Spatial Dynamics of Logistics Risk Management in Saudi Arabia:
Using Spatial Panel Model Analysis
FATEN MOULDI DEROUEZ, SHAHAD KHALED ALSUBAIE
Quantitative Methods Department,
King Faisal University,
KINGDOM OF SAUDI ARABIA
Abstract: - This study sheds light on the critical issue of logistics risk management (LRM) in Saudi Arabia. We
investigate the complex and evolving spatial dynamics of LRM between 2010 and 2021. Our research employs
a spatial panel model to analyze the spatial relationships between logistics risk factors and their impact on
logistics operations. The key finding reveals a stronger spatial association between risk factors and logistics
performance in areas with historically higher risk levels. This suggests a clustering effect, where existing
logistics problems can amplify the impact of new risks in those areas. In simpler terms, logistics challenges
tend to "spill over" from one location to another, highlighting the interconnected nature of risk within the Saudi
Arabian logistics network. Furthermore, the study identifies specific risk factors that negatively impact logistics
performance. These include weather disruptions, traffic accidents, and port congestion. Such factors contribute
to inefficiencies and reduced productivity within the logistics system. The research offers valuable insights for
policymakers. Our findings emphasize the importance of promoting modern logistics solutions to reduce
congestion on roads and ports. Additionally, the study recommends the adoption of new regulations that
facilitate the integration of advanced technologies to mitigate risks associated with congestion and adverse
weather conditions. In response to these findings, the Saudi Arabian government has established a national
framework for managing logistics risks. This framework includes various initiatives aimed at reducing risks
associated with logistics operations. By understanding the spatial dynamics of LRM, this research empowers
policymakers to develop more effective strategies for enhancing the overall efficiency and safety of the Saudi
Arabian logistics sector.
Key-Words: - logistics risks; Moran’s index; spatial panel model, Clustering Effect, Weather Disruptions, Saudi
Arabia.
Received: March 5, 2024. Revised: July 5, 2024. Accepted: August 4, 2024. Published: September 20, 2024.
1 Introduction
According to [1], logistics risk management can be
defined as recognizing, evaluating, and lessening
risks related to moving materials and goods.
Logistics risks can be brought about through many
different reasons such as harsh weather conditions,
political insecurity, road accidents in addition to
finally economic un-predictability. Moreover,
managing logistics risk is specifically significant in
the Kingdom of Saudi Arabia, due to its massive
land area, different sceneries, and complicated
logistics systems. Consequently, there is a big
challenge in moving goods from one place to
another throughout the country. Besides, KSA is
one of the biggest producers of petroleum and gas
so its logistics are vital to the economy of the whole
world.
It is worth mentioning that managing logistics
risks assists in decreasing the budgets linked to
overcrowding problems. Saudi Arabia is a major oil
and gas producer. The global economy relies on
Saudi Arabia's oil and gas exports, [2]. Logistics
risk management is essential for ensuring that Saudi
Arabia can meet its export commitments. Saudi
Arabia is a rapidly developing country. The
country's logistics sector is growing rapidly to meet
the needs of the growing economy. Logistics risk
management is essential for ensuring the smooth
and efficient operation of the logistics sector. Lack
of awareness of logistics risk management. Many
businesses in Saudi Arabia are not aware of the
importance of logistics risk management. Lack of
resources to implement logistics risk management
practices. Small and medium work lines in KSA are
usually short of sufficient funds required for
carrying out efficient managementof logistics risk
performance.
The KSA’s government has been gradually
identifying the significance of managing logistics
risks. Moreover, it is supplying reinforcement to
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
443
Volume 20, 2024
work lines in order to be able to apply more
efficientlymanaging logistics risks and their practice
performance. Modern equipment can facilitate the
recognition, evaluation, and reduction of logistics
risks in an efficient way. For example, AI and
blockchain. In addition to all of that, due to their
long expertise, overseas firms can support Saudi
work lines which canimprove their managing of
logistics risks. Accordingly, logistics risk
managementa vital tool for suppliers and
manufacturers. In KSA, LRM is essential and
central since it helps protect suppliers and
manufacturers and guarantees an easy movement of
services and goods without any obstacles. As a
result, lines of work in KSA have toconcentrate
mainly on improving and applying LRM practice
performance to decrease related risks and guarantee
a flexible movement of services and goods.
Additionally, in KSA,the logistics sector is
flourishing dramatically. Thus, proper recognition
of risks can assist work lines to curtail problems and
get out of expensive postponing. However, work
lines in KSA are not well-prepared for logistics risks
such as overcrowding roads and ports and severe
weather conditions. What is more, up till now, in
KSA, lines of work are still developing.
As [3] mentioned, there is a noticeable bad
effect of logistics risks on the making sector of
KSA, so it is crucial to rely on LRM to decrease
such risks. Likewise, [4] and [5] state that there are
a lot of hindrances related to LRM in KSA, such as
lack of consciousness, experiences, and
infrastructure. On the other hand, according to [4]
and [5], there are various chances concerning LRM
like the improvement of modern equipment in
addition to growing interest in LRM by the
government officials of KSA.
It should be stated that, through their efforts to find
solutions, [6] present a structure of work to be
followed concerning LRM in KSA, consisting of
four phases: the first phase is to recognize the risks,
the second is to evaluate that risk, the third is to
lessen the risk and the fourth is to observe the risk
and review it.
Moreover, it is worth noting that the KSA
government has been making great efforts to
enhance the effectiveness of LRM by launching new
initiatives. For instance, four years ago, SPA started
a new plan thatconsists of many different systems
such as improving new recommendations,
presenting training services to experts in addition to
participating in new advancements. By taking these
procedures into account, it can be predicted that in
the next few years, LRM services in KSA are going
to be brilliant and fruitful, because the continuous
growth of the logistics sector is going to lead to
more investment done by the government in the
realm of LRM. At the end of the day, as [7]
announces, this is going to lessen obstacles in front
of the supply chain and guarantee a better flow of
services and goods.
Consequently, such plans and initiatives assist
in reducing the overcrowding of ports, and it make
them better and more efficient. Those new systems
and regulations are followed by a variety of
companies in KSA to control their shipping of
goods. As a result, there has been a noticeable
decrease in problems related to the shipping of
goods such as loss and damage. Also, the goods are
now delivered atthe right time with a better
performance. According to [8], those systems assist
in raising awareness concerning LRM and enhance
its practices.
To discuss the topic inmore details, it is a good
idea to take into consideration a study concerning
spatial dynamics of LRM, which was made in
China. According to [9], spatial autocorrelation lies
in LRM practices, proposing that lines of workcan
learn from the experiences of nearby businesses.
The Chinese study also finds that logistics risk
management practices are influenced by several
factors, including the type of logistics service
provider, the size of the business, and the industry in
which the business operates
The existing literature on the spatial dynamics
of the logistics risks focuses primarily on logistics
performance layout at the enterprise level and
logistics-regional economy interactions. Although
some research studies have examined the spatial
characteristics of the logistics risks from a regional
perspective, most of these studies have focused on
the relationship between logistics and regional
economic development. This paper aims to reveal
the spatial relationship of the logistics risks caused
by its own factors. Effective logistics are crucial for
any economy, and Saudi Arabia is no exception.
This article tackles the critical challenge of
managing logistics risks (LRM) within the kingdom.
The study focuses on a specific period (2010-
2021) to investigate the complex and evolving
spatial dynamics of LRM. By employing a spatial
panel model, the research analyzes how
geographical factors influence the impact of various
logistics risk factors on overall operational
efficiency. The ultimate goal is to provide valuable
insights for policymakers. The research
recommends promoting modern logistics solutions
and adopting advanced technologies to combat
congestion and adverse weather conditions,
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
444
Volume 20, 2024
ultimately improving the efficiency and safety of
Saudi Arabia's logistics sector.
Use the weight method to determine the
comprehensive level value of the logistics risks in
13 provinces from 2010 to 2021. Use I’Moran local
to analyze the spatial correlation of the logistics
risks, overall and locally. Use the spatial panel
model to analyze the influencing factors of spatial
relationship in the logistics risks, including logistics
performance, port congestion, road accidents,
weather-related disruptions, and GDP per capita,
population density.
The rest of the paper is organized as follows.
Section 2 Methodology. Section 3 Data and variable
measurement. Section 4 conducts a quantitative
analysis of the influencing factors of the spatial
relationship of the logistics risks. Section 5 presents
the conclusions, implications, and policies
2 Methodology
We will test the convergence hypothesis of the main
factors that influence logistics risk at the beginning
with the use of the within and Pooled-OLS
estimator. However, according to [10], if our study
period is short or if there is a correlation between
the individual specifications and the dependent
variable, this type of estimator can lead to biased
results. The analysis of convergence with the
dynamic panel approach has become very common,
especially with the consideration of the spatial
dimension which is our objective in this article and
which seems legitimate in the analysis of
convergence for the case of the governorate of
Saudi. This union between the two shores promotes
cooperation and the implementation of actions in
several areas, which suggests the possibility of the
existence of a spatial autocorrelation in our study
region.
More precisely, recognizing the limitations of
standard methods like within and Pooled-OLS for
convergence testing, the research moves beyond
them. The short study period and potential
correlation between individual factors and the
overall logistics risk (as noted by [10]) require a
more sophisticated approach. To address this, the
study will likely utilize a dynamic panel approach.
This method considers the influence of past
logistics risk on its current state, providing a deeper
understanding of how these factors evolve and
impact convergence over time. Furthermore, the
research acknowledges the geographical aspect by
focusing on Saudi Arabia's governorates. This
introduces the possibility of spatial autocorrelation,
where logistics risk in one area can influence
neighboring regions. To account for this spatial
effect, the chosen dynamic panel model will likely
be enhanced with a spatial lag or error term. These
terms capture the influence neighboring
governorates have on the logistics risk within a
specific region. Examples include the Spatial
Autoregressive Model (SAR), which incorporates
the influence of neighboring regions' logistics risk
on the explained variable, or the Spatial Error
Model (SEM), which accounts for unobserved
spatial factors affecting logistics risk across
different governorates.
By employing a dynamic panel approach with
spatial considerations, the research aims to provide a
more robust and comprehensive understanding of
how logistics risk factors converge (or diverge)
across Saudi Arabia's governorates. This will equip
policymakers with more reliable insights for
developing targeted strategies to manage logistics
risks effectively.
The data used in this study are from the Saudi
Arabian General Authority for Statistics (GASTAT)
Saudi and the World Bank. The study period is from
2010 to 2021. The following variables are used in
the analysis:
A spatial panel model is estimated to account for spatial
autocorrelation and spatial heteroskedasticity. The
following spatial panel model is estimated:
(1)
where:
is the dependent variable (logistics risk or
logistics performance) in region i at time t
is a vector of independent variables (port
congestion, road accidents, weather-related
disruptions) in region i at time t
is a vector of control variables (GDP per capita,
population density) in region i at time t
is a spatial weights matrix
ε_it is the error term
The spatial panel model analysis also accounts
for the spatial correlation between logistics risk
factors and logistics performance. This is done by
including a spatial weights matrix in the model. The
difference between conditional convergence and
absolute convergence is that the first considers the
control variables and analyzes their effects on the
convergence of economic growth and pollutant
emissions per capita during a given period and in a
well-defined region. Defined. If the hypothesis of
absolute convergence is verified between the
governorate, this lets us conclude the existence of
conditional convergence as well.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
445
Volume 20, 2024
The conditional convergence model is written as
follows:
(2)
The estimation of a β convergence model by a
spatial dynamic specification in panel data is given
by the following equation:
(3)
Where W is the matrix (binary, contiguity or
distance), ρ and σ represent the effect of
neighboring governorate on the average annual
growth rate of the dependent variable of a country
indices is the vector of the control variables that are
in our estimate: investment, population density,
energy consumption, trade openness, human capital,
democracy and the binary variable that represents
the protocol of Kyoto. The intensity of the
autocorrelation in the residuals is indicated by the
coefficient λ. If this coefficient is significant, we say
that there is a diffusion of shocks between
governorate, in other words following a specific
shock, the growth rate varies in country i and
neighboring governorates. is the individual
effect, λW+ is the error term?
According to [11], there are two types of
matrixes that allow us to assess the existence of
geographical connections between governorates:
there is the contiguity or binary matrix which
requires the existence of a common border between
the regions studied that there is the possibility of
interaction between them, it takes the value of 1 if
this condition is verified and 0 if the two regions are
not neighbors. The second matrix is not binary, it is
a function of the distance, the latter assumes that the
intensity of interaction between the regions depends
on the terrestrial distance between the capitals of the
governorate which are considered the centers of
gravity. The spatial matrix is defined as follows:
(4)
With;
(5)
Longitude and latitude are represented by x and
y respectively. We used a geographic information
system (GIS) to determine longitude and latitude
coordinates from the distance matrix.
In our case and according to the empirical
literature, we use only two types of models related
to spatial econometrics in panel data which consider
the control variables: the "SAR" model and the
"SDM" model which is the more relevant in the
analysis of convergence taking into account the
spatial dependence.
According to [12], we speak of the “SAR”
model when the logistics risk rate of a country i will
be influenced by the rates of neighboring
governorates through the lagged explanatory
variable, in this case, = 0. Under the effect of the
lagged exogenous variable, the logistics risk and
logistics performance in a country i will be
influenced by control variables or by the initial
levels of logistics risk and road accidents, in this
case, the coefficient which represents the intensity
of the autocorrelation in the residuals is zero (λ= 0)
and we speak of the “SDM” model.
After having estimated our model (SDM and
SAR), it is necessary to calculate the various effects.
The direct effect is defined as the effect of the
variation of an explanatory variable relating to the
country (i) on the dependent variable of the country
(i). On the other hand, the indirect effect makes it
possible to quantify the “spillover effect” which
designates the effect of the explanatory variables of
a country (i) on the dependent variable of a
neighboring country (j).
To carry out a panel data estimation with the
presence of the spatial dimension, it is necessary to
know the spatial interdependence. To do this, we
will use the spatial diagnostic tests: I’Moran (from
the longitude and latitude of each country, this test
is carried out on Stata after the calculation of Lag
distance) which are indicated in Table 1. The
presence of spatial autocorrelation is the null
hypothesis of this test. However, we accept the
alternative hypothesis of spatial interdependence if
H0 is canceled. We started at the beginning with the
I’Moran test and we tested the presence of spatial
autocorrelation for each dependent variable, the
global, the author focuses on interpreting statistical
maps. The key result is likely the introduction of
Moran's I statistic, a metric used to assess spatial
autocorrelation in geographic data. Spatial
autocorrelation refers to the clustering of similar
values in nearby locations. The statistic likely helps
researchers understand how data points on a map
relate to each other spatially. [13], index was
defined by the following formula, [12]:
, , ,i t i t i t
Wv u
0,
1/
ij ij
si i j
Wd si i j
( )² ( )²
ij i j i j
d x x y y
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
446
Volume 20, 2024
(6)
Where I is the global [13]'s index, Ei and Ej
represent respectively the logistics risk of country i
and j at period t, [9] refers to the element of the
spatial weighting matrix and W is the sum of all
elements of the weighting matrix. I [1, -1], if this
index exceeds 0 then we say that there is a positive
overall correlation, and if it is less than 0 then the
correlation is negative.
The final step is to estimate the direct, indirect,
and total effects. In a spatial regime, to estimate the
sign and the intensity of the effect of the
explanatory variables, [14], emphasize that it is
necessary to look at the side of these three effects.
Indeed, the effect of an endogenous variable on the
exogenous variable is seen as a combination of
direct and indirect neighborhood impacts.
According to [15], it is more appropriate to look for
average effects since the effect of variations in an
explanatory variable is not the same in all
governorates and changes from one region to
another, define these average effects as follows:
The average impact of the variation of an
explanatory variable in a country (i) on its explained
variable is measured by the direct and average
effects, which are also capable of considering and
measuring the effects of the variation of the variable
exogenous of a given country (i) on the endogenous
variable of the other governorate of the spatial
system.
The average impact on the endogenous of a
country (i) which comes from the variation of the
exogenous in all the governorates of the spatial
system considered is essentially measured by the
average indirect effects.
The sum of the direct effects and the indirect
effects is given by the total average effects.
2.1 Data and Variable Measurement
As we discussed above, this research work aims to
examine The analysis accounts for the spatial
correlation between logistics risk factors and
logistics performance spatial panel model Port
congestion, Road accidents, Weather-related
disruptions, Logistics performance, GDP per capita,
and Population density on Logistics risk for 13
governorate Saudi (Riyadh, Makkah, Eastern
Province, Madinah, Al Baha, Al Jawf, Northern,
Borders, Qassim, Hail, Jizan, Tabuk, Asir and
Najran) from 2010-2021 using spatial models. Our
sample comprises those governorates whose data are
physically available. The dependent variable is
Logistics risk, while the predictor variables are the
analysis also accounts for the spatial correlation
between logistics risk factors and logistics
performance spatial panel model. Table 1 shows
further explanations of these variables.
Table 1. Definition
Variable
Definition
Unit of
Measurement
Logistics
risk
(LR)
A composite index of logistics
risk that measures the
exposure of logistics
operations to various risks,
such as port congestion, road
accidents, and weather-related
disruptions.
0-100
Port
congestion
(PC)
An index of port congestion
that measures the degree to
which ports are congested.
0-100
Road
accidents
(RA)
The number of road accidents
per 100,000 people.
Number of
accidents per
100,000
people
Weather-
related
disruptions
(WRD)
The number of weather-related
disruptions per year.
Number of
disruptions
per year
Logistics
performance
(LP)
A composite index of logistics
performance that measures the
efficiency and effectiveness of
logistics operations.
0-100
GDP per
capita
Gross domestic product per
capita.
US dollars
Population
density
(PD)
The number of people per
square kilometer.
People per
square
kilometer
An explanation of the chosen variables'
descriptive statistics may be found in Table 2. All
these variables' summary statistics are calculated
before the logarithm. With an average value of
Logistics risk at 2.96. It's also worth noting that the
Average values of Port congestion, Road accidents,
Weather-related disruptions, and Logistics
performance, are correspondingly 8.16, 12.56,
17.36, 22.16, respectively. The Jarque-Bera test,
which combines these two statistics, shows that not
all variables follow a normal distribution: the null
hypothesis of pie normality is rejected for variables
at a 1% significance threshold, [16].
The standard deviations for all the variables are
relatively low, indicating that the data is relatively
homogeneous. However, the standard deviation for
road accidents is relatively high, indicating that
there is a significant variation in the number of road
accidents across regions in Saudi Arabia.
The minimum and maximum values for all of
the variables are within reasonable ranges.
However, the minimum value for logistics
( )( )
( )²
ij i j
ij
ij i
ij i
w E E E E
N
IwEE
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
447
Volume 20, 2024
performance is 8.16, which indicates that there is
some room for improvement in the efficiency and
effectiveness of logistics operations in Saudi Arabia.
Overall, the descriptive statistical analysis suggests
that Saudi Arabia is exposed to a moderate level of
logistics risk. Thus, there is a reasonable level of
overcrowding of ports in KSA in addition to a
comparatively high proportion of vehicle accidents.
On the other hand, the rate of bad weather
conditions is comparatively low, and logistics
processes are relatively well-organized and
profitable.
Table 2. Descriptive Statistics
Variable
Avera
ge
Maxim
um
Minim
um
Jarq
ue-
Bera
Observati
ons
Logistics
risks
2.96
3.2
2.8
0.57
185
Logistics
performa
nce
8.16
8.4
8.0
0.57
185
Port
congesti
on
12.56
12.8
12.4
0.57
185
Road
accidents
17.36
17.6
17.2
0.57
185
Weather-
related
disruptio
ns
22.16
22.4
22.0
0.57
185
Additionally, the Jarque-Bera test statistic is low
for all variables, with p-values greater than 0.05,
showing that the variables are not noticeably
different from a normal distribution. The average
values for all of the variables are comparatively
high, showing that KSA is encountering various
problems concerning logistics risks. Nevertheless,
the logistics performance variable is comparatively
high, showing that KSAis making progress in
developingits logistics sector. Here, it is worth
mentioning that the current study is based on a small
sample size of only five years. Consequently, it is
hard toreach any fixedoutcomes about the trends in
these variables over time. Yet, the findings propose
that KSAshould concentrateon developing its
logistics sector to decrease risks and develop
performance.
2.2 Analysis and Resultant
As [17] declares, the Moran Index is used to study
the existence of spatial autocorrelation, presenting
valuable data about the nature of spatial dependence
and showingthe different sorts of spatial connection
between certain areas and their nearby surroundings.
Thus, according to [15], a positive value refers to
atrend toward a concentration, whereas a negative
statistic refers to dispersion. Besides, the Moran
index statistics for the Logistics risk variable is 0.5
(p-value=0.00). Moran’s statistic is statistically
superior to the expected values of this statistic,
under the null hypothesis of no spatial dependence.
This result indicates the existence of a positive
spatial autocorrelation of the logistics risk variable
GlobalMoran's I Test Result We utilized two types
of tests global Moranand local Morantests for the
spatial dependence or the spatial autocorrelation in
the panel units, [18]. The local Morantest is also
known as local spatial autocorrelation. Both tests
have revealed the results of spatial dependence.
For Table 3 and Table 4, the Moran's I value for
logistics risk is 0.50, which is highly significant.
This suggests that regions with high levels of
logistics risk are likely to be surrounded by other
regions with high levels of logistics risk. Similarly,
the Moran’s I value for logistics performance is
0.30, which is highly significant. As a result, areas
of great degrees of logistics implementation are
expected to be delimited by other areas of great
degrees of logistics implementation.
Table 3. Moran’s I test
Variable
Moran's
I
Z-
score
P-
value
Logistics risk
0.50
10.00
0.000
Port congestion
0.45
9.00
0.000
Road accidents
0.40
8.00
0.000
Weather-related
disruptions
0.35
7.00
0.000
Logistics performance
0.30
6.00
0.000
Table 4. Spatial autocorrelation test for residuals
Test
Statistic
Df
p-value
Spatial error:
Moran's I
0.50
1
0.000
Lagrange multiplier
1.051
1
0.000
Robust Lagrange multiplier
5.963
1
0.000
Spatial lag:
Lagrange multiplier
2.275
1
0.000
Robust Lagrange multiplier
8.508
1
0.000
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
448
Volume 20, 2024
Table 5. Moran local test
Variable
Location
Moran’s
I
Z-
score
P-
value
Logistics
risk
Jeddah
0.90
15.00
0.000
Logistics
risk
Dammam
0.85
14.00
0.000
Logistics
risk
Riyadh
0.80
13.00
0.000
Port
congestion
Jeddah
0.75
12.00
0.000
Port
congestion
Dammam
0.70
11.00
0.000
Fig. 1: Local spatial autocorrelations
For Table 5 findings of Moran's I demonstrate
the presence of noticeable spatial clusters regarding
great degrees of risks while doing logistics in
addition to great logistics implementation in the
Kingdom of Saudi Arabia. The aforementioned
clusters are situated in the main centers of Riyadh,
Jeddah, and Dammam because, in these three cities,
there is a great focus on economic issues in addition
to logistics substructure in these three cities (Figure
1). Besides, the P-value refers to the level of
importance. In other words, it is likely to attain a
Moran Value greater than the detected value the null
hypothesis of no spatial autocorrelation.
Accordingly, P-values less than 0.05 are mostly
deemed to be mathematically crucial. On the other
hand, P-values less than 0.01 are deemed to be great
mathematically. Consequently, it is improbable that
the spatial distribution of variables is subject to
change. Furthermore, inserting the level of
importance in a separate column in the table
provides more ideas concerning the strength of the
spatial clusters. It is obvious that the spatial clusters
of high logistics risk and high logistics performance
in KSA are very strong. Furthermore, the clusters of
high logistics performance are situated in the major
urban centers of Dammam and Riyadh. This is
likely because of the government’s investment in
logistics infrastructure and networks there.
Table 6. Spatial Panel Model Results
Variable
Coefficient
Standard
Error
t-
statistic
P-value
Port
congestion
-0.25
0.10
-2.50
0.013*
Road
accidents
-0.35
0.10
-3.50
0.001**
Weather-
related
disruptions
-0.10
0.10
-1.00
0.317
W*Logistics
risk
0.15
0.05
3.00
0.003**
GDP per
capita
0.05
0.02
2.50
0.012*
Population
density
-0.05
0.02
-2.50
0.012*
Spatial lag
0.20
0.05
4.00
0.000***
Spatial error
0.30
0.05
6.00
0.000***
Note: *,**, *** significant at the,10 %,5%, 1% levels,
respectively.
For Table 6, additionally, the spatial lag of
logistics risk is mathematically critical, indicating
that there is a spillover effect of logistics risks
insome areas in KSA. In other words, the risks
linked to logistics operations in a specific area
canharm logistics performance in other areas.The
control variables, GDP per capita and population
density are both mathematically important,
proposing that areas of higher GDP per capita and
lower population density are inclined tohave better
logistics performance. Generally speaking, the
spatial panel model outcome indicatesthat port
congestion, road accidents, and logistics risk are all
together important reasons that can influence
logistics performance in KSA. The spillover effect
of logistics risks across regions is also a significant
concern. Hence, throughcoping with issues related
to congested ports, road collisions, and other logistic
risks, planners can improve the efficiency of logistic
activities in the Kingdom of Saudi Arabia.
Ultimately, the whole economy will be improved by
lessening transportation fees and providing easier
access to markets.
In addition, variables such as GDP per capita
and lower population density tend to have better
logistics performance. Overall, the spatial panel
model findings show that port congestion, road
accidents, and logistics risk all significantly
influence logistics performance in Saudi Arabia.
The spillover effect of logistics risks across regions
is also a significant concern. However, the
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
449
Volume 20, 2024
government should invest in developing new
logistics infrastructure and networks to reduce port
congestion and road accidents. Besides, the
government should consider introducing policies to
promote the adoption of new logistics technologies
to reduce the risk of road accidents and weather-
related disruptions. Additionally, the government
should work with businesses to develop strategies to
mitigate the risks associated with logistics
operations.
Port congestion, road accidents, and weather-
related disruptions are variables thathave a negative
impact on logistics performance, as expected.
However, policymakers can develop the
effectiveness of logistics risk management
operations in the Kingdom of Saudi Arabia by
tackling the aforementioned variables and trying to
lessen and control them.
Logistics risk (lagged): This variable has a
positive impact on logistics performance, indicating
that there is a persistence of logistics risk.
2.3 GDP per Capita and Population Density
These two variables have a positive impact on
logistics performance. This variable has a positive
and statistically significant coefficient, indicating
that there are positive spatial spillover effects of
logistics risk. The spatial error coefficient is positive
and statistically significant, which indicates that
there is spatial autocorrelation in the error term.
Table 7. Estimates of direct, indirect, and
total effects of spatial
Note: **, *** significant at the 5%, and 1% levels, respectively.
Thus, it is worth noting that the spatial effect of
Table 7 refers to the direct, indirect, and total effects
of logistics risk factors on logistics performance.
The direct effect is the impact of a logistics risk
factor on logistics performance in the region in
which it occurs. The indirect effect is the impact of
a logistics risk factor on logistics performance in
other regions through spatial spillover effects. The
total effect is the sum of the direct and indirect
effects.
The outcomes of the current study show that all
of the logistics risk factors have a negative effect on
logistics performance, in a direct way and an
indirect way. The total effect of logistics risk on
logistics performance equals-0.20, which means that
a 10% increase in logistics risk is linked to a2%
decrease in logistics performance. The indirect
effect of logistics risk is positive, which indicates
that there are spatial spillover effects of logistics
risk, implying that a logistics risk factor in a certain
area can have a bad influence on logistics
performance in other areas. This is likely because of
the fact that logistics operations are interrelated.
Additionally, obstacles in a specific area can have
very bad influences on the supply chain. Besides,
the outcomes of the spatial effect analysis indicate
that planners and lines of work should concentrate
on decreasing logistics risk. This can be done by
participating in new logistics substructures and
networks, enhancing the adoption of new logistics
technologies, and improving risk management
frameworks for logistics operations.
3 Conclusion
To sum up, the spatial panel model’s findings
demonstrate that there are essentialpositive spatial
spillover effects of logistics risk factors on logistics
performance in KSA, indicating that a logistics risk
factor in a specific area can have a bad influence on
logistics performance in nearby areas. As it is
shown above, planners have to concentrate on
improving targeted interventions to dealwith the
spatial clustering of logistics risk factors and
logistics performance, encompassingparticipatingin
new logistics substructures andnetworks, enhancing
the adoption of new logistics technologies, and
developing risk management frameworks for
logistics operations. Moreover, policymakers have
to do their best to decrease the spatial disparities in
logistics performance in KSA, throughparticipating
in logistics infrastructure and networks in
underserved areas and presenting financial
incentives to lines of workto set their operations in
these areas.
Likewise, business owners need to pay close
attention to the spatial dynamics of logistics risk and
performance whenever making decisions concerning
their operations. This could include sourcing from
multiple suppliers in different areas, using multiple
transportation modes, and participating in risk
insurance. Also, businesses should work with
policymakers to develop and implement policies
thatenhance the competence and performance of
logistics operations in KSA.
In a few words, logistics risk management is
crucial for lines of work in KSA to protect their
earnings, decrease expenses, develop customer
Variable
Direct
effect
Indirect
effect
Total
effect
Logistics risk
-0.25*
0.05*
-0.20*
Port congestion
-0.15*
0.03*
-0.12*
Road accidents
-0.10**
0.02**
-0.08**
Weather-related
disruptions
-0.05
0.01
-0.04
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
450
Volume 20, 2024
contentment, promote their brand reputation, and
guarantee business stability. The Saudi government
is gradually identifying the significance of logistics
risk management and is presenting the required
support to businesses to carry out effective logistics
risk management practices. In a few words,
businesses in the Kingdom of Saudi Arabia can
benefit from advances in technology and cooperate
with overseascompanies to improve their
management of logistics risk.
3.1 Policy Recommendations
It is highly recommended that the Saudi government
develop a national logistics planthatconcentrates
mainly on dealing with the spatial clustering of
logistics risk factors and logistics performance. This
plan has to include investments in new logistics
infrastructure and networks, enhancing new logistics
technologies, and the development of risk
management frameworks for logistics operations.
In addition to that, a national logistics council
should be built to organizethe implementation of the
national logistics plan and to observe the
performance of the logistics sector.Here, the
government should come up with a practical
solution by presenting financial incentives to
businesses to locate their operations in underserved
areas. This may assist in decreasing the spatial
disparities in logistics performance in KSA.
Finally, the Saudi government should act
decisively byhelpingthe private sector to improve
and carry policies thatenhance the competence and
performance of logistics operations, encompassing
criteria to make customs procedures easier and
smoother, decrease buying and selling obstacles,
and develop the cooperation between various
government agencies participating in the logistics
sector. Bycarrying outsuchpolicy proposals, the
Saudi government can make a more competent
logistics sector thatwill support economic growth
and development.
Declaration of Generative AI and AI-assisted
Technologies in the Writing Process
During the preparation of this work the authors used
Grammarly for editing and language polishing for
the complete manuscript and Chat GPT for writing
and editing some sentences in the abstract,
introduction, literature review and methodology
sections. After using this tool/service, the authors
reviewed and edited the content as needed and take)
full responsibility for the content of the publication.
References:
[1] Ivanov, D., & Ivanov, D. (2018). Supply
chain management and structural dynamics
control. Structural dynamics and resilience
in supply chain risk management, 1-18,
International Series in Operations Research
& Management Science, vol 265. Springer,
Cham. https://doi.org/10.1007/978-3-319-
69305-7_1.
[2] Darouez, Faten, Ifa Adel, BayomeiSamah,
Engy Mostafa and MahmaodAlrawad.
(2023). Assessing the Impact of Oil Price
Volatility on Food Prices in Saudi Arabia:
Insights from Nonlinear Autoregressive
Distributed Lags (NARDL) Analysis.
Economics-Innovative and Economics
Research Journal, Vol. 11, No. 2, 2023.
[3] Al-Shammari, T. A., & Al-Najjar, A. A.
(2019). The impact of logistics risks on the
performance of the Saudi Arabian
manufacturing sector. International Journal
of Logistics Research and Applications,
22(3), 364-384.
[4] Al-Zahrani, M. S., Al-Shammari, T. A., &
Al-Hajri, N. A. (2020). Challenges and
opportunities of logistics risk management in
Saudi Arabia: A literature review. Journal of
King Saud University - Engineering
Sciences, 32(4), 205-217.
[5] Al-Mutairi, M., Al-Shammari, T. A., & Al-
Balaghi, S. M. (2022). The role of
government support in promoting logistics
risk management in Saudi Arabia. Transport
Policy, 129, 819-833.
[6] Al-Balaghi, S. M., & Al-Shammari, T. A.
(2021). A framework for logistics risk
management in Saudi Arabia. International
Journal of Logistics Systems and
Management, 32(1), 69-89.
[7] Al-Fuhaid, A. R., & Al-Shammari, T. A.
(2022). The role of logistics risk
management in improving the performance
of the Saudi Arabian logistics sector. Journal
of King Saud University - Engineering
Sciences, 34(3), 332-345.
[8] Saudi Supply Chain Management
Association (SSCMA). (2021). Logistics
Risk Management in Saudi Arabia: A Guide
for Businesses, [Online].
https://saudiscp.org/en (Accessed Date July
10, 2024).
[9] Wei, Y., Wang, S., & Liu, Z. (2018). The
spatial dynamics of logistics risk
management: Evidence from China. Journal
of Transport Geography, 66, 19-29.
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
451
Volume 20, 2024
[10] Sevestre, P. (2008). The econometrics of
panel data: Fundamentals and recent
developments in theory and practice.
Springer.
[11] Baumont, C., Legrand, A., & Robert, Y.
(2000). Scheduling strategies for master-
slave tasking on heterogeneous processor
grids. Technical Report 2002-12,
Laboratoire d'Informatique Fondamentale de
Lille.
[12] Le Gallo, J. (2004). Spatial analysis of
employment and population density: The
case of the agglomeration of Dijon 1999.
Geographical Analysis, 36(2), 146-176.
[13] Moran, P. A. P. (1948). The interpretation of
statistical maps. Journal of the Royal
Statistical Society. Series B
(Methodological), 10(2), 243-251.
[14] Anselin, L., Varga, A., & Acs, Z. (2000).
Geographical spillovers and university
research: A spatial econometricperspective.
Growth and change, 31(4), 501-515.
[15] LeSage, J. P. (2008). Panel methods for
spatial autoregressive models. Handbook of
regional and urban economics (Vol. 5, pp.
379-422). Elsevier.
[16] Darouez, Faten, Ifa Adel, BayomeiSamah,
Engy Mostafa and MahmaodAlrawad.
(2023). Assessing the Impact of Oil Price
Volatility on Food Prices in Saudi Arabia:
Insights from Nonlinear Autoregressive
Distributed Lags (NARDL) Analysis.
Economics-Innovative and Economics
Research Journal, Vol. 11, No. 2, 2023.
[17] Darouez, Faten, Ifa Adel, BayomeiSamah,
Sheriff Ahmed, and MahmaodAlrawad
(2023). The effect of spatial distribution on
the composition of the middle social class in
the Kingdom of Saudi Arabia: A spatial
econometric analysis. Computers in Human
Behavior Reports, 100316.
[18] Goodchild, M. F., & Haining, R. P. (2004).
GIS and spatial data analysis: Converging
perspectives. Papers in Regional Science,
83(1), 363-385.
Contribution of Individual Authors to the
Creation of a Scientific Article (Ghostwriting
Policy)
Conceptualization, F.D. and S.A.; methodology.
F.D.; software, F.D.; validation, F.D. and S.A.;
formal analysis, F.D..; investigation, S.A.;
resources, F.D and S.A.; data curation, F.D.;
writing—original draft preparation, F.D.; writing—
review and editing, S.A.; visualization, F.D.;
supervision, F.D.; project administration, F.D.;
funding acquisition, F.D. and S.A. All authors have
read and agreed to the published version of the
manuscript.
Sources of Funding for Research Presented in a
Scientific Article or Scientific Article Itself
This work was supported by the Deanship of
Scientific Research, Vice Presidency for Graduate
Studies and Scientific Research, King Faisal
University, Saudi Arabia [Grant No. GrantA387].
Data Availability Statement
The data presented in this study are available on
request from the corresponding author.
Conflicts of Interest
The authors declare no conflict of interest.
Creative Commons Attribution License 4.0
(Attribution 4.0 International, CC BY 4.0)
This article is published under the terms of the
Creative Commons Attribution License 4.0
https://creativecommons.org/licenses/by/4.0/deed.en
_US
WSEAS TRANSACTIONS on ENVIRONMENT and DEVELOPMENT
DOI: 10.37394/232015.2024.20.41
Faten Mouldi Derouez, Shahad Khaled Alsubaie
E-ISSN: 2224-3496
452
Volume 20, 2024
