AI and Robotics in pharmaceutical industries.

1. A R T I F I C I A L I N T E L L I G E N C E ,
R O B O T I C S A N D
C O M P U T A T I O N A L F L U I D
D Y N A M I C S
- Sneha Gaurkar
M.Pharm II sem
Department of Pharmaceutics
Smt. Kishoritai Bhoyar College Of
Pharmacy, kamptee

2. Contents :
 Artificial Intelligence
Introduction
Applications
Advantages
Risks
Challenges
Future aspects
 Robotics
Introduction
Types
Advantages
Challenges
 Computational fluid
dynamics
 Introduction
 Applications
 Advantages
 Disadvantges
 Current challenges and future
aspects
2

3. A R T I F I C I A L
I N T E L L I G E N C E
3

4. ARTIFICIAL INTELLIGENCE
Introduction to Artificial Intelligence :
 According to the father of Artificial Intelligence, John McCarthy, it is, “The
science and engineering of making intelligent machines, especially
intelligent computer programs”.
 Artificial Intelligence is a way of making a computer, a computer-controlled
robot, or a software think intelligently, in the similar manner the intelligent
humans think.
 AI is accomplished by studying how human brain thinks, and how humans
learn, decide, and work while trying to solve a problem, and then using the
outcomes of this study as a basis of developing intelligent software and
systems.
4

5. Also, intelligence distinguish us from everything in the
world. As it has the ability to understand, apply
knowledge.
 Also, improve skills that played a significant role in our
evolution. We can define AI as the area of computer
science.
 Further, they deal with the ways in which computers can
be made. As they made to perform cognitive functions
ascribed to humans.
5
Cont…

6. Definition Of Artificial Intelligence :
 AI may be defined in simple words that “It is the study of ideas which
enable computers to do the things that make people seem intelligent”.
 Artificial intelligence refers to the ability of a computer or a computer
enabled robotic system to process information and produce outcomes in a
manner similar to the thought process of humans in learning , decision
making and solving problems.
 By extension the goal of AI system is to develop system capable of tacking
complex problems in ways similar to human logic and reasoning.
 AI is the branch of science which deals with the helping machines to find
the solutions of the complex problems in a more human like fashion.
6

7. Artificial intelligence
Machine learning
Deep
learning
7

8. Simple Neural Network
Deep Neural Network
8
For example…

9. Properties Of AI Software :
 A software generates new molecular structures by
combining properties of existing drugs.
 Automatic chemical design helps drug discovery for a
faster & accurate manner.
 A treatment method from the ground up using a deep
learning neural network.
 A system to use historical , biological & chemical data to
imagine novel molecules with the potential to fight major
diseases.
9

10. Applications of artificial intelligence (AI) in different subfields
of the pharmaceutical industry
10

11. AI Techniques In Use For Drug Discovery
Deep learning technique known as a generative
adversarial network (GAN) by Baltimore based company,
Insilico Medicine.
 GPU (graphics processing unit)-accelerated deep learning
to target cancer and age-related illnesses by above
organization.
 Benevolent Bio’s deep learning software, powered by the
NVIDIA DGX-1 AI supercomputer (it ingests & analyzes
the information to find connections and propose drug
candidates).
11

12. AI In Drug Discovery
12

13. Examples of AI tools used in drug discovery
Tools Details Website URL
DeepChem MLP model that uses
a python-based AI
system to find a
suitable candidate in
drug discovery
https://github.com/d
eepchem/deepchem
DeepTox Software that predicts
the toxicity of total of
12 000 drugs
www.bioinf.jku.at/res
earch/DeepTox
DeepNeuralNetQSAR Python-based system
driven by
computational tools
that aid detection of
the molecular activity
of compounds
https://github.com/M
erck/DeepNeuralNet-
QSAR
13

14. Cont…
14
Tools Details Website URL
ORGANIC A molecular
generation tool that
helps to create
molecules with
desired properties
https://github.com/as
puru-guzik-
group/ORGANIC
PotentialNet Uses NNs to predict
binding affinity of
ligands
https://pubs.acs.org
/doi/full/10.1021/ac
scentsci.8b00507
Neural graph fingerprint Helps to predict
properties of novel
molecules
https://github.com/
HIPS/neural-
fingerprint
AlphaFold Predicts 3D
structures of
proteins
https://deepmind.co
m/blog/alphafold

15. AI in clinical trial design
 Preclinical discovery of molecules as well as predicting lead compounds
before the start of clinical trials by using other aspects of AI, such as
predictive ML and other reasoning techniques, help in the early prediction
of lead molecules that would pass clinical trials with consideration of the
selected patient population.
 Drop out of patients from clinical trials accounts for the failure of 30% of
the clinical trials, creating additional recruiting requirements for the
completion of the trial, leading to a wastage of time and money. This can
be avoided by close monitoring of the patients and helping them follow
the desired protocol of the clinical trial
 Mobile software was developed by AiCure that monitored regular
medication intake by patients with schizophrenia in a Phase II trial, which
increased the adherence rate of patients by 25%, ensuring successful
completion of the clinical trial
15

16. AI in pharmaceutical product management
AI in market positioning
Market positioning is the process of creating an identity of
the product in the market to attract consumers to buy them,
making it an essential element in almost all business strategies
for companies to establish their own unique identity
With the help of technology and e-commerce as a platform,
it has become easier for companies to get a natural recognition
of their brand in the public domain.
Other techniques, such as statistical analysis methods,
particle swarm optimization algorithms (proposed by Eberhart
and Kennedy in 1995) in combination with NNs, provided a
better idea about markets. They can help decide the marketing
strategy for the product based on accurate consumer-demand
prediction
16

17. Cont…
AI in market prediction and analysis
 The advances in digital technologies, referred to as the ‘Fourth industrial
revolution’, is helping innovative digitalized marketing via a multicriteria
decision-making approach, which collects and analyzes statistical and
mathematical data and implements human inferences to make AI-based
decision-making models explore new marketing methodology
 AI-based software engages consumers and creates awareness among
physicians by displaying advertisements directing them to the product site by
just a click.
 In addition, these methods use natural language-processing tools to analyze
keywords entered by customers and relate them to the probability of
purchasing the product.
 Pharmaceutical companies are also introducing their online applications such
as 1 mg, Medline, Netmeds, and Ask Apollo, to fulfill the unmet needs of the
patients.
17

18. Advantages of AI
a. Error Reduction
 We use artificial intelligence in most of the cases. As this helps us in
reducing the risk.
 Also, increases the chance of reaching accuracy with the greater degree of
precision.
b. Difficult Exploration
 In mining, we use artificial intelligence and science of robotics. Also, other
fuel exploration processes.
 Moreover, we use complex machines for exploring the ocean. Hence,
overcoming the ocean limitation.
18

19. Cont…
c. Daily Application
 As we know that computed methods and learning have become common
place in daily life.
 Financial institutions and banking institutions are widely using AI. That is
to organize and manage data.
 Also, AI is used in the detection of fraud users in a smart card based
system.
d. Digital Assistants
 “Avatars” are used by highly advanced organizations. That are digital
assistants.
 Also, they can interact with the users. Hence. They are saving human
needs of resources.
 As we can say that the emotions are associated with mood. That they can
cloud judgment and affect human efficiency. Moreover, completely ruled
out for machine intelligence.
19

20. Cont…
e. No breaks
 Machines do not require frequent breaks and refreshments for humans. As
machines are programmed for long hours.
 Also, they can continuously perform without getting bored.
f. Increase Work Efficiency
 For a particular repetitive task, AI-powered machines are great with amazing
efficiency.
 Best is they remove human errors from their tasks to achieve accurate
results. Reduce cost of training and operation.
 Deep Learning and neural networks algorithms used in AI to learn new things
like humans do.
 Also, this way they eliminate the need to write new code every time.
20

21. Risks of Artificial Intelligence
a. High Cost
 Its creation requires huge costs as they are very complex machines. Also,
repair and maintenance require huge costs.
b. No Replicating Humans
 As intelligence is believed to be a gift of nature. An ethical argument
continues, whether human intelligence is to be replicated or not.
c. Lesser Jobs
 As we are aware that machines do routine and repeatable tasks much
better than humans.
 Moreover, machines are used of instead of humans. As to increase their
profitability in businesses.
21

22. Cont…
d. Lack of Personal Connections
 We can’t rely too much on these machines for educational oversights.
That hurt learners more than help.
e. Addiction
 As we rely on machines to make everyday tasks more efficient we use
machines.
f. Efficient Decision Making
 As we know computers are getting smarter every day.
 Also, they are demonstrating not only an ability to learn but to teach
other computers.
22

23. Challenges to the adoption of Artificial
Intelligence (AI) in Pharma
23
The
unfamiliarity
of the
technology
Overfitting
or
underfitting
Low
accuracy of
the training
data
Breaking down
data silos and
streamlining
electronic records
Lack of
proper IT
infrastructure
Hesitant
to change
Data quality,
governance,
security, and
interoperability

24. Cont…
A. The unfamiliarity of the technology: For many pharma companies, AI still
seems like a “black box” owing to its newness and esoteric nature.
B. Lack of proper IT infrastructure: Most IT applications and infrastructure
currently in use are not developed or designed with AI in mind. Even worse,
pharma companies have to spend lots of money to upgrade their IT system.
C. Breaking down data silos and streamlining electronic records: Much of the
data is in a free text format, and the data management is messy and
unorganized across the heterogeneous databases. This means that pharma
companies have to go above and beyond to collate and put this data into a
standard form that can be analyzed.
D. Low accuracy of the training data: Even though algorithms have a higher
threshold for minimizing errors, there are still some categorical errors from
training sets.
24

25. E. Overfitting or underfitting: With algorithm prediction, there is a concern
with overfitting or underfitting.
Overfitting means when a model consists of lower quality
information/technique but generates higher quality performance.
Underfitting models fail to recognize the underlying trend in the datasets
and generalize the new data. Both result in inaccurate results.
F. Data quality, governance, security, and interoperability: Issues around
data will always be at the heart of successfully promoting AI solutions.
Healthcare is the least digitized sector, which needs to take a systematic
approach to develop common data standards and processes to maximize
the value of existing data. Healthcare providers and AI companies need to
put in place robust data governance, ensure interoperability and
standards for data formats, enhance data security and bring clarity to
consent over data sharing.
25
Cont…

26. G. The need for transparent algorithms to meet drug development
regulations: Transparency in health care is quite a task given the
complexity of the processes involving artificial intelligence.
H. Hesitant to change: Pharma companies are known to be traditional and
resistant to change.
26
Cont…

27. Future prospects
 AI technologies are now used at every stage in pharmaceutical R&D, as is
evidenced by the numerous high-value deals between ‘big pharma’ and
providers of AI technologies.
 As a recent example, Eversheds Sutherland advised global
biopharmaceutical company AstraZeneca on its long-term collaboration
with BenevolentAI to use AI and machine learning for the discovery and
development of potential new treatments for chronic kidney disease (CKD)
and idiopathic pulmonary fibrosis (IPF).
 Further to this, even if the prospect of machines inventing medicines is
some way off, the use of AI to improve efficiency and speed up drug
discovery is a realistic prospect today.
27

28. Cont…
 AI technologies may also be utilised in areas of manufacturing which are
amenable to predictive modelling and predictive support, such as
process optimisation.
 The use of such AI tools within smart factories and smart manufacturing
is enabling less down-time in the manufacturing process and driving
stronger results. And AI tools are also beginning to be deployed on a
more wide-spread basis within the ‘back office’ such as within
accounting technology. As AI tools and computing power become more
accessible and affordable, as with any technology, it will become very
much part of the process within pharma and manufacturing – the
future will be ‘AI-enabled’.
28

29. 29
R O B O T I C S

30. Introduction
 The International Organization for Standardization gives a
definition of robot in ISO 8373: "An automatically controlled,
reprogrammable, multipurpose, manipulator programmable in
three or more axes, which may be either fixed in place or
mobile for use in industrial automation applications.“
 Robotics is a domain in artificial intelligence that deals with
the study of creating intelligent and efficient robots.
 Robots are the artificial agents acting in real world
environment.
30

31. Types of Robots
• There are three types of industrial robots most commonly used in
pharmaceutical manufacturing.
 In their simplest form, Cartesian robots consist of two linear slides placed
at 90-degree angles to each other, with a motorized unit that moves
horizontally along the slides in the axis (z), which moves up and down in
the vertical plane.
 The quill holds the robot’s end-effector, such as a gripper. A fourth axis
(t, or theta) allows the quill to rotate in the horizontal plane.
31
1. Cartesian

32. Cont…
 The chief advantage of Cartesian robots is their low cost, although their
restricted range of motion limits their usage.
 They are often incorporated into automation subsystems or machines
dedicated to a single purpose, such as assay testing
32

33. 2. SCARA Robots
SCARA stands for “selective compliance articulated robot arm.”
 This refers to the fact that a SCARA’s arm segments, or links, are
“compliant,” that is, they can move freely, but only in a single geometrical
plane.
 Most SCARAs have four axes. Even though three- and five-axis SCARAs are
also found, the terms “SCARA” and “four axis robot” are often used
interchangeably to refer to a four-axis SCARA.
 The first two links of a SCARA swivel left and right around the first two
axes in the horizontal plane. The third link is the quill, which moves up and
down in the vertical plane along the third axis.
 The quill also rotates horizontally in the fourth axis, but cannot tilt at a
vertical angle.
 Some SCARAs have a metal shaft that is actually a hollow air- balance
cylinder.
33

34.  The function of this cylinder is to counterbalance the weight of the end-
effector and payload and thus reduce settling time—the time in which
robot has to wait after it moves to a given point before it can carry out its
next movement.
 Faster settling times result in faster cycle times.
 The unique design of SCARAs gives them a high degree of rigidity, which in
turn allows them to move very fast and with precise repeatability.
 SCARAs excel at high-speed pick-and- place than other material-handling
tasks.
34

35. 3. Articulated Robots
 Articulated robots not only have more joints than SCARAs, they have both
horizontal and vertical joints, giving them increased freedom of
movement. Whereas a Cartesian robot has a cube- shaped work envelope
and a SCARA has a cylindrically shaped one, the work envelope of an
articulated robot is spherical.
 With their greater flexibility of movement, articulated robots can perform
almost any task that can be performed by a human arm and hand.
 The most common articulated robots have six axes.
 The first link rotates in the horizontal plane like a SCARA, while the second
two links rotate in the vertical plane.
35

36.  In addition, six-axis articulated robots have a vertically rotating “forearm”
and vertically rotating “wrist” joint, which let them perform many of the
same types of movements as a human forearm and wrist.
 The forearm and wrist joint of six-axis articulated robots allow them to pick
up an object no matter how it is oriented off the horizontal plane, then
place it at any angle of approach that might be required.
 They also allow the robot to perform many other tasks that would
otherwise call for the dexterity of a human operator.
36
Comparison between robot and human movement. (Image: Kawasaki)

37. NANO-ROBOTS
 Nano-robots are so tiny machines that they can traverse the human body
very easily.
 When a nano-robot enters into the body of a patient would seek for
infected cells and would repair them without causing any damage to the
healthy cells.
 The nano-robot will remain outside the cell while the nano- manipulators
will penetrate into targeted or damaged cell thus avoiding any possibility of
causing damage to the intracellular skeleton.
37

38.  Thus these nano-robots when enter into human bloodstream
provide cell surgery and extreme life prolongation.
 Each nano-robot by itself will have limited capabilities, but
the coordinated effort of a multitude will produce the desired
system level results.
 Coordination is needed across the board for communication,
sensing, and acting and poses a major research challenge.
38

39. Determining the Type of Robot Needed
 The first step in automating a process with a robot is to establish the
process parameters, including
(1) The required type and size of end-effector, or end-of-arm tooling
(2) Cycle time
(3) Repeatability
(4) Reach
(5) Payload capacity
 Taken together, these will usually determine whether a Cartesian,
SCARA or articulated robot is necessary.
 Another essential consideration is the environment in which the robot
will be operating.
39

40. 40
Accuracy Tirelessness
Return on investment
Reliability
Affordability
Speed Advantages of Pharmaceutical Robots
And Production
Quality
Flexibility
Safety
Reduced chances of contamination
Work continuously in any environment
And Savings
Advantages

41. Challenges
1. Expense: The initial investment of robots is significant, especially when
business owners are limiting their purchases to new robotic equipment.
The cost of automation should be calculated in light of a business'
greater financial budget. Regular maintenance needs can have a
financial toll as well.
2. Dangers and fears: Although current robots are not believed to have
developed to the stage where they pose any threat or danger to society,
fears and concerns about robots have been repeatedly expressed in a
wide range of books and films. The principal theme is the robots'
intelligence and ability to act could exceed that of humans, that they
could develop a conscience and a motivation to take over or destroy the
human race.
41

42. 3. Expertise: Employees will require training in programming and
interacting with the new robotic equipment. This normally takes time
and financial output.
4. Return on investment (ROI): Incorporating industrial robots does not
guarantee results. Without planning, companies can have difficulty
achieving their goals.
5. Safety: Robots may protect workers from some hazards, but in the
meantime, their very presence can create other safety problems. These
new dangers must be taken into consideration.
42

43. 43
C O M P U T A T I O N A L
F L U I D D Y N A M I C S

44. Introduction
 CFD is a branch of fluid mechanics that applies numerical methods for
solution of physical problems involving fluid flows.
 CFD is now recognized to be an important part of the computer-based
engineering design tools and is used extensively today in many industries.
 In general, computational methods can be used to analyze various design
and process parameters, while, at the same time, reducing the involved
experimental work and associated costs.
 Computational modeling has been found useful in the study of various
pharmaceutical unit operations, for example, mixing and dissolution
44

45. Definition
 Computational fluid dynamics (CFD) is a branch of physics
that deals with the study of the mechanics of fluid: liquid,
plasmas and gasses and forces acting on them.
 CFG is based on Navier-Stroke equations that describe how
pressure, velocity, density and temperature of a moving fluid
are related.
45

46. APPLICATION OF CFD IN PHARMACEUTICS
•The application of CFD to a few key unit operations and processes in
the pharmaceutical industry
46
CFD for mixing
CFD for solids handling
CFD for separation
CFD for dryers
CFD for packaging
CFD for energy generation and energy-transfer
devices

47.  CFD methods can be applied to examine the performance of static mixers
and to predict the degree of mixing achieved, thus indicating whether more
mixing elements are required shows surface mesh and blade orientation for
a Kenics mixer depicts the mass fraction concentration of the two species
being mixed.
 The degree of mixing is shown as the color proceeds from distinct inlet
streams (red and blue) to the fully mixed outlet stream (green).
 A CFD solution can be used to derive the pressure drop, hence the power
required.
47
CFD for mixing
Example: Mixing of Powders

48. CFD techniques can be applied to analyze such flows and minimize or
eliminate the risk of erosion.
 CFD also can be applied to analyze the unsteady and chaotic flow behavior in
fluidized beds.
 Simulation of such a flow field requires unsteady flow calculations and small
time increments.
 As a result, performing calculations can take an extensive amount of time.
 Simulations of gas–solid flows in complex three-dimensional reactors can
take months of computational time and are not practically feasible
48
CFD for solids handling

49. CFD techniques are used for analyzing
separation devices such as cyclones and
scrubbers.
 The following example incorporates CFD
methods to optimize and predict performance
of an existing cyclone design.
 CFD solutions depict particle paths for various
particle sizes.
 In this example, CFD techniques were used to
perform what-if analysis for optimization of the
design.
The performance computed with CFD closely
matched that observed in physical testing
wherein 90% of 10ưm particles were removed,
but only 10% of 1ưm particles were separated
from the air stream.
49
CFD for separation

50.  We used CFD to analyze the performance of an
industrial spray dryer before making major structural
changes to the dryer.
 This strategy minimizes the risk of lost profit during
changeover, especially if the improvement does not
materialize.
 CFD was applied to examine configuration changes,
thus minimizing risk and avoiding unnecessary
downtime during testing shows the velocity
distribution (skewed flow).
 This flow is a result of uneven pressure distribution in
the air dispersing head.
 CFD models were applied to determine optimum
equipment configuration and process settings.
50
CFD for dryers

51.  CFD can be applied to conduct virtual experiments before changes are made
to the filling lines or to the package geometry.
 This method allows a wide range of conditions to be tested and leads to an
optimized filling process, depicts the filling of a container.
 The figures shown are typical of solution results that are used to optimize
filling processes to increase throughput and reduce foaming.
(a)filling process, liquid surface location, strong splash;
(b) filling process, liquid surface location, no splash.
51
CFD for packaging
a) b)

52.  CFD techniques can be applied to analyze thermal and flow fields within such
devices.
 CFD modeling methods also can be applied to gain insight into flame
characteristics.
 Maintaining flame stability and burner efficiency is very critical to the proper
functioning of a process heater, power plant, or furnace. Flame length, shape,
and size can influence the process.
 If the flame is too long, then it can impinge on critical regions of the
apparatus and cause thermal damage.
 If the flame is too short, then it may wear out the burner tip.
 Replacing the burner or associated apparatus results in downtime and loss of
product revenue.
52
CFD for energy generation and energy-transfer devices

53. Advantages Of CFD
 A great time reduction and cost reduction in new designs.
 There is a possibility to analyze different problem whose experiments are
very difficult and dangerous.
 The CFD techniques offer the capacity of studying system under
conditions over its limits.
 The level of detail is practically unlimited.
 The product gets added value.
 The possibility to generate different graph permits to understand the
features of the result. This encourages buying a new product.
 Hi-Tech CFD is a computer aided engineering company which provides
total solutions to engineering problems in the field of Computational
Fluid Dynamics (CFD),Computational Electromagnetic, Computational
Structural Mechanics, Dynamics and Controls.
53

54. Disadvantages Of CFD
 Accuracy in the result is doubted i.e. in certain situations we will not
obtain successful result.
 It is necessary to simplify mathematically the phenomenon to facilitate
calculus. If the simplification has been good the result will be more
accurate.
 There are several incomplete models to describe the turbulence,
multiphase phenomenon, and other difficult problems.
 Untrained user of CFD has the tendency to believe that the output of the
pc is always true
54

55. Softwares of CFD:
CFD Simulation :CFD simulation software helps to predict the impact
of fluid flows on a given product throughout design and manufacturing
as well as during end use.
Passage® software is a collection of finite element programs solving
complex flow, heat transfer and other related problems in product
design and manufacturing.
PASSAGE Discrete Element Method (DEM) Software is for predicting
the flow particles under a wide variety of forces. It can be used alone or
together with our PASSAGE/FLOW Software.
DEM and FLOW CFD Software
55

56. Current Challenges/ Future Aspect
 The integration of CFD methods can shorten product-process
development cycles, optimize existing processes, reduce energy
requirements, and lead to the efficient design of new products and
processes.
 Unit operations in the pharmaceutical industry handle large amounts of
fluid. As a result, small increments in efficiency, such as those created by
implementing CFD solutions, can lead to significant product cost savings.
 Key processes in the pharmaceutical industry can be improved with CFD
techniques.
 The aerospace and automobile industries already have integrated CFD
methods into their design process.
 The chemical process and the pharmaceutical industries now are
beginning to integrate this technology.
 The full potential for process improvements using CFD solutions is yet to
be realized 56

57. References:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7577280/
https://www.slideshare.net/SyedImran151/pharmaceutical-
robotics
SHEPARD JEFF ,Robot axes, drive safety, and power architectures
MARCH 8, 2021 https://www.microcontrollertips.com/robot-axes-
drive-safety-and-power-architectures/
https://www.slideshare.net/PRAJAKTASAWANT33/computer-aided-
drug-development-249384130
57

58. 58
References:
Challenges to the adoption of Artificial Intelligence (AI) in Pharma
By Editorial -April 23, 2021 https://roboticsbiz.com/challenges-to-
the-adoption-of-artificial-intelligence-ai-in-pharma/
https://www.slideshare.net/SyedImran151/pharmaceutical-
robotics
https://www.slideshare.net/PrasathP13/artificial-intelligence-
138582081
http://www.technalysis.com/pharmaceutical_industry.aspx
https://alfrescostaticfiles.s3.amazonaws.com/alfresco_images/ph
arma/2014/08/22/86fd0a7c-40ff-47ef-8e11-22ac54892744/article-
9676.pdf

59. THANK YOU !!!
Skb cop