See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/344161045

Review on Smart Supply Chain Management using IoT and Machine Learning

Research · September 2020

DOI: 10.15680/IJIRSET.2019.0809040

CITATION READS

1 801

3 authors, including:

Dhruvilsinh Jhala
Narsee Monjee Institute of Management Studies
3 PUBLICATIONS 3 CITATIONS

SEE PROFILE

All content following this page was uploaded by Dhruvilsinh Jhala on 12 September 2020.

The user has requested enhancement of the downloaded file.

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

Review on Smart Supply Chain Management

using IoT and Machine Learning
Dhruvilsinh Jhala1, Utsav Jariwala2, Deepti Barhate3
U.G. Student, Department of Computer Engineering, Mukesh Patel School of Technology Management & Engineering,
Shirpur, Maharashtra, India1
U.G. Student, Department of Computer Engineering, Mukesh Patel School of Technology Management & Engineering,
Shirpur, Maharashtra, India2
Assistant Professor, Department of Computer Engineering, Mukesh Patel School of Technology Management &
Engineering, Shirpur, Maharashtra, India3

ABSTRACT: SCM includes the integrated process planning and implementation needed to optimize the flow of
resources, data and financial capital. There is a delay in the transmission of data in traditional supply chain
management mode that seriously affects the effectiveness of supply chain management. Each company's objective is to
minimize costs for each process, so by using automation maximum efficiency is achieved. Supply chain management is
the backbone of global trade. This paper presents a framework of real-time monitoring of smart supply chains and
predicting the quality of goods in real-time. It is based on enabling a set of Internet-of-Things (IoT) and Machine
Learning based functions across the supply chain. IoT-based mechanisms such as RFID technology can improve
transportation and distribution efficiency. By taking real-time data and using different models of Machine Learning we
can improve the productivity of every process.

KEYWORDS: Supply Chain, Internet Of Things (IoT), Machine Learning.

I. INTRODUCTION

The idea of supply chain management emerged in the late 90’s which was widely used by industrialists for maintaining
and tracking goods. Supply chain management helps in maintaining the supply of the goods from manufacturer to the
final destination in an efficient way and also helps in maintaining the inventory.

Previously, producers were the supply chain drivers - dealing with the pace at which items were produced and
appropriated. Today, clients are calling the shots, and makers are scrambling to meet client requests for
choices/styles/highlights, speedy request satisfaction, and quick conveyance. Supply chain management is an essential
and significant component of the organisation to boost the organisations profit. Effective supply chain management has
become a possibly efficient way to secure competitive advantage and improve the efficiency of organizations.
Companies learning how to enhance their supply chain management will become the fresh success stories on the global
marketplace.

In this rapidly growing world, food requirements have become one of the main concerns. Food and Agriculture
Organization estimates that by 2050 the world's population will be around 9.1 billion. So, controlling food waste
have become very important. At the end of the 20th century, Kantor et al. (1997) researched the loss of food in the
United States of America. He found the food supply is around 355,883 million pounds, and 27% of that food
supply loses around 96,088,41 million pounds when calculated as numerical values [7]. The principal causes of
damage to food products during transportation are microbial illnesses, biological alterations, physical harm due to
inadequate circumstances and mechanical harm caused by maltreatment [4].

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9809

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

There are various technologies used for efficient supply chain management which are Machine Learning, IoT and
Block-chain. Machine learning helps in predicting the requirement of the goods, IoT helps in tracking the goods and
block-chain helps in maintaining the goods. The Internet of Things is a technology of interconnected computer systems,
machines, individuals that have distinctive recognizable information and transmit information over a network. In IoT,
RFID system i.e. Radio Frequency Identification is used which helps in identifying the goods. So, by combining these
technologies a supply chain network is formed which is smart and reliable.

IoT based Supply Chain

One of the main reasons which affect supply chain systems is our ability to know about sudden changes in real-time.
Here Internet of Things (IoT) comes to power in order to make supply chain management system more efficient. By
sensing changes in real-time we have the ability to take necessary action towards it and remove the time gaps that are
formed [1].

Figure 1: Gaps in supply chain [1]

As shown in Fig 1, in every supply chain the time gaps are formed which means information obtained from source
is delayed and this directly affects the decision-making ability of the system. Also, the decision made at any instance
takes time to implement which is decision-time gap. Therefore, to overcome these gaps in any supply chain
management system IoT is used. As an automatic terminal for collecting data, the RFID scheme is also the entry of the
Internet of Things information.

Application of IoT in Supply Chain

1) Authenticate the Location of Goods at Any Time

Specific storage containers, materials to be transported or products can be linked to sensors. The IoT device will
communicate its place that GPS satellites pick up and use to monitor goods location in real-time.

2) Track Speed of Movement and When Goods Will Arrive

Tracking motion flow and traffic flow makes predicting how goods move through the supply chain much easier.
Manufacturers, Suppliers, and warehouses can prepare to receive products, reducing handling times and ensuring
products are processed efficiently.

3) Monitor Storage Conditions of Raw Materials and Products

It is necessary to store certain products such as food in optimal conditions. Sensors can monitor temperature,
atmospheric exposure, light levels, humidity and other variables in the environment. If certain limits are exceeded,
these sensors trigger an alarm to the company.

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9810

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

4) Streamline the Problematic Movement of Goods

By tracking the goods and route planning over it, we can recognize where and when goods are in delayed through
IoT devices. This helps the supply chain to be accelerated by strategic planning and appropriate paths.

5) Locate Goods in Storage

Goods are tagged with sensors when they are in a warehouse, which makes it much easier to locate particular
products in a big warehouse and ensuring precise identification and goods distribution.

RFID system

RFID is an automatic non-contact identification technology called "radio frequency identification" that can
automatically identify the remaining or moving entities. The RFID system comprises primarily of 3 components reader,
tags and antennas [2].

Figure 2: Working of RFID system [2]

Tags: Tags are applied to all product quantities, from individual parts to shipping labels. The RFID tag itself is a
power-free microchip. Each product contains unique product code. When a product passes through the RFID scanners,
data from the tag is read that can contain any quantity of data, such as:
● Shipment Identity
● Order status
● Serial number
● Product Location logs
Reader: RFID reader is a gadget used to collect data from an RFID tag, which is used to monitor individual objects.
Radio waves are used to move information from the tag to a reader.

Antenna: An antenna provides the tags with data and commands

Supply Chain for Agricultural Products

According to a Billerud AB, study, about 10% of the new fruits and vegetables coming into European markets from
various parts of the continent are affected in the transport process. It is a major loss not only to the safety of food but
also economically of about 10 billion euros annually. By enhancing the transport method, damages can be reduced. The
principal causes of damage to food products during transportation are microbial illnesses, biological alterations,
physical harm due to inadequate circumstances and mechanical harm caused by maltreatment [3] [4]. To overcome
such problem following should be used:

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9811

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

- Smart Thermostats for Better Quality Assurance

Once an industry sends products for shipment, some temperature-sensitive products can become too hot, leading to a
despicable foodborne disorder that puts the public in danger. Temperature-controlled packaging may assist with
investment.

- Smart Consumer Appliances for Food Alerts and Inventory Reminders

Nearly every food item the customer brings home can be monitored by new intelligent products like refrigerators.
Smart appliances can also sense the temperature of the product and control it to maintain quality.

- Enhanced Precision Farming

Precision farming is a fresh concept embraced by farmers to define and establish the precise parameters required to
boost productivity using digital tools. These digital tools are used by farmers to collect data and other variables which
influence crop productivity. The information collected helps farmers to determine trends that assist to plan and predict.

Application of Machine Learning in Supply Chain

Machine learning is a sub-area of Artificial Intelligence that allows systems to learn and help improve from experience
automatically. By using different machine learning models we can predict the probability if the product will arrive on
time, if not then it calculates delay time depending on the situations, the probability of customer returning the goods,
and much more. By many machine learning prediction models we are now able to predict and overcome the loss by
taking action towards it [6][10].
- We can predict if the shipment likely to be delivered on time by using real-time data.

- If there are many returns from a specific area or for a specific product then we can find out the reason through
data analysis.

- Predicting the quality of product on arrival by sensing the temperature and weather conditions on the way by
using RFID sensors.

- By analyzing the patterns of customer demand for a product in a particular region we can ship the product
directly to customer eliminating the cost of the warehouse.

Next Generation Supply Chain

Today, nearly all sectors and most businesses are facing a more competitive climate – that is, increased demand
uncertainty, smaller product life cycles, increased demand for mass customization, higher competitive intensity, fewer
warehouses, more third-party service, globalization, etc [8]. The next generation of SCM should possess flexibility to
satisfy all the terms listed above. Robotics is revolutionizing all regions which operate in materials handling. It offers
great support to companies related to distribution such as Amazon or Ali Baba with unique positions same as supply
chain providers [5]. Jobs are a big and essential factor in any supply chain. So by increasing use of robotics, the amount
of human interaction decreases, which results in more cost effective supply chain system. A research by the
Information Services Group discovered that by 2019, 72 percent of companies would use robotic process automation to
decrease expenses, enhance productivity, and enhance compliance.

II. LITERATURE REVIEW

Marwa Chamekh et. Al [1] explain IoT based Tracking System For Supply Chain Management.
This paper describes the use in the business sector of IoT-based supply chain management. It presents an IoT based
traceability model and illustrates the distinct supply chain needs such as IoT data history, integrity, accuracy,

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9812

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

originality, optimal updating, and scalability by storing and querying information on IoT networks such as product
locations, sensor system communications, and camera manufacturing.

Bo Yan et. Al [2] introduced RFID and the IoT in the supply chain and discussed RFID, which can uniquely
identify an entity compared with other automatic identification technologies.The EPC technique is used to integrate
product data effectively with a general network dependent on the assimilation of the RFID scheme and network
appliance system. Data for the products stored in tags can be read by readers mounted on various supply chain
circumstances. Supply chain members are therefore able to request, update or share data on time through the network.

Yu Gu Et. Al [3] presented research in SCM for Fresh Agricultural Products using IoT. This paper describes how
the management of the SCM of fresh agricultural products is crucial to improving fresh produce companies'
competitiveness, and the appearance of items offers a new ability to handle the fresh agricultural supply chain.

Zhibo Pang et. Al [4] proposed the Tracking Service for Supply Chain by WAN and Sensors. This paper creates a
fresh system solution that is based on the global deployment of our wireless sensor system and RFID methods. It is
concerned with the coherent architecture WAN-SAN that helps offer the practical solution with low energy use,
reasonable prices and significantly better tolerance of communications in the sector.

Dr. Sini V Nair Et. Al [5] presented Supply Chain Management System of Amazon. This paper states that supply
chains in reaction to product life cycles or experience with a new industry can and should develop over time. Amazon
follows a customer-centered flywheel-business model that is particularly defined by focusing on selection, comfort
and cost and that the new customer-centered supply chain world will involve supply chain executives to adopt the
shift, drive it within their organisations, and learn how to handle it with their clients rather than concentrating
exclusively on expenses.

Rahul Agrawal Et. Al [6] explains how Machine Learning transforms Supply Chain Management. He
recommended architecture for Machine Learning Models in Supply Chain Management. By using different machine
learning models we can predict the future state of every product. By Machine Learning about 90 percent cases of
delays and returns were predicted correctly.

Linda Scott Kantor Et. Al [7] explained the wastage of food in the food supply chain on each and every level.
Initially it starts from farm itself, then during processing and transport, finally during retail and after consumer
purchases it. Many different techniques are given in this paper to reduce and overcome food wastage by recycling it
or using its byproduct. Field gleaning, Perishable food rescue or sal, Food rescue, Non-perishable food collection are
the types of efforts to reduce waste of food.

Vikas Misra Et. Al [8] explains how prerequisites have been proposed for the upcoming age of SCM frameworks
and how SCM operator based methods are a solitary answer for the organization of the upcoming age of SCM in a
global environment. The requirements for the next generation of SCM systems is described based on every
characteristic. Distributed Information Management System (DIMS) is one of the main factors as it manages
information over such big network.

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9813

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

Table 1: Summary of Literature Review Table

Sr. Author Title Source Technol Findings

No. ogy
Used
1 Marwa Chamekh, IoT Based Tracking IEEE-2018 IoT IoT based traceability model for
Sadok El Asmi, System For Supply Supply Chain Management.
Mohamed Hamdi, Chain Management
Tai-Hoon Kim

2 Bo Yan, Guangwen Supply chain IEEE-2009 IoT Introduces a model which can
Huang information efficiently fix information
transmission based on asymmetry in the supply chain,
RFID and internet of and create a network for supply
things chain transmission data.

3 Yu Gu, Tiaobin The IOT research in IEEE-2011 IoT How IoT has a beneficial
Jing supply chain function in the future
management of fresh development of the supply chain
agricultural products for new agricultural products.

4 Zhibo Pang, Jun Global fresh food IEEE-2010 IOT and Introduces system solution based
Chen, Zhi Zhang, tracking service Machine on our worldwide deployable
Qiang Chen, enabled by wide area Learning wireless sensor system and RFID
Lirong Zheng wireless sensor technologies by using WAN-
network SAN coherent architecture.

5 Sharanya, Dr. Sini Supply Chain IJLTEMAS- IoT and Study of Supply Chain
V Nair Management System 2018 Machine architecture of Amazon and
of Amazon Learning suggestions to improve it.

6 Rahul Agrawal, Using Machine White Paper, Machine Suggest architecture to improve
Avneet Saxena Learning to Transform TCS-2018 Learning Supply Chain efficiency using
Supply Chain Machine Learning
Management

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9814

 ISSN(Online): 2319-8753
ISSN (Print): 2347-6710

International Journal of Innovative Research in Science,

Engineering and Technology
(A High Impact Factor, Monthly, Peer Reviewed Journal)

Visit: www.ijirset.com
Vol. 8, Issue 9, September 2019

7 Linda Scott Kantor, Estimating and Econ papers- Analysis of wastage in American
Kathryn Lipton, Addressing America’s 2011 Food Supply Chain and suggested
Alden Manchester, Food Losses solutions to overcome it.
and Victor Oliveira

8 Vikas Misra, Supply Chain Journal of Study of Supply Chain

M.I.Khan , Management Systems: Strategic Management system and
U.K.Singh Architecture, Design Innovation proposing the next generation of
and Vision and SCM - Distributed Information
Sustainability Management System.
-2010

III. CONCLUSION

In this modern world we require new technologies with Supply Chain Management to meet with the growing demand
of the consumers. This paper presents the concept of Smart Supply Chain based on Internet-of-Things and Machine
Learning along with traditional supply chain architecture. It is now time to include more automation in the business
world. Various attempts on this path are found in the literature which helps supply chain management and
transportation in an efficient way. By technology mentioned in this literature we can monitor goods or raw materials as
well as update server data. The gathered information will be used to analyze the work for future reference. Initially, this
will benefit the company by saving them funds and allowing for wider product diversification, thus allowing for higher
productivity.

REFERENCES

[1] Marwa Chamekh, Sadok El Asmi, Mohamed Hamdi, Tai-Hoon Kim, “IoT Based Tracking System For Supply Chain Management”,2018 6th
International Conference on Wireless Networks and Mobile Communications (WINCOM),2018.
[2] Bo Yan, Guangwen Huang, “Supply chain information transmission based on RFID and internet of things”, 2009 ISECS International
Colloquium on Computing, Communication, Control, and Management – 2009
[3] Yu Gu, Tiaobin Jing, “The IOT research in supply chain management of fresh agricultural products”, 2011 2nd International Conference on
Artificial Intelligence, Management Science and Electronic Commerce (AIMSEC) – 2011.
[4] Zhibo Pang, Jun Chen, Zhi Zhang, Qiang Chen, Lir Zheng, “Global fresh food tracking service enabled by wide area wireless sensor
network”,IEEE Sensors Applications Symposium (SAS) – 2010.
[5] Sharanya, Dr. Sini V Nair,“Supply Chain Management System of Amazon”, International Journal of Latest Technology in Engineering,
Management & Applied Science (IJLTEMAS), Volume VII, Issue I, January 2018.
[6] Rahul Agrawal, Avneet Saxena, “Using Machine Learning to Transform Supply Chain Management”, TCS White Papers – 2018.
[7] Linda Scott Kantor, Kathryn Lipton, Alden Manchester and Victor Oliveira, “Estimating and Addressing America’s Food Losses”, Food
Review/ National Food Review, 1997, vol. 20, issue 1, 2011.
[8] Vikas Misra, M.I.Khan, U.K.Singh, “Supply Chain Management Systems: Architecture, Design and VisionJournal of Strategic Innovation and
Sustainability vol. 6(4), 2010

Copyright to IJIRSET DOI:10.15680/IJIRSET.2019.0809040 9815

View publication stats