Trucking Business Really Is?
Min, H., Lambert, T., 2002. Truck driver shortage revisited. Transport. J. 5–16.
Mitra, S., 2009. A Joint Study Report by TCI and IIM Calcutta.
Narenji, M., Ghomi, S.F., Nooraie, S., 2011. Grouping in decomposition method for multi-item capacitated lot-sizing problem with immediate lost sales and joint and
item-dependent setup cost. Int. J. Syst. Sci. 42 (3), 489–498.
Nooraie, S.V., Parast, M.M., 2015. A multi-objective approach to supply chain risk management: integrating visibility with supply and demand risk. Int. J. Product.
Econ. 161, 192–200.
Numadic, T., 2017. Why are Indian truck drivers poorly paid despite soaring demand and short supply?
Ölçer, A., Odabaşi, A., 2005. A new fuzzy multiple attributive group decision making methodology and its application to propulsion/manoeuvring system selection
problem. Eur. J. Operational Res. 166 (1), 93–114.
Omidvari, M., Lavasani, S.M.R., Mirza, S., 2014. Presenting of failure probability assessment pattern by FTA in Fuzzy logic (case study: distillation tower unit of oil
refinery process). J. Chem. Health Safety 21 (6), 14–22.
Padhi, S.S., Wagner, S.M., Mohapatra, P.K., 2016. Design of auction parameters to reduce the effect of collusion. Decision Sci. 47 (6), 1016–1047.
Padhi, S.S., Wagner, S.M., Niranjan, T.T., Aggarwal, V., 2013. A simulation-based methodology to analyse production line disruptions. Int. J. Product. Res. 51 (6),
1885–1897.
Parikh, J., Khedkar, G., 2013. the impacts of diesel price increases on India’s trucking industry.
Pennington, R., Tuttle, B., 2007. The effects of information overload on software project risk assessment. Decision Sci. 38 (3), 489–526.
Perego, A., Perotti, S., Mangiaracina, R., 2011. ICT for logistics and freight transportation: a literature review and research agenda. Int. J. Phys. Distribut. Logis.
Manage. 41 (5), 457–483.

54
K.K. Dadsena et al. Transportation Research Part A 121 (2019) 37–55

Piao, H., Wang, H., Chu, S., 2010. Modularity design of logistics systems, automation and logistics (ICAL). In: 2010 IEEE International Conference on IEEE, pp.
183–188.
Raghuram, G., 2015. An overview of the trucking sector in india: significance and structure.
Rodrigue, J.P., Comtois, C., Slack, B., 2009. The geography of transport systems. Routledge.
Rajapakshe, T., Dawande, M., Gavirneni, S., Sriskandarajah, C., Panchalavarapu, P.R., 2014. Dedicated transportation subnetworks: design, analysis, and insights.
Product. Operations Manage. 23 (1), 138–159.
Sangwan, S., 2017. The Indian Transportation Market is expected to be worth $300B by 2020’.
Sangwan, T., Liangrokapart, J., 2015. Risk Identification for Outbound Road Freight Transportation Service. Innovations and Strate Logistics and Supply Ch, 391.
Shafiee, M., 2015. A fuzzy analytic network process model to mitigate the risks associated with offshore wind farms. Expert Syst. Appl. 42 (4), 2143–2152.
Sherali, H.D., Dalkiran, E., Glickman, T.S., 2011. Selecting optimal alternatives and risk reduction strategies in decision trees. Operations Res. 59 (3), 631–647.
Srivastava, S.K., 2006. Logistics and supply chain practices in India. Vision 10 (3), 69–79.
Tang, C.S., 2006. Robust strategies for mitigating supply chain disruptions. Int. J. Logis.: Res. Appl. 9 (1), 33–45.
Tokar, T., 2010. Behavioural research in logistics and supply chain management. Int. J. Logis. Manage. 21 (1), 89–103.
Üster, H., Kewcharoenwong, P., 2011. Strategic design and analysis of a relay network in truckload transportation. Transport. Sci. 45 (4), 505–523.
Vasco, R.A., Morabito, R., 2016. The dynamic vehicle allocation problem with application in trucking companies in Brazil. Comput. Operations Res. 76, 118–133.
Wyman, O., 2011. All rights reserved Factoring «Risk» into Transportation and Logistics Sourcing.
Young, P.C. and Tippins, S.C., 2000. Managing business risk: an organization-wide approach to risk management. Amacom.
Yazdi, M., Daneshvar, S., Setareh, H., 2017. An extension to Fuzzy Developed Failure Mode and Effects Analysis (FDFMEA) application for aircraft landing system.
Safety Sci. 98, 113–123.
Yoshimoto, R., Nemoto, T., 2005. The impact of information and communication technology on road freight transportation. IATSS Res. 29 (1), 16–21.
Zavadskas, E., Antucheviciene, J., Turskis, Z., Adeli, H., 2016. Hybrid multiple-criteria decision-making methods: a review of applications in engineering. Scientia
Iranica. Trans. A Civil Eng. 23 (1), 1.
Zeng, B., Yen, B.P.-C., 2017. Rethinking the role of partnerships in global supply chains: a risk-based perspective. Int. J. Product. Econ. 185, 52–62.
Zhang, Y., Zuo, F., 2016. Selection of risk response actions considering risk dependency. Kybernetes 45 (10), 1652–1667.
Zolfagharinia, H., Haughton, M., 2014. The benefit of advance load information for truckload carriers. Transport. Res. Part E: Logis. Transport. Rev. 70, 34–54.

55