Fin Irjmets1631603841

e-ISSN: 2582-5208
International Research Journal of Modernization in Engineering Technology and Science

( Peer-Reviewed, Open Access, Fully Refereed International Journal )
Volume:03/Issue:09/September-2021 Impact Factor- 6.752 www.irjmets.com
MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN FOOD INDUSTRY

Dr. Vikas Saxena*1, Amardeep Gautam*2
*1Associate Professor, Department Of FBM&ED, NIFTEM.
*2Student Of M.Tech. (FPOM), Department Of FBM&ED.
ABSTRACT
Artificial intelligence (AI) is the theory and design of computer systems capable of performing tasks normally
requiring human intelligence. With increased competition and growing demand in the food industry, the
industry has begun to use artificial intelligence technologies to maximize profits and discover new ways to
reach and serve consumers. Artificial intelligence has been successfully implemented in applications such as
sorting fresh produce, supply chain management, monitoring food safety compliance, efficient on-site cleaning
systems, predicting consumer preferences, and developing new products with greater efficiency and savings in
time and resources. The article focuses on travel and the application of machine learning and artificial
intelligence in various sectors of the food industry and provides an insight into the future prospects of machine
learning and artificial intelligence in the food industry.
Keywords: Artificial Intelligence; Machine Learning; Food Industry; Agriculture; Food Technology &
Processing; Supply Chain; Marketing.
I. INTRODUCTION
Whenever we discussing the food industry generally don’t consider technology in the food sector. But today in
the food industry, technology plays an important role in food production and delivery processes. With the use
of technology, we find foods through applications in our smartphones and we can get manufacturer information
also from that. The technology in the food industry also helps in improving product shelf life, designs various
types of packaging, and helps in maintain product quality and safety. With the use of technology, the production
costs are decreased and the quality of food significantly increased.
The fourth manufacturing revolution or industry 4.0 consists of the use of AI in food and drinks manufacturing.
AI is already penetrated many segments of food and drink production. Machine learning is transforming a
business in terms of food supply chain management, new product development, predictive maintenance, and
food safety compliance.
Food manufacturing is an operation that aligns industry. For staying in the competitive environment it needs to
update its technology and discover new opportunities to improve its operations and keeping more focus on
innovation technology. AI can make smart and fast-informed decisions that are based on what is learned from
complex calculations and data analysis.
Artificial intelligence and machine learning in the food industry help to optimize and computerize the process
which is carried out in the industry and saves money and reduces human error to a significant level.
Restaurants, Bars, and café like Starbucks as well as food manufacturing also endorsed artificial intelligence
and machine learning. Secure and precise product manufacturing lines with excellent speed and consistency
than human workers can be achieved by the use of AI technologies framework in the industry.
Around the world AI is also used in supply chains, AI is used in every process from start to end means from the
manufacturing process to the supply chain process in activities like tacking, overseeing stocks of product levels,
etc.
In this paper, we will discuss Artificial intelligence, machine learning, the evolution of AI & ML, AI applications
in the food industry, Present Scenario of AI in the Food industry, AI in food Safety, Benefits, and challenges of AI
in the food industry, Impact of AI on Food industry, and Future prospects of AI in the food industry.
II. WHAT IS ARTIFICIAL INTELLIGENCE (AI)
Artificial intelligence (AI) is the imitation of human intelligence and behavior in machines that are programmed
to think and behave like humans and copy their actions. AI may be called any machine which has traits like
human minds in terms of problem-solving and learning.
www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

[585]
e-ISSN: 2582-5208
The objective of artificial intelligence intake account of learning, analyzing, and perception. Artificial
intelligence is used in different industries including healthcare, finance, automobiles, and the food industry.
Mainly two types of AI are there weak AI and strong AI. Weak AI is having a tendency to do simple and single
task-oriented, while strong AI has a tendency to carries out more complex tasks.
The distinctive feature of artificial intelligence is that it has a tendency to diminishing and take actions that are
better for achieving a specific purpose. Machine learning is a subset of artificial intelligence that mentions that
the computer can learn automatically and adjust to new data without the intervention of human beings.
2.1 Different Fields of Artificial Intelligence
2.2 Types of Artificial Intelligence:-

AI is classified into two categories i.e., weak and strong
Weak artificial intelligence: - a system that is designed to carry one specific task. Examples of weak AI
systems are video games, Amazon’s Alexa, Windows10 Cortana, Apple’s Siri. You can ask them any question and
they will answer it for you.
Strong artificial intelligence: - a system that is designed to carry out complex tasks which are considered to
be human-like. These systems have a complicated and complex design. These systems are programmed to
control those situations in which human interfere is required to solve them. Applications of strong AI-based
systems are self-driving cars and hospitals.
III. WHAT IS MACHINE LEARNING (ML)
Machine learning is the branch of science in which computers are learned by self as we humans do, and they
develop their learning with time in self mode, by using the data and information which we provide them in the
form of observations and interactions with the real world.
Machine learning is a subset of artificial intelligence that is based on the idea that the system can gain self-
knowledge from data that is fed in it and identify the patterns and take decisions with minimal/no interference
of human beings.
3.1 Types of Machine Learning:-
Based on how an algorithm learns to become more accurate in its predictions, there are four types of ML. Data
scientist select a type of ML-based on that what type of data they want to predict.
[586]
e-ISSN: 2582-5208
1. Supervised learning- defined variable and algorithms with labeled training data is supplied by data
scientists to access the correlation. In this type of machine learning, the algorithm of both input and output are
specified,
2. Unsupervised learning- train on unlabeled data is used in this type of machine learning. Data
algorithms of input and output are predetermined.
3. Semi-supervised learning- it is the mixture of supervised and unsupervised learning, mostly the
labeled training data is supplied by the data scientist. In this learning, the system can explore data on its own
and can generate its understanding for the dataset.
4. Reinforcement learning- it is used to teach a machine by completes a multi step-process by clearly
defined rules. Data scientist creates an algorithm to finish a job. In this, algorithms decide their channel for
what steps should take to complete the task.
IV. HISTORY OF ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING
In modern business, machine learning has a great impact but it’s important to know that it is not new it has a
vast history.
1943- Neural network was born by neurophysiologist Warren McCulloch and neurophysiologist Warren
McCulloch in their paper about neurons.
1950- A Turing test had been created by Alan Turing for the computer to convince that it is a human, not a
computer.
1952-Arthur Samuel created a program by which checkers had been played.
1958- Perceptron the first artificial neural network came into existence by Frank Rosenblatt.
1959- In this year Bernard Widrow and Marcianh off brought two models ADELINE (binary pattern detector)
and MADELINE (eliminator of echo on the phone line).
Till the late 1970s no further more innovation came into existence as MADELINE had great success then Von
Neumann's architecture placed the same memory for data and instruction.
1982- A bidirectional line network suggested by john Hopfield brought the growth of interest in the neural
network. In the same year, Japan announced an advanced neural network which brought more interest for the
researcher.
1986- Backpropagation used by neural networks came into existence by three researchers from the Stanford
psychology department by extending the algorithm created by Widrow and Hoff in 1962.
1997- A chess-playing computer IBM deep blue came into existence to beat the chess champion.
1998- AT&T bell laboratories researched digital recognition which brought great accuracy in the detection of
handwritten products from the postal service of the US.
2012- Jeff dean of Google introduced a deep neural network by which detection of patterns in image and video
was focused. In the same year, the ImageNet competition was won by AlexNet, which led to the use of GPUs and
Convolution Neural Network in machine learning.
2014- Facebook created a deep neural network that can identify with some precision as a human can. In the
same year, Google bought a deep mind which can play a basic video game by resembling the same level as a
human.
2015- Elon musk and others created a non-profit organization to innovate safe Artificial Intelligence for a better
future for human beings. This year Amazon introduced a machine learning platform to show how big
companies are involved in machine learning. CNN architecture created U-net a specialized biomedical image
segmentation is introduced.
2017-amazon used Nvidia to boost up Amazon web service machine learning.
V. PRESENT SCENARIO OF AI
5.1 Status of AI in India
In reference to a research conducted by Analytics India Magazine (AIM) in association with Jigsaw Academy, I'll
be discussing the developments of the Artificial Intelligence market in India and around the world. The firms
[587]
e-ISSN: 2582-5208
included in this research are Indian Corporations (domestic conglomerates and domestic IT firms), Hi-Tech
Electronic/Chip & Semiconductor firms, MNCs (including engineering and technology firms), Banking &
Financial (Public sector, Private sector, and captives), Communications & Media enterprises, Energy, Industrial,
Healthcare, Mobile Payments (Fintech), and e-commerce firms.
In this report, the salary figures are represented in INR and the Market value figures are represented in USD.
 As of July - August 2020, the value of the Indian Artificial Intelligence market is stated to be worth $ 6.4B.
 The AI market share and size concerning the Types of Companies are the highest throughout the broad MNC
IT/Technology/Electronic category, which comprises High-end Software and Hardware technology, IT
services, Semi-conductor, and Electronic firms. The combined market share is 36.2%.
 Throughout the IT Services sector, the market size by Industries or Sectors is the highest which is followed
by the Technology Sector that includes both Hardware and Software firms, with a market share of 41.1%
and 23.3% respectively.
 The BFSI sector has 9.6% of the market share of AI services.
 There is nearly 91000 AI personnel that are working throughout enterprises in India, with an average
income of INR 14.7 Lakhs.
 The highest median salary is INR 16.7 Lakhs which is drawn by AI professionals in Mumbai.
 As of July 2020, in India, there are nearly 16500 open positions related to AI that are to be filled. Bangalore
has the highest proportion of open jobs just as it does for other Data Science and IT services roles.
The reports include various sectors but the contribution to the Indian AI market by the FMCG sector is $55.2M
in market size and 0.9% in market share. Both Domestic and FMCG companies are covered by this sector. The
largest provider to this sector is 'Unilever'. Apart from Unilever, there are other contributors to the AI market in
India which include the MNC firms like Pepsi, Kraft, Coca Cola, and Proctor & Gamble (P&G), etc., and Domestic
FMCG players like the ITC. There are about 170 AI-related start-ups in India as per a report by Zinnov. These
start-ups have obtained an investment of $36 million. Out of these, 64 are situated in Bangalore. These start-
ups are working in the areas of finance, e-commerce, healthcare, etc. A start-up called 'Tuplejump' assists its
clients to visualize the data while making a decision. This start-up is acquired by Apple. ClearTax is working on
developing a solution for e-filing using documents directly. AIndra is specializing in developing devices with
computer vision ability for applications like detection of cervical cancer, facial recognition, etc. IT majors like
Infosys, TCS, etc. are developing solutions that are AI-based for their needs. TCS has developed a Virtual
Assistant that could interact with the clients regarding insurance products (Tata Consultancy Services, 2012).
One of the intriguing aspects of this Virtual Assistant is that it can interact with others in spoken natural
language. Using AI-based solutions, Infosys has automated various IT support processes. IBM has installed
Watson for Oncology products in several hospitals to be used in the treatment of cancer patients. Some of the
examples of start-ups are given in table no 1.
Table 1: Indian Start-ups in Artificial Intelligence.
Start-up URL Task
Edge Networks www.edgenetworks.in Matches the job seekers with jobs available
Fluid AI www.fluid.ai Provides customer information on the products in an

interactive way
Flutura www.flutura.com Monitors health of machines to advise on

maintenance
Heckyl www.heckyl.com Analyses stock related information to advises on

stock trading
Mad Street Den www.vue.ai Helps customers locate products using captured
photos

[588]
e-ISSN: 2582-5208
ShopR360 www.shopr630.com Video analytics solution which can distinguish staff

and customers
SigTuple www.sigtuple.com Affordable diagnosis solution using a microscope, cell

phone, and cloud
Social Cops www.socialcops.com Interpretation of data
VPhrase www.vphrase.com Converts structured data such as graphs, etc. into

words
Freenome www.zephyrhealth.com Analyses genetic material to find disease signature at

an early stage
JISTEM USP, Brazil Vol. 15, 2018, www.jistem.fea.usp.br

Since a while, the government has been investing in R&D in AI. With financial support from the United Nations
Development Programme (UNDP), the Ministry of Electronics and Information Technology (MEITY) started a
knowledge-based computer systems project in 1986. Later on, the Ministry started another program called
National Programme on Perception Engineering. This program has developed some prototype systems like
robotic arms, etc. Technology Development for Indian languages Programme of the Ministry has taken up some
projects, too. A large number of funds are used on machine translation, speech-to-text, and text-to-speech
systems.
DRDO has been investing in AI projects at the Centre for Artificial Intelligence and Robotics (CAIR) for defense
and civil applications. Funding of R&D by industry is limited to few companies only. Nonetheless, the situation
is changing now. Recently, Infosys has contributed Rs. 50 million for research in AI to Indraprastha Institute of
Information Technology (IIIT).
Some companies have recently declared that they will be releasing the workers or hiring fewer workers. This
will have a serious impact on job opportunities as the percentage of people employed in low-skill jobs is high.
We can see that the effect of automation has begun to reflect on the operations of IT behemoth Infosys. As per
Krishnamurthy Shankar, the head of Human Resources, because of automation of lower-end jobs, the firm has
"released" about 8000 to 9000 employees in the past 1 year. Services that are most likely to be affected are the
BPO type of services and the downside of this is that a huge number of people are employed in this type of
service. At the same time, the demand for IT professionals is going to decrease due to the automation of various
tasks. The most affect IT professionals will be those who do routine white-collar jobs.
5.2 Status of AI in World:-
As per the latest Artificial Intelligence Index report, adoption of AI among large companies, in 2019, has
increased by 47% compared to 2018. According to the same report, last year, the global private AI investment
exceeded $70 billion. This highlights the increasing interest in AI software and start-ups. The Artificial
Intelligence Index report states that every 3.4 months, the AI computing power keeps doubling.
To estimate the present level of adoption of technology, it is necessary to look into the applications that have
been developed across the world.
● Healthcare: -For quite some time, AI has been used in the domain of healthcare. As per a report, more
than 100 start-up companies are working on the application of AI in the domain of healthcare. The virtual nurse
is an application that has been used in various types of environments. The virtual nurse can perform tasks that
are normally performed by a human nurse. Molly is one of Sensely's virtual nurses that uses conversational AI
to draw the attention of the people towards better health. Sensely is a start-up that contributes to AI
development.
Watson for Oncologists which is developed by IBM helps a doctor in searching for the relevant material from a
large number of documents or papers that could be useful for the case at hand. Watson for Oncologists is very
helpful as it analyses both structured and unstructured data. If the case at hand is similar to any case reported

[589]
e-ISSN: 2582-5208
in any document/publication and a specific treatment plan is effective then the reported information may be
useful for the treatment in the present case.
● Education: -According to the predictions done in a study, there are possibilities that AI might
transform education in the next 4 to 5 years. Today, it is possible to make bots that are similar in appearance to
a human being. A professor at Georgia Tech created an AI-powered system to work as his assistant in teaching
to assist him in a course that was taught by him. The AI-powered system interacted with students just like a
human teaching assistant. It has become possible to determine the subjective answers given by the students.
This helps in saving an ample amount of time and brings uniformity to the assessment. Bots are used to assess
the subjective answers given by the students appearing in the GRE test which is administered by ETS. Though
there is the availability of a large amount of digital learning material, the usage of computer-based education
remains limited due to the lack of the possibility of personalization. A human teacher can personalize the
content as per the needs of the student since every students' level of knowledge and aptitude differs from one
another.
● Cyber Security: -The situation in Cyberspace keeps on changing rapidly and is not easy to predict,
which makes its environment dynamic. Since AI systems have the flexibility to respond to the changing
environment, it is increasingly used in all the stages in cyber defense namely early warning, prevention,
detection, and response. As the processes are automated by AI, it helps in reducing human intervention.
Because of developments in AI technology, today, it has become possible to create artificial police agents that
help in monitoring the entire network. Intelligent agents are used to carrying out this work. An intelligent agent
is an application that can sense, reason, and act autonomously. These agents can also interact with each other to
collaborate to carry out a task.
● Finance: -Banks use Chabot to perform simple tasks like account activation or balance checking, etc. It
also helps the customers who are not fully aware of IT systems and would like to interact in natural language.
Banks like HDFC Bank, Axis Bank, State Bank of India, and ICICI Bank use AI-based applications to provide
customer services in India.
The Singapore Government has announced a grant of $19.9 million for the banks in Singapore to promote AI
and data analytics as vast amounts of data on stock trading is publically available. Therefore, it is an excellent
area to use AI. Sentient Technologies Inc. is working on an AI system for taking decisions in stock trading. It has
a team of engineers who have worked for Microsoft, Apple, Google, Amazon, etc. Other firms like Wealthfront
and RBS, etc. are exploring the idea of using AI in hedge funds. Some applications that have been built, help in
the preparation of tax returns. An example is ‘AskMyUncleSam’ Chabot which uses AI to answer tax questions
and financial queries.
● Information Browsing: -Today, AI is embedded in the normal systems of daily use and is not visible to
the users. For example, the Google search engine uses AI algorithms to predict the content that the user may be
interested in. It helps the user by bringing the most relevant documents at the top of the search that makes it
the best search engine. Search engine companies use the information on the user preferences to display the
most relevant advertisements apart from information retrieval. Google monitors the news articles that the user
reads and uses this knowledge to display the news that would interest the user.
● Transport: - One of the high-potential applications of AI would be self-driving cars. Self-driving cars
are more likely to cause accidents much less than human-driven cars.
● Virtual Assistants: -Virtual Assistant, today, is used by many around the world. It is a remote office
assistant. It assists the users in carrying out the tasks and also performs particular tasks on their own in an
autonomous way. In the past years, many Virtual assistants have been created by many companies. For
example, Siri from Apple, Alexa from Amazon, Cortana from Microsoft, and Google's Assistant. It is also easy to
interact with a virtual assistant as the users can use natural language while interacting. Even an illiterate or
blind person can use it as one can interact with spoken words. It is also possible to use it while driving because
of the speech recognition feature. Many suppliers are trying to install these assistants in vehicles. Siri has
already been integrated with some cars. Nissan is working on integrating Microsoft's platform into its cars.
Also, Hyundai is integrating with Alexa and Google Assistant.
[590]
e-ISSN: 2582-5208
● E-Commerce: - There are many product and service recommender systems like the one Amazon uses
on its shopping portal. It helps the users to discover new products and services by keeping a track of items that
have been purchased by the user over a while. It also identifies certain patterns to keep a note of the items
purchased by the user to decide what products and services might be of interest to the user. These patterns are
not fixed and hardcoded in the system. They are created by using machine learning techniques.
Today, Chabot is used widely by many companies. It is used to order products and services online. A Chabot has
also been developed to order coffee from Starbucks. It is convenient as the customer can order coffee using
spoken natural language to describe the type of coffee they want and the order is sent to the closest Starbucks
unit. Payment is also easy to do using the pre-registered credit/debit card.
VI. APPLICATION OF ARTIFICIAL INTELLIGENCE IN THE FOOD INDUSTRY
Artificial intelligence will be playing an important role in food production in the future. To meet the consumer
demands Food and beverages companies shifting very rapidly on Ai technologies for operations and logistic
efficiencies. Production, financial, marketing, consumption, packaging, distribution, and storage having a major
impact due to the use of AI.
Before 2021, the annual growth of food and beverage companies is estimated at a compound growth rate of
42% in the global AI market. Food companies generating increased their revenue and productivity by using
Artificial intelligence systems.
The food industry is gaining significant development from the latest Artificial intelligence which is used for
sorting of fresh produces, keeping top grade health, and maintain food safety compliances, the latest AI helping
in developing new products, and with the help of the AI supply chain activities are also improving.
The food industry is divided into 4 major segments are
1. Agriculture
2. Food Technology and Processing
3. Supply Chain
4. Marketing.
We will discuss the application of AI in each sector one by one.
6.1 Application of AI in Agriculture:-
In recent times the use of Artificial intelligence (AI) has been evident in the agriculture field. To improve its
yield includes pest management and disease infestation, improper soil treatment, low output, and the lack of
knowledge technology in farmers. The main concept of artificial intelligence is its accuracy, effectiveness, better
performance, and flexibility. Artificial intelligence used in;
1) Soil management
2) Crop management
3) Weed management
4) Disease Management
5) Pest Management
6.1.1 AI in Soil Management
Soil management and irrigation are very important issues in agriculture. They lead to poor quality and crop
failure.
Management-oriented modeling (MOM) reduces nitrate leaching because it includes a set of cause feasible
management alternatives, each alternative is evaluated by a simulator, and a best-weighted alternative which is
fulfilling the user criteria is select by an evaluator. Soil Risk Characterization Decision Support System
(SRCDSS) involves three stages for its development which are knowledge acquisition, conceptual design, and
system implementation. Artificial Neural Network (ANN) model based on characteristics of existing rough
resolution of soil maps incorporate with hydrographic parameters obtained from Digital Elevation Model
(DEM) predicts the soil texture (sand, clay, and silt contents). A remote sensing device is set in a Higher-Order
Neural Network (HONN) estimate and characterized the dynamics of soil moisture.

[591]
e-ISSN: 2582-5208
Table 2: AI in soil management techniques.
Technology Strength Limitation
MOM Minimizes nitrate leaching, maximizes Takes time, limited only to nitrogen
production
Fuzzy Logic: Can classify soil according to associated Needs big data
SRC-DSS risks
DSS Reduce erosion and sedimentary yield Requires big data for training
ANN Can predict monthly mean soil temperature Considers only as a factor for soil
performance
Can predicts soil texture Requires big data for training
Can able to predict soil moisture The predictions will fail with time
as weather conditions are hardly
predictable
Cost-effective saves time has 92% accuracy Requires big data
Can estimate soil Nutrients after erosion Its estimate is restricted to only
NH4
6.1.2 AI in Crop Management
Crop management is a practice in agriculture that starts with sowing the seed and carries on with keeping an
eye on the growth, harvesting, and storage of crops and distributions.
In other words, it is the activity used to improve the yield of crops. An agricultural management system named
Precision Crop Management (PCM) is designed to find outcrop and soil inputs as per the requirement of a field
with the aim of optimization of profitability and protect the environment. To overcome the water deficit caused
by soil, weather or limited irrigation farmers should use the combination of different crop management
strategies.
High crop quality yield can be achieved by understanding the weather pattern properly. For estimation of
operation behavior of field weather data, labor availability, machinery capacities are used by PROLOG (an AI
programming language). PROLOG can analyze crop production, gross revenue, and net profit for separate files
as well as for combined fields.
Crop prediction methodology helps forecast the suitable crop by analyzing various parameters of soil and
atmosphere parameter.
Demeter is a computer-controlled speed rowing machine that is equipped with a set of videos cameras and a
global positioning sensor for navigation. It having the ability to plan the harvesting operations for the whole
field and then accomplish its task by crop rows cutting, turning to cut successive rows, and detecting
unexpected obstacles.
The use of AI in harvesting cucumber consists of separate hardware and software
Components of the robot incorporate the autonomous vehicle, the manipulator, the end-effector, the two
computer vision systems for detection and 3D imaging of the fruit and the environment, and, finally, a control
scheme that generates collision-free motions for the manipulator during harvesting.
Table 3: AI in crop management techniques
CALEX Can formulate scheduling Time-consuming
guidelines for Crop management
activities
PROLOG Removes less used farm tools Location-specific

[592]
e-ISSN: 2582-5208
from the farm
ROBOTICS Can harvest up to 40 hectares of Expensive: use a lot of fuel
(Demeter) the crop
ROBOTICS Has 80% success rate in Slow picking speed and accuracy
harvesting crops
ANN Predicts crop yield Only captures weather as a

factor for crop yield
Above 90% success rate in A little number of symptoms

detecting crop nutrition were considered
disorder
Can predict the response of Consider only soil temperature
crops to soil moisture and and texture as factors
salinity
Can accurately predict rice yield Time-consuming
FUZZY Predict cotton yield and improve It is relatively slow

Cognitive map crop for decision management
FUZZY LOGIC + ANN Reduce insects that attack crops Shows inability to differentiate
between crop and weed
6.1.3 AI in Weed Management
Implementation of herbicides in the field has consequences on human health as well as on the environment. A
report says that if the weed invasion is not managed then it caused a 50% reduction in dried beans yield and
wheat yield about 48% loss is there due to weed.
According to the Weed Science Society of America (WSSA) report, weed effects include flooding during the
hurricane, some species of weeds can cover their way during a rampant fire, some weeds can cause
unrecoverable liver damage if they eat up, and they muscle out plants or crops by competing for water,
nutrients, and sunlight. Several weeds are toxic and can cause allergic reactions or even they can be a threat to
public health. Herbicide use can be reduced with the help of Modern AI methods by proper and precise weed
management.
A rule-based expert system is designed to identifying and eliminate the weed in crops like wheat, barley, oats,
and triticale by Pasqual. Machine vision combined with a trained neural network is used to identify weeds in
five distinct species by Burks by doing a comparison between three different neural network systems
backpropagation, and radial propagation and founds that the backpropagation system gives the best accuracy
of 97%.
Table 4: AI in weed management techniques
ANN, GA High performance, reduce trial Require big data
and error
Optimization using invasive Cost-effective, enhanced Adaptation challenge with new
weed optimization (IVO), ANN performance data
Mechanical Control of Weeds, Saves time and removes Expensive, constant use of the
ROBOTICS, Sensor machine resistant weeds heavy machine will reduce soil

[593]
e-ISSN: 2582-5208
learning productivity
UAV, GA Can quickly and efficiently Has little or no control over
monitor weeds weeds, expensive
Saloma expert system for High adaptation rate and Requires big data and usage
evaluation, prediction, and weed prediction level expertise
management
Support Vector Machine (SVM), Quickly detects stress in the crop Only detects low levels of
ANN that will prompt timely site- nitrogen
specific remedies
Digital Image Analysis (DIA), Has above 60% accuracy and Time-consuming
GPS success rate
UAV (Unmanned High rate of weed detection Expensive and requires vast
Aerial Vehicle) within a short time human expertise
Learning Vector Quantization High weed recognition rate with The method of data used
(LVQ), ANN short processing time affected the AI’s performance
6.1.4 AI in Disease Management
Disease in the crop is also a subject of problem for farmers. To detect imperfect plants and to take important
actions for curing particular expertise and experience is required. For controlling and diagnosing the disease
computer-aided systems are used worldwide. Integrated disease control and management model which
combined with physical, chemical, and biological measures should be adopted by farmers for effective control
on disease and reduce losses. An innovative application based on the Fuzzy Logic of rule promotion is applied
for drawing intelligent inferences for crop disease management.
Table 5: AI applications in disease management
Computer vision system (CVA), Works at high speed can Dimension-based detection may
Genetic algorithm (GA), ANN multitask affect good species
Rule-Based Expert, DataBase Accurate results in the tested The inefficiency of DB when
(DB) environment implementing on a large scale
Fuzzy Logic (FL), Web GIS Cost-effective eco friendly Inefficiency due to scattered
distribution takes time to locate
and disperse data
FL Web-Based, Web-Based Good accuracy, responds swiftly Limited usage as it requires
Intelligent Disease Diagnosis to the nature of the crop disease internet service, its potency
System (WIDDS) cannot be ascertained as only 4
seed crops considered
FL and TTS converter Resolves plant pathological Requires high-speed internet,
problems quickly uses a voice service as its
multimedia interface
An expert system using rule-base Faster treatment as diseases are Time-consuming, needs constant
in disease detection diagnosed faster, cost-effective monitoring to check if the pest
based on its preventive approach has built immunity to the
preventive measures
ANN, GIS 95% accuracy Internet-based, some rural
farmers will not have access

[594]
e-ISSN: 2582-5208
FuzzyXpest Provides pest information for Internet-dependent

farmers, also supported by
internet service, high precision in
the forecast
Web-Based Expert System High performance Web and internet-based
ANN Has above than 90% prediction ANN doesn't kill infections or
rate reduces their effect
6.1.5 AI in Pest management
Insect pest invasion is one of the most terrifying problems in agriculture which can cause a huge economic loss.
Over the past decades, researchers are trying to develop a computerized system that could identify the pests
and could able to suggest control measures.
SMARTSOY by Batchelor et al., CORAC by Mozny et al., Fuzzy logic-based system by Saini et.al, IPEST by Hayo
are some of the AI-based technologies used for pest management. For pest management is Tea, TEAPEST is
used.
6.2 AI applications in Food Processing
Food processing and management can cause huge alterations in the food economy. Various factors, like quality
control, type of food, current trends, customer psychology, and human wellness, play an important role in
influencing the food processing industry. This lead to the need for the inclusion of technologies to improve
production values, waste management, and satisfying market demand.
AI has been installed for the following applications -
1. Sorting fresh produce,
2. Food safety compliance monitoring,
3. Effective cleaning in place systems,
4. New product development
5. Food packaging
6. Food waste
6.2.1 AI in sorting fresh produce
When it comes to sifting and sorting vegetables, food processing industries depend on manual sorting which
leads to a lack of efficiency and increased costs. Such an issue can be solved with the use of Artificial
Intelligence. With the proper use of Artificial Intelligence, food processing companies can strive in food
cataloging, using a combination of cameras, lasers, and machine learning to enable food sorting with enhanced
efficiency. For example, if artificial intelligence involving sensor-based optical sorting solutions is deployed
then the time-consuming processes of sorting fresh produce can be removed which will lead to higher yield
with better quality and lesser wastage. To deal with various product sizes and reduce waste and costs, AI can be
used as a mode to better calibrate machines. A sorting plant is required to segregate millions of vegetables
depending on their shape, size, and color.
Until the end of the 20th century, about 90% of the food was being sorted by humans, says a leading sorting and
collection solutions provider in Norway, TOMRA. TOMRA uses X-ray, NIR (Near Infra-Red) spectroscopy, Laser,
cameras, and a unique machine-learning algorithm to analyze different aspects of a vegetable or a fruit for
sorting. Similarly, Aromyx uses AI to quantify taste and smell to increase production efficiency to create "a new
quantitative and visual standard to represent senses of taste and smell as actionable data". Another similar
technology includes electric noses which can be used instead of human noses to distinguish various foods'
aromas and odors by using AI sensors. For example, Japan's Kewpie uses Google's Tensorflow AI to detect
defective ingredients during processing. Initially, it was used for sorting foods, and gradually it was developed
into an anomaly detector. Qcify, a dutch company, uses machine vision systems to classify almonds and
pistachios, etc.

[595]
e-ISSN: 2582-5208
6.2.2 AI in Food safety compliance monitoring
The key avenue for AI is the reduction of pathogens and detecting toxins in food production. A Kankan
subsidiary uses AI-enabled cameras to ensure safety compliance amongst the food workers in the food facility.
This enables object-recognition and facial-recognition software to understand whether the workers can adapt
with good personal hygiene as required by food safety law. It also extracts images of the screen for review in
case a violation or violations are found which can be rectified in real-time. This technology can give an accuracy
of more than 96%. In line with the effects of Covid-19, the company added developed facial recognition abilities
to account for the compulsory use of a mask, and new body temperature detection.
The Japanese company Fujitsu has developed an AI-based model that is used for monitoring hand washing in
kitchens following strict rules set by the Japanese health ministry. This technology will help in reducing the
need for visual checks, especially where food safety must be increased. Next-Generation Sequencing (NGS) is
another attempt by the food safety associations to increase the speed and accuracy with which any kind of
threats to food safety are detected and dealt with in a production chain.
6.2.3 Effective cleaning in place systems
The most challenging task in food processing sectors or industries is maintaining proper hygiene. Due to the
use of automation processes, many food processing companies claim that they are clean like white paper. As
per the University of Nottingham, about 30% of the energy and water supplies of a food processing plant are
used in the cleaning of equipment. They claim that they can save nearly $133 million per year and also save
time, energy, and water to clean equipment because of their AI-based technology.
AI-enabled technologies to optimize the cleaning process by using ultrasonic sensing and optical fluorescence
imaging which measures food residue and microbial debris in a piece of equipment.
6.2.4 AI in New product development
To understand consumer flavor preferences and predict their response to new tastes, AI technology uses
machine learning and predictive algorithms. The data received can be segregated into demographic groups
which will help the companies to bring new products to the market that match the preferences of their target
audience.
Coca-Cola has set up self-service soft drink fountains across many restaurants and other places. These fountains
allow people to customize their drinks. Theoretically, consumers can create hundreds of different drinks by
adding various flavors to their base beverages through these machines. These self-service fountains created a
huge amount of customer preference data which Coca analyses by using AI. With the help of this data, the first
product they launched was 'Cherry Sprite.'
In 2019, news headlines covered that IBM created and deployed AI-based solutions for McCormick and Cerealto
Siro Foods. IBM research developed an interactive tool for flavorists that helped them in recommending new
seasoning formulas, drawing on the existing proprietary data of McCormick's. IBM Services provided a
trendspotting resource that gave inspiration for new products. This trendspotting resource uses publicly
available information such as social media. In both instances, IBM departed from the approach which was built
on the flavor pairing hypothesis that was used with its Chef Watson project.
6.2.5 AI in Food packaging
Food packaging and handling are essential in the procedure of moving the raw food materials from farms to
distribution centers. Food packaging and handling are responsible for the preservation and accurate
representation of foods.
Most of the foods sold are usually packaged. In such matters, AI can be more helpful in ensuring protection and
health across the food industry. It can also help in the product development of food and beverages by
recognizing the most profitable and eco-friendly packaging for food products. As there can be effects of
packaging on foods, packaging of the food in certain ways is limited. For instance, specific inks cannot be used
in the packaging of some high-fat foods like cheese as it can cause chemical contamination. In cases where glass
and plastic are used for packaging, great care is taken to ensure that no contamination or deformation might

[596]
e-ISSN: 2582-5208
affect the food. Also, glass must have proper coatings. Visual sensors can also be used to find the best way of
packaging food to minimize the plastic needed to cover food products with abnormal shapes.
6.2.6 AI in Food Waste
Artificial Intelligence can also be important in managing food waste and help reduce world hunger. According
to USDA, around 30-40% of food supplied in the US becomes food waste. USDA’s Economic Research Service
estimates that in 2010, 31% of food waste found at the retail and consumer levels is equal to 133 billion pounds
of food. It is also projected that over 2 billion people will suffer from hunger and will be undernourished
globally by 2050.
Food waste AI will help us to understand overproduction and allocate the correct resources to meet supply and
demand properly while also adapting to various population needs. AI can also control the number of production
materials used in correlation with weather forecasts to create the maximum amount of product needed for
distribution.
A huge number of foods are rejected by suppliers and farmers because they don't suit certain criteria needed. In
such a case, using a food tracking app with AI can be helpful. With the help of such an app, the food from the
farmers can be connected directly to food pantries, stores, and restaurants which will result in the reduction of
waste food as the less desired and aesthetically unacceptable crops can be sold before they become useless. AI
can also sort these crops and put a different transportation route to make sure it is delivered correctly.
As a result, the contribution of artificial intelligence and machine learning in the food industry will create more
efficient systems that will benefit the economy as well as improve the health of customers, the field, and the
world.
6.3 Application of AI in the Supply chain:-
As demand uncertainty increases, supply risk increases and competition increases, supply chain (SC)
advantages often depend on a firm's ability to integrate and manage various processes or materials or part
sources to produce finished goods. Delivery makes customers. As these functions are often enhanced by better
end-to-end transparency of SC processes, many leading companies have tried to add sources of information and
exchange real-time information with SC partners. As a result, SC Management (SCM) is becoming stronger and
focuses on replacing assets (eg inventory, warehouse, vehicles) with information. SC professionals recognize
the importance of data for SC's success and are looking for ways to better manage information and develop
better business solutions. One such capability is AI, which has been around for years but is not fully used by
SCM.
Artificial intelligence and machine learning (ML) have transformed the supply chain industry, deepening the
gap between winners and losers. By eliminating deeply ingrained inefficiencies and uncertainties, AI and
machine learning enhance business transparency in all aspects of the supply chain, providing details and
methodologies that no one else can replicate at scale. AI in the supply chain provides the powerful optimization
needed for more accurate capacity, performance, and quality planning. Reduce costs and increase productivity
while maintaining safer working conditions.
Artificial intelligence is used in the supply chain to reduce delays and maximize profit margins by keeping an
eye on each supply chain operation and control them. To beat the competition companies used AI for finer
forecasting the demand and to checking stock availability and superior management of pricing.
Nowadays people are discussing the transparency of products, Artificial intelligence will also be used to track
the products from the starting step i.e., farm to the final step i.e., consumer to make sure transparency. By
appropriate use of AI in food production and distribution channels the companies will be sure that the flow of
goods will be streamlined and cost-effective.
6.3.1 Chatbots for Operational Procurement
Chabot can be described as an AI-based computer program that mimics human conversations. They are also
known as digital assistants who understand human potential. The bots translate and process user requests and
provide the right quick responses.
For daily activities, Chatbots can be used to:

[597]
e-ISSN: 2582-5208
 Speak to providers during minor discussions.
 Set up and submit actions to providers for governance and compliance.
 Apply for a purchase.
 Research and answer internal questions about the functionality of a product or set of suppliers/suppliers.
 Receiving / submitting / documenting invoices and payments / order requests.
6.3.2 Machine Learning (ML) for Supply Chain Planning (SCP)
Procurement planning is an important function within the SCM strategy. Having the tools for smart work of
building concrete plans is a must in the current business world.
The ML, used within the SCP can assist in in-product forecasting, demand, and delivery. Used properly with
SCM work tools, ML can transform the elegance and efficiency of purchasing decisions. Using ML technology,
SCM specialists - who oversee the SCP - will be providing excellent scenarios based on intelligent algorithms
and machine-to-machine analysis of large data sets. This type of power can increase the supply of goods while
measuring supply and demand, and would not require human analysis, but rather a set of effective parameter
measures.
6.3.3 Machine Learning for Warehouse Management
If you look closely at the SCP domain, its success depends largely on proper storage and inventory-based
management. Regardless of forecasting demand, supply errors (excess or understock) can be catastrophic for
almost any customer-based company/seller.
"A prediction engine with machine learning continues to look to see which combination of algorithms and data
streams have the most powerful prediction of different predictive hierarchies".
The ML provides an infinite loop for prediction, which carries an ever-improving effect. This kind of skill can
reshape warehouse management as we know it today.
6.3.4 Autonomous Vehicles for Logistics and Shipping
Intelligence in Logistics and Shipping has become a form of internal focus in asset management in recent years.
Faster and more efficient deployments reduce lead times and travel costs, increase environmental resources,
reduce labor costs, and - most importantly - widen the gap between competitors.
If private cars were developed for power - as some business analysts and tech gurus have thought - the impact
on asset efficiency could be stellar.
6.3.5 Natural Language Processing (NLP) for Data Cleansing and Building Data Robustness
NLP is an AI and Machine Learning tool, with an amazing ability to interpret large amounts of foreign language
data systematically.
The NLP, applied with appropriate work done, can create data sets with suppliers, and understand missing
data, due to a language problem. In the view of CSR or Sustainability & Governance, NLP technology could
reverse an audit and past actions that were not possible due to language barriers between consumer bodies.
6.3.6 ML and Predictive Analytics for Supplier Selection and Supplier Relationship Management (SRM)
Selection of suppliers and finding suitable suppliers is a growing concern for strengthening purchasing
sustainability, CSR, and asset sales ethics. Supplier-related risks have become a football and a series of visible
products around the world. One-dimensional smoothness in the performance of suppliers ' bodies, as well as
negative PR, is directly related to your company.
However, what if you had a very good provider of supplier selection and risk management, during all single-
provider communications?
Data sets, generated by SRM actions, such as supplier audits, audits, and credit ratings provide an important
basis for the continuation of decisions about the provider.
With the help of Machine Learning and understandable algorithms, this (besides) data collection performance
can be activated.
The choice of suppliers can be speculative and sound more than ever; builds a successful platform from the first
partnership. All of this information will be easily accessible for human testing but is done using a machine-
[598]
e-ISSN: 2582-5208
readable machine; to provide multiple providers of advanced providers based on any parameters, where the
user wishes.
6.4 Applications of AI in Marketing
Now day's managers are focusing on all innovative ways to interact with their customers. But in the fast-
moving and evolving environment of digital marketing, it is getting difficult to get people and get results.
Artificial intelligence can look like more science fiction than fact to many, but AI is no longer at a distance
concept. By 2021, $57 billion will be spending on Artificial intelligence platforms by companies. This is a matter
of concern for every type or size of organization to think about how AI can be useful for them to overcome the
competition.
Here is the list of some real-world examples of Applications of AI which help the organization to improve
customer satisfaction by increasing their profit as well.
Table 6: AI applications in Marketing
Name of the Feature Benefits
application
Saffron Predictive customer service Increased prediction efficiency by 50% to 88%
Albert Marketing automation Increased in the dealership by 29 times within
three months
Persado 1:1 dynamic e-mails Citi’s e-mails open rate increased by 70% and
click to open rate by 114%
Conversica Lead scoring- Automated sales CenturyLink's ROI came up to $20 for every
assistant software dollar invested
Layer 6 AI Predictive analytics Netflix AI recommendations platform helps
them save $1bn making a significant ROI
Wordsmith Content generation Yahoo has added over 100 years equivalent of
incremental audience engagement by working
with Wordsmith
IBM Watson Content curation Watson helped Wimbledon maintain its digital
viewership and increase video views by 25%
even after having international sports activities
going on in
Parallel
Sentiment Social semantics Euler Hermes' inbound leads increased by
38.1% with a projected conversion of 51.98%
Artificial intelligence in marketing used for:-
1. Sales forecasting
2. AI-powered customer insights
3. Content generation
4. AI-enhanced PPC advertising
5. Dynamic pricing
6. Automated image recognition
7. Ensure data quality and privacy
6.4.1 AI in Sales forecasting
Every business has an aim to know what they have to do next and how they can do that to meet their customer
needs and increase their sales. Artificial intelligence makes it easier for marketing managers to understand the
customer's wants based on the data which is collected by them and their past purchases. Marketers can predict

[599]
e-ISSN: 2582-5208
what their customers want from them and what they going to buy next by the means of AI systems. It will also
help to increase the organization's sales revenue because marketers already get to know that which product
they have to promote more.
Example- Qurious software is an AI-based software that is generally used by marketers.
6.4.2 Get AI-Powered customer insights
Data collection is the main reason for using AI by businesses. To make customer-centric decisions marketers
use AI to collect the insights of customers and to understand them. By analyzing a large amount of online
available data from social media platforms, blogs, or other platforms AI opens the source for the external
market. After collecting the customer insights data the segmentation of customers can be done easily and
effectively. Marketers now have a clear vision on which customer segment has to be targeted.
6.4.3 Content generation
Now day's content can be created by AI-powered tools that will be more effective and efficient. However the
main center of content creation is in the way of human creativity, but with the use of AI-powered tools, you can
increase the efficiency of content creating a team by doing some automated tasks like email content, social
media content curation.
Examples – a tool named Rocco can be used to create social media content to increase followers.
6.4.4 AI-enhanced Pay-per-click (PPC) advertising
Artificial intelligence can also be used in PPC advertisements. Some of the top AI applications in PPC are:-
 Smart bidding
 Micro-moment targeting
 Responsive ads
 Performance analysis
 Dynamic search ads
 Price optimization
 Account management
With the help of AI, marketers can keep track of more advertising platforms that cannot be reached via in-house
teams and by competitors.
6.4.5 Dynamic Pricing
This AI technology is also called personalized pricing. This is a pricing strategy used where the product's price
is set based on demand and/or supply. Customers can gain a profit from personalized pricing when there is a
low demand for a product.
For example – the prices of ride-sharing apps increase as demand rises and when hotel rooms go unsold.
6.4.6 Automated Image recognition:-
Artificial intelligence is used by big brands like Google, Facebook, Pinterest, and Amazon to recognize people
and objects from images and videos and it has an accuracy of 99.75%.
From point of view of marketing, image recognition has greater coordination with the online content and store
visit. Facial recognition software is used by many stores to trace visits of customers in stores and provide them
better service by linking their profiles. With the help of this, the interaction between customers and the store
will increase and it keeps their excitement level high when they are store.
6.4.7 Ensure Data Quality and Privacy
When AI is successfully applied, it contributes to improving data quality and maintaining privacy in an
organization. Typically, marketers work with a huge amount of data which can be a difficult task to handle.
Hence, to make the AI solution work we must have quality data all in one place.
It is similar to using data management software in a data warehouse to abolish data quality complications that
might occur due to improper management of data.
AI gives you the best way to collect data while blocking cyberattacks and protecting valuable data.

[600]
e-ISSN: 2582-5208
VII. ADVANTAGE AND DISADVANTAGE OF ARTIFICIAL INTELLIGENCE
IN THE FOOD INDUSTRY
Nowadays supply chain management is getting better because of artificial intelligence as it provides
transparency. AI-based technology gives a better understanding of situations which helps to get good revenue.
AI reduces time to deal with market uncertainties in business. AI will reduce labor costs, improve quality, and
increase the speed of the process.
7.1 Advantages of AI
Reduction of human error: - Artificial intelligence makes decisions based on the previous information or a set
of algorithms which helps in the reduction of the errors and the chance of accuracy is increased.
Available 24x7:- with the help of AI machines we can be available 24x7 without taking any break because an
average human can work 6-8 hours a day and then it needs to take a rest for refreshing themselves.
Digital assistance:- with the help of chatbots and Ai based billing machine in the restaurant can save the time
of users and also provide personalized services.
Faster decision:- artificial intelligence also help to make decisions because the information of our preferences
is already fed into these systems.
New inventions:- AI is helping to create new almost in every sector and in the food industry it also helps in
developing a new product.
7.2 Disadvantages of AI
High cost of creation:- The formation of these types of systems requires a huge cost, as in AI new technology
or hardware and software is introduced every day so to keep it updated and to meet the latest requirements
machines need to be repaired and maintain which takes a lot of costs.
Making humans Lazy: - with the help of AI most of the works are getting automated and humans get addicted
to these inventions that's make them lazy.
Unemployment: - As most of the time are preferring robots for doing repetitive jobs and want to make less
human intervention because AI robots can do the same work with more accuracy than a human.
Lacking out-of-box thinking: - Machines only perform those works which are programmed in them and for
they are designed to do, no more thinking like we humans do.
VIII. FUTURE PROSPECTS OF ARTIFICIAL INTELLIGENCE IN THE FOOD INDUSTRY
The food industry is undergoing a quick evolution. As per the McKinsey report, the food industry will reach
$100 billion globally in the upcoming five years all thanks to artificial intelligence and machine learning
innovation. Technology companies taking advantage of the opportunity to reformulate how food is being
cooked and can be planned in the diet.
IBM’s chef Whatson with Bon Appetite used artificial intelligence and developed an application that can read
the chemical compositions of hundred ingredients and analyze them to create thousands of recipes all you have
to do just feed the ingredients data in-app.
People who are suffering from obesity and diabetes are depending on dieticians to decide what they should
need to consume or what not to eat. The diet charts are made from physiological reports. But here AI and ML
play a major role nowadays machine learning and data analytics will be used to create plans for a specific
patient.
If we discuss the food industry in terms of the agricultural industry then we can find that there are more
chances of AI to grow. Mechanical engineers of Britain Institute state that 550 billion liters of water are
annually wasted in-process the production of crops. In the future artificial intelligence can reduce this number
and solve this problem somehow. If AI is successfully overcoming this problem then the production of food may
be increased by 60% or maybe more. There will be a lot of possible solutions for wastewater because Machine
learning and AI are still growing.

[601]
e-ISSN: 2582-5208
"AI technology can now be applied to every stage of the food chain and can be the key changing these
processes, from initial production and farming through to the final product served," (Nicola Sewell, marketing
manager at AI-enabled food waste tracker winnow).
AI can be used to give the direction about climate change to grow plants at the places where they can bloom.
A huge amount of data are collected via sensors, also from marketing and warehousing and this data is utilized
and evaluated by artificial intelligence. With the use of suitable algorithms machines are in a position to take
decisions independently after identifying data patterns.
In the future with the use of AI and robotics, there will be an automation workplace that consists of the more
prevalent use of technology combined with elevated end vision systems which include the usage of
hyperspectral cameras with AI to expand the spectral abilities up and far away from the human ability.
IX. CONCLUSION
Artificial intelligence is making the food industry more efficient and better and promises to yield many more
new changes shortly. AI's role in the food industry is becoming increasingly important due to its ability to
reduce waste, predict markets for products, enable round-the-clock efficient and effective monitoring, augment
sanitation, manage cost and increase revenue. Through the execution of AI and ML in the food industry are
already taken to the new level by facilitating hardly any human errors and reduce waste in most of the
products, by lowering the cost of storage/delivery and transportation; and making voice searching, quick
response and personalize orders according to customer. Researchers are ongoing to proper solutions to the
challenges of artificial intelligence and increase its adoption.
X. REFERENCES
[1] Olena Kovalenko Project Manager @ SPD-Group, leading AI/ML team and writing articles for our
blog.January 29, 2020[https://spd.group/machine-learning/machine-learning-and-ai-in-food
industry/#Using_AI_in_Food_Industry_Machine_Learning_applications_in_Food_Manufacturing]
[2] Ed burn(Machine Learning),
[3] [https://searchenterpriseai.techtarget.com/definition/machine-learning-ML]
[4] By Jake Frankenfield, Reviewed By Gordon Scott Updated Mar 8, 2021
[5] [https://www.investopedia.com/terms/a/artificial-intelligence]
[6] By Daniel Faggella, February 26, 2020,[https://emerj.com/ai-glossary-terms/what-is-machine-
learning/]
[7] Early History of Machine Learning
[8] [https://www.doc.ic.ac.uk/~jce317/historymachinelearning.html#:~:text=Early%20History%20of%2
0Machine%20Learning,The%20first%20case&text=In%201950%2C%20Alan%20Turing%20created
%20the%20world%2Dfamous%20Turing%20Test.&text=1952%20saw%20the%20first%20compute
r,network%20in%201958%2C%20called%20Perceptron]
[9] A brief history of machine learning
[https://www.dataversity.net/a-brief-history-of-machine-learning/#]
[10] Sunil Kumar Srivastava [https://orcid.org/0000-0002-7752-1328]
[11] Government of India, Ministry of Electronics and Information Technology, New Delhi, India
[12] By Gouravmoy Bannerjee, Uditendu Sarkar, Swarup Das, Indrajit Ghosh, [Artificial Intelligence in
Agriculture: A Literature Survey] www.ijsrcsams.com
[13] By Ngozi Clara Eli-Chukwu, [Applications of Artificial Intelligence in Agriculture:
[14] A Review] www.etasr.com
[15] Chidinma-Mary-Agbai, [Application of artificial intelligence (AI) in food industry]
https://doi.org/10.30574/gscbps.2020.13.1.0320
[16] Vijay Kakani , Van Huan Nguyen , Basivi Praveen Kumar , Hakil Kim ,Visweswara Rao Pasupuleti[A
critical review on computer vision and artificial intelligence in food industry]
www.journals.elsevier.com/journal-of-agriculture-and-food-research/
[17] Pixelplex, 20, Nov 2020, [From farm to table: The future of Artificial Intelligence]

[602]
e-ISSN: 2582-5208
[18] https://pixelplex.io/blog/ai-in-farming-and-agriculture/
[19] Thomas Hayes | Jan 17, 2020, [Artificial intelligence for new product development]
https://www.vitafoodsinsights.com/product-development/artificial-intelligence-new-product-
development
[20] Aidan Connolly, 10 august 2020, [Artificial intelligence can save the Food Industry.]
https://www.linkedin.com/pulse/can-artificial-intelligence-save-food-industry-aidan-connolly
[21] Sagar Sharma, 15 July 2020, [How Artificial Intelligence is Revolutionizing Food Processing Business?]
https://towardsdatascience.com/how-artificial-intelligence-is-revolutionizing-food-processing-
business-d2a6440c0360
[22] Hokey Min., 24 march 2009, Artificial intelligence in supply chain management: theory and
applications [https://doi.org/10.1080/13675560902736537]
[23] Terry Brown, 28 April 2020, Impact of AI in Supply Chain Management
[24] [https://itchronicles.com/artificial-intelligence/ai-in-supply-chain-management/]
[25] Lauren Moreno date: august 28, 2019, 10 impressive examples of ai in marketing,[
https://blog.socialmediastrategiessummit.com/10-examples-of-ai-in-marketing/]
[26] Jade, June 01, 2021, AI Marketing: What, Why and How to use Artificial Intelligence in Marketing
[https://www.mageplaza.com/blog/ai-marketing-what-why-how.html]
[27] Sunil Kumar, 25nov, 2019 [https://towardsdatascience.com/advantages-and-disadvantages-of-
artificial-intelligence-182a5ef6588c]
[28] HCL technologies [https://www.hcltech.com/technology-qa/what-are-the-advantages-of-artificial-
intelligence]
[29] By Suchayan Mandal Express News Service, 13th September 2020 [www.newindianexpress.com]
[30] By Lucy Britner, 22 April 2020 [https://www.just-food.com/analysis/machine-dreams-the-future-of-
artificial-intelligence-in-food-and-beverage-operations_id143542.aspx]
[31] By Friederike Arndt, February 4, 2020 [https://blog.drinktec.com/cross-industry/artificial-
intelligence-is-making-its-presence-known-in-the-beverage-industry/]
[32] Rick Graff & Greg Powers, Senior Product Manager, Automations & Vice President, Solutions
Architecture[https://www.gray.com/insights/ai-and-robotics-into-the-future-of-food-beverage-
engineering/]

[603]

Fin Irjmets1631603841

Uploaded by

Copyright:

Available Formats

Fin Irjmets1631603841

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fin Irjmets1631603841

Uploaded by

Copyright:

Available Formats

e-ISSN: 2582-5208

International Research Journal of Modernization in Engineering Technology and Science

MACHINE LEARNING AND ARTIFICIAL INTELLIGENCE IN FOOD INDUSTRY

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

2.2 Types of Artificial Intelligence:-

Fluid AI www.fluid.ai Provides customer information on the products in an

Flutura www.flutura.com Monitors health of machines to advise on

Heckyl www.heckyl.com Analyses stock related information to advises on

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

ShopR360 www.shopr630.com Video analytics solution which can distinguish staff

SigTuple www.sigtuple.com Affordable diagnosis solution using a microscope, cell

Social Cops www.socialcops.com Interpretation of data

VPhrase www.vphrase.com Converts structured data such as graphs, etc. into

Freenome www.zephyrhealth.com Analyses genetic material to find disease signature at

JISTEM USP, Brazil Vol. 15, 2018, www.jistem.fea.usp.br

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

ANN Predicts crop yield Only captures weather as a

Above 90% success rate in A little number of symptoms

FUZZY Predict cotton yield and improve It is relatively slow

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

FuzzyXpest Provides pest information for Internet-dependent

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science

You might also like