Page 1 of 15

Archives of Business Research – Vol. 12, No. 12

Publication Date: December 25, 2024

DOI:10.14738/abr.1212.18101.

Deore, A. R., Shafighi, N., & Hajibashi, A. A. (2024). Impact of Machine Learning on Supply Chain Optimization. Archives of Business

Research, 12(12). 112-126.

Services for Science and Education – United Kingdom

Impact of Machine Learning on Supply Chain Optimization

Aakash Ramesh Deore

Najla Shafighi

Anahita Amini Hajibashi

ABSTRACT

Supply chain optimization or SCO is the progression that manages the

manufacturing industry of Germany and the rate of construction. The study aims to

and the consequence of machine learning or ML on SCO of the manufacturing

companies in Germany focusing on automobile and Internet of Things (IoT)

equipment manufacturing companies. From the study, it will be concluded that the

SCO system desires more precision in the supervision system from the employees of

the business industry.

Keywords: Artificial intelligence, Machine learning, Costs, Green supply chain

management, Command forecasting, Advancement in technology, Sustainable

development, and Manufacturing Industry

INTRODUCTION

Supply chain optimization in the manufacturing industry of Germany involves systematically

improving processes, resources, and logistics to enhance the supply chain's efficiency, cost- effectiveness, and overall performance. As per the views of Anjomshoae et al.(2022),this supply

chain optimization is a process or adjustment of supply chain management operations, and it

always makes sure that SCM can work with full efficiency. In Germany, supply chain

optimization plays a crucial role as there are advanced technologies and advanced

manufacturing sectors available. The main focus of this process is on the flow of materials,

information, and products from suppliers to manufacturers, and eventually to customers

(Aranda et al. 2019; Hemant, and Shafighi, 2023).

SCM is the optimization system in which the process of optimization of a business industry

becomes improved and the rate of the production of the business company becomes enhanced.

As per the recommendation of Abbas et al. (2020), with the help of the SCM system, the data,

information, and products related to finance became more protected. In order to track the

performance of the supplier and negotiate contracts as well as the proactive mitigation of

potential risk in SCM of manufacturing companies in Germany, the use of ML is highly

applicable. There are five kinds ofSCM that impact a business and marketindustry and these are

resources, planning, delivery of the products, making, and the return. As per the viewofAbdella

et al. (2020),SCM makes the optimization system of a business industry stronger and increases

the profit of the business organization. Therefore, the SCM is an important factor in the growth

and development of an industry. The use of the SCM in a business industry increases the

creativity in the product's optimization system, which makes the products more acceptable to

customers worldwide (Hemant, and Shafighi, 2023).

Page 2 of 15

113

Deore, A. R., Shafighi, N., & Hajibashi, A. A. (2024). Impact of Machine Learning on Supply Chain Optimization. Archives of Business Research, 12(12).

112-126.

URL: http://doi.org/10.14738/abr.1212.18101

AI and machine learning are set to disrupt the status quo across the automobile industry in

Germany. However, machine learning can improve the reduction ofthe emissions of vehicles. In

addition, algorithms of ML can determine various ways and reduce fuel consumption by

analyzing different datasets on driving patterns. The automobile sector has a huge amount of

data that can be utilized for proving training models of ML. Apart from this, there are several

tasks within the automotive industry that can benefit from the implementation of ML like

autonomous driving and predictive maintenance. One of the main reasons that the German

automobile industry has been rising for the past few years is the nature of home demand (Aich

et al. 2019). The supply chain changes frequently with the evolving requirements of meeting

the new demands within the maintenance of a smooth sailing flow. Lately, supply chain

optimization or SCM has been challenged by a number of factors such as inadequate inventory

planning, demand fluctuation, uncertainties in logistics, and above all the backlogs of orders.

Moreover, communication gaps also exist where there is a shortage in the supply at different

times (Pallathadka et al. 2023). In international business, sustainability implementation has

become a cause of concern along with the staff shortage in the post-Brexit era

(Commonslibrary.parliament, 2023). There has been an increase in the input prices which is

hard to deal with.

1. The main focus of the research is to determine the impact of Machine Learning in

Supply Chain Optimization in order to maintain the business in a German

manufacturing industry. Is it possible to rectify pivotal areas by applying machine

learning in SCO of German Manufacturing companies?

2. What is the existing state of adoption of machine learning in the German

manufacturing industry?

3. What are the challenges and barriers in adopting ML in SCO of the manufacturing

industries of Germany?

4. How would Sustainable development generate a substantial effect on Supply chain

Optimization?

5. Will advancement in Technologies will bring modernization to Supply chain

optimization?

6. What effect does supply chain optimization in the Manufacturing industry have on

attaining sustainable growth when machine learning and technological advancements

are combined?

LITERATURE REVIEW

Evaluating the Key Areas for Implementing ML in the SCM of Manufacturing

Companies in Germany

The role of ML in improving the efficiency of the supply chain is significant and requires the

support of data science and data computing for the growth and development of the

manufacturing industry of Germany. As per the opinion of Ivanov et al. (2019), data science

plays the vital role in the growing performance of manufacturing companies as such companies

require huge data for training of machines. According to Kamble et al. (2021), using ML can help

manufacturing companies to optimise production through getting insight of future trends and

possible changes in customer demand. Machine learning is the key component of technology

that can change the future of the economic development of Germany. The potentiality of ML to

improvethebusinessprospectofGermancompanies canbe seenthrough the automationoftasks

and the making of better decisions. Customer is the most important asset of an organisation and

Page 3 of 15

114

Archives of Business Research (ABR) Vol. 12, Issue 12, December-2024

Services for Science and Education – United Kingdom

the use of ML to satisfy the demand of them can increase the value of a brand (Frazzon et al.

2019). Manufacturing companies in Germany use ML for resolving the SCM related complaints

of customers and answering FAQs through automated operation. Effective supply chain

optimization ensures the practice of raw materials to perform reliably and assemble lines of

contract for effective functioning (Cole et al. 2019).

Disadvantages of Using Machine Learning in the Supply Chain Optimization in German

Manufacturing Companies

Data is a prime component of machine learning, and the wrong input can change the course of

production negatively. Khan et al. (2020) mentioned a lack of skills in employees can hinder the

successful use of machine learning in companies which leads to loss instead of betterment. With

the use of ML in the supply chain optimization of Germany, companies need to have an idea of

the German economy and the way it can be better. However, accessing the economic

information in an unsystematic manner can bring treats unfavorable situations. The regulation

of using machine learning is strict and unable to follow it can also bring unfavourable situations

for the company leaders. Kilimci et al. (2019) mentioned machines are highly unpredictable,

and failure of machinery can happen at any time during the operation. Thus, the dependency of

manufacturing industries on the machine learning system can be highly disadvantageous.It has

been suggested that relying heavily on technology and machines can increase challenges for

logistics and SCM for manufacturing companies during cyber-attacks, system failures, or the

occurrence of glitches.

Analysis of the ML Adoption Process in SCM of Manufacturing Companies in Germany

The gradual yet effective initiation can be seen in the manufacturing sector in Germany for

managing the supply chain effectively. In this context, Ni et al. (2020) explained, ML can be

adopted in manufacturing companies, primarily to analyse the lifecycle of machines. In order to

predict the failure of machines, German companies are adopting ML. apart from this and logistic

optimization is getting help due to ML adoption. Thus, manufacturing companies in, Abbas et al.

(2020) mentioned, in-store house-related information, and the use of ML has helped many

business professionals to save money and time. The automated alarming system in each step of

anomaly can help manufacturing companies in Germany maintain safety in storehouses. The

adoption of machine has opened various paths for the German companies towards regulating

the skills and capabilities in terms of globalizing the supply chain services. As opined by Aslam

et al. (2021), off shoring the production credibility is assured by the industrialized economy

which is readily available to build the capabilities for supply of energy. Germany is known to be

a powerhouse of technology due to the devotion of the research and design project that lends

support towards sustainability. As per the critical analysis by McMaster et al. (2021), the

startup booms in Germany has attracted the technical foundations with a background of AI and

machine learning. In such a context, nearly 7% of the German economic output has centred the

technical convulsions with respect to hardware segments.

Saving costs, reducing waste production and improving quality production can be possible

through the proper visibility of market conditions. ML is able to help in getting an insight into

all the essential requirement, thus, companies in Germany is adapting ML for improving lean

production. In order to make effective decisions for the business, the leaders of German

manufacturing firms are adopting ML (Xuet al. 2022).

Page 4 of 15

115

Deore, A. R., Shafighi, N., & Hajibashi, A. A. (2024). Impact of Machine Learning on Supply Chain Optimization. Archives of Business Research, 12(12).

112-126.

URL: http://doi.org/10.14738/abr.1212.18101

Challenges in the Adoption of Machine Learning in Supply Chain Optimization in

Manufacturing Companies of Germany

In order to access the benefit of machine learning, the availability of sufficient data is important.

In this context, Stockheim et al. (2023) mentioned, data required to train machines and the

absence of data can prevent the process of adoption of ML for professional purposes. On the

other hand, the use of machine learning requires specialized skills, as it is a highly technical and

complex process. Gupta and Gupta (2019) mentioned, having a workforce with no necessary

expertise can bring difficulties in implementing it. Managing the supply chain with the use of

ML requires skills of handling machines and data and the lack of understanding of that skill can

hinder the adoption of ML in manufacturing companies in Germany.

The adoption of ML is undoubtedly effective, although the process is costly in terms of the

requirements of both the hardware and software. Ivanov (2021) mentioned, ML need time to

be implemented in the manufacturing companies to maintain the proper alignment for better

use. Along with time consumption, the cost of labour training and resource collection is also

high for the adoption of ML in German manufacturing companies. In contrast to these Thiems

2022) mentioned, among all external challenges, the most prominent one is the difficulty of the

model of ML in interpreting while working. The German manufacturing industry is renowned

for its focus on quality testing, improvement planning and innovation strategy, and the use of

machine learning this industry to offer help in SCM needs to have the efficacy to deal with every

aspect. However, the models of machine learning often make decisions on their own to

automate service by hindering any three of the focused areas of German companies

(Steinberget al. 2023). Thus, the challenges are highly prominent and require the support of

some effective strategies to mitigate the threats and barriers for the manufacturing companies

in Germany.

METHODOLOGY

Primary Quantitative Method has been used in the research conduction that is based on

concrete ideas assigned to the industrial domain of Germany. Further, primary quantitative

data focuses on objectiveprotocols while conducting the study that mainly includes the resilient

supply chain and security of supply scale optimization in Germany (Piao et al. 2023). Primary

data was collected through a survey of 100 respondents. These were the key manufacturing

units in Germany across crisis-hit sectors. Automobile and IOT sector in Germany have been

chosen as the targeted group to collect data. Therefore, employee 100 employees have been

chosen as sample size to gather data. In this study, there was random sampling method was

used, and the sample size was 80. During this survey it is better to shorten the view and focus

on the German Automobile and Internet of Things (IOT) equipment manufacturing sectors.

DATA ANALYSIS

The frequency analysis and pie chart analysis have been done among the candidates. The

regression analysis has been done for data collection purposes and the coefficient, ANOVA table

and model summary table have been evaluated in this research.

Correlation Analysis

Correlation Test for “Machine Learning and Supply Chain Optimization” Variables:

Page 5 of 15

116

Archives of Business Research (ABR) Vol. 12, Issue 12, December-2024

Services for Science and Education – United Kingdom

Table 1: Correlation test for "Machine Learning and Supply Chain Optimization"

variables

(Source: SPSS)

The above table provides two tables of descriptive statistics and correlation values for two

variables that are related to "Supply Chain Optimization" and "Machine Learning.". In the

descriptive statistics table, the mean value for the supply chain optimization variable is 3.59.

Here, the standard deviation value for this variable is 1.01 where the number of observations

was 100. On the other hand, for the variable Machine Learning, the mean value is 3.51 which

are positive, and a higher positive value offered a higher scale of positive relationship among

variables. On the other hand, for this variable, the standard deviation value is 1.065 and this

value also signified this test.

This data provides information on the variability as well as central tendency related to mean

value giant these two variables that relate to the "Supply Chain Optimization" and "Machine

Learning" segments. In that case, the mean value suggested responses from the average number

of participants (Elshami et al. 2021). Both variables have higher mean values that suggested

higher responses from participants. On the other hand, through the help of this standard

deviation, this study offers information on the spread or of responses in the overcaution

numbers. In that case, both variables have lower values that demonstrated that less amount of

variability in the mean aspects.

In terms of Pearson correlation test refers to the test assorted with the relation where this test

measured the relationship between different two variables (Vallejos et al. 2020). The above

table also illustrated correlation coefficients between "Supply Chain Optimization" and

"Machine Learning." Here, Pearson correlation coefficients and their significance levels for a

two-tailed test have been done for these two variables for this study. Interms ofthe relationship

between supply chain optimization as well as Machine Learning, the person correlation

coefficient value is 0.720 that signified at a level of 0.01. On the other hand, if the value is

positive then there is positive relation has taken place between the two variables whereas if

there is a negative value that means -1 then it suggested a negative relationship between two

Page 6 of 15

117

Deore, A. R., Shafighi, N., & Hajibashi, A. A. (2024). Impact of Machine Learning on Supply Chain Optimization. Archives of Business Research, 12(12).

112-126.

URL: http://doi.org/10.14738/abr.1212.18101

different variables in this study (Kadoch et al. 2023). This correlation test indicates that a strong

positive relationship is observed between these two factors due to its positive value which is

0.720. This positive result stated that if there are increasing uses of machine learning, the

optimization in the supply chain segment. In that case, the significance level of 0.001 provides

information about this correlation that is significantly high at 0.01 level for the 2-tailed

bivariate test. This result also demonstrated that this kind of positive correlation is seen by a

lower chance (Gao et al. 2019). Therefore, the above data offers an effective positive

relationship between these two variables where the mean values for these variables are similar.

In that case, correlation results provide information that increasing uses of machine learning in

the supply chain aspects, this brought improvement in the optimization segment where an

organization gets different cost-effective supply chain practices.

Correlation Test for “Supply Chain Optimization and Advancement in Technologies”

Variable:

Table 2: Correlation test for “Supply chain optimization and Advancement in

Technologies” Variable

(Source: SPSS)

The above two tables provide results of descriptive statistics as well as correlation values for

two variables. Here, by analysing these tables, the correlation between two variables has been

discussed. In descriptive analysis, this result offers the distribution of these two variables in the

data set. Here, the mean value for the supply chain optimization variable was 3.59 which is

higher. This higher value stated that participants suggested this variable at a higher scale. On

the other hand, for this variable, the standard deviation value was 1.101 which is positive. On

the other hand, with the advancement in the technology variable, the mean value is 3.71 which

are similar to the other variable. In this context, the value for standard deviation was 1.176

among the observation number was 100. In that case, this statistical value not only offers a

tendency of the relationship between these variables but also shows the variability by the

participants against these two variables (Coccia, 2021). On the other hand, both these variables

had higher mean values which refer tothe acceptance ofthese two variables among participants

Page 10 of 15

121

Deore, A. R., Shafighi, N., & Hajibashi, A. A. (2024). Impact of Machine Learning on Supply Chain Optimization. Archives of Business Research, 12(12).

112-126.

URL: http://doi.org/10.14738/abr.1212.18101

helping in intelligent management, visualizations and automation purposes. Therefore, IoT

software is helping in information collection purposes and delivering proper visibility for real- time data analysis purposes.

Then again, the regression model equation is estimated as y=mx+C

Where Supply Chain Optimization is equal to 0.744 * Machine Learning +0.978 Here, Supply

Chain Optimization is denoted as y dependent variable

Machine Learning is denoted as an x-independent variable 0.744 is the slope m and 0.978 is the

constant C

Therefore, the model has indicated that there is a 0.978 unit change in y with every unit change

in x predicted.

Regression Analysis for Advancement in Technologies and Supply Chain Optimization:

Table 5: Regression Analysis for Advancement in Technologies and Supply Chain

Optimization

(Source: SPSS)

The above table is showing ANOVA table, where the F value is 240.702 and the significance

value is 0.001, which is less than 0.05. Along with this, the R square value is 0. 711 and R-value

is 0.843 which is close to 1. Therefore, modern technology is creating a positive impact on