The document discusses the use of design of experiments (DoE) in pharmaceutical development. It begins with an introduction to DoE, covering its history, terminology, why and how it is used. The document then discusses the types of DoE including factorial designs and fractional factorial designs. Examples of DoE applications in areas like oral drug delivery and inhalation drug delivery are provided. The advantages of DoE include maximizing process knowledge with minimal resources and establishing cause-and-effect relationships. The document concludes that DoE is a useful statistical tool that can promote quality in pharmaceutical development.

1. Seminar on
Design Of Experiments (DoE) In Pharmaceutical
Development
Presented by:
Imdad H. Mukeri
M. Pharm (Pharmaceutics)
Center of Pharmaceutical Sciences, IST,
JNTUH, Hyderabad-500085
11/15/2022 1
CPS, IST, JNTUH

2. TABLE OF CONTENS
 Introduction
 History of DoE
 Terminology & Evaluation of DoE
 DOE: Why to use it ?
 DOE: How to use it ?
 Steps and Guidelines for Planning and Conducting DoE
 Types Of DoE with example
 Advantages of DoE
 Uses of DoE
 Conclusion
 References
11/15/2022 2
CPS, IST, JNTUH

3. INTRODUCTION
 Design of Experiments (DOE) mathematical
methodology used for planning and conducting
experiments as well as analyzing and interpreting data
obtained from the experiments.
 Simply means to make as Perfect, Effective, or
Functional as possible.
 The design of experiments ensures
I. Formulation quality,
II. Saves time,
III. Labor
IV. And money
11/15/2022 3
CPS, IST, JNTUH

4. History Of DOE
 The agricultural origins 1918 – 1940s
R. A. Fisher & his co-workers
Profound impact on agricultural science
Factorial designs, ANOVA
 The first industrial era, 1951 – late 1970s
Box & Wilson, response surfaces
Applications in the chemical & process industries
 The second industrial era, late 1970s – 1990
Quality improvement initiatives in many companies
CQI and TQM were important ideas and became
management goals
Taguchi and robust parameter design, process robustness
11/15/2022 4
CPS, IST, JNTUH

5. 11/15/2022 5
CPS, IST, JNTUH

6. 11/15/2022 CPS, IST, JNTUH 6

7. Evolution of DOE
• Depends on one’s knowledge
and experience
• Depends on one’s luck
1. Trial and error
method
• Does not examine all
permutations and combinations
• Can not examine interactions
2. One factor at a
time
• Can predict the results of
experiments not yet performed
• Can predict the best conditions
to meet multiple goals
3. Design of
experiments
11/15/2022 7
CPS, IST, JNTUH

8. DOE: Why to use it ?
 Reduce time to design/develop new products &
processes
 Improve performance of existing processes
 Improve reliability and performance of products
 Achieve product & process robustness
 Perform evaluation of materials, design alternatives,
setting component & system tolerances
11/15/2022 8
CPS, IST, JNTUH

9. DOE: How to use it ?
VARIABLE / FACTOR
1. Independent Variables
A. Quantitative : Numeric values and continuous. e.g. Time,
Temperature, Amount of polymer, Plasticizer,
Superdisintegrants etc. such as 1%, 2%, 3% concentration
B. Qualitative : (also known as categorical variables) e.g.
Type of polymer, component or machine.
2. Dependent Variables :
• Characteristics of the finished drug product are Dependent
Variables Or Response Variables. e.g. Drug release profile,
Percent drug entrapment, Pellet size distribution, Moisture
uptake etc.
11/15/2022 9
CPS, IST, JNTUH

10. 11/15/2022 10
CPS, IST, JNTUH

11. Generally, factors that cause product variation can be
categorized in three main groups
External/environmental (such as temperature,
humidity and dust)
Internal (wear of a machine and aging of materials)
Unit to unit variation (variations in material,
processes and equipment)
11/15/2022 11
CPS, IST, JNTUH

12. General practical steps and guidelines for planning
and conducting DOE are listed below
State the objectives
Response variable definition
Determine factors and levels
Determine Experimental design type
Perform experiment
Data analysis
Practical conclusions and recommendations
11/15/2022 12
CPS, IST, JNTUH

13. Types Of DOE
1. Mixture designs (used for formulation characterization
and optimization)
2. Factorial (process) designs (each factor can be
adjusted independently of the others)
3. Full factorials designs (all possible combinations of
factor levels.)
4. Fractional factorial designs (Represent a part of the
relevant full design)
11/15/2022 13
CPS, IST, JNTUH

14. 1. Mixture designs
 Mixture designs are used for formulation characterization and
optimization, applied when the overall amount of a
composition is defined, e.g. for a tablet with fixed mass.
 Mixture designs component proportions are not independent
to each other.
 Mixture-process designs are used to investigate interactions
between formulation and process variables.
Applications in pharmaceutical technology include
 determination of diluent proportions in solid formulations,
 selection of appropriate solvent-cosolvent combinations in liquid
forms, etc
Responses = f(component proportions) ........ Eq. (1)
11/15/2022 14
CPS, IST, JNTUH

15. 2. Factorial (process) designs
 Parameters that can be adjusted independently of each
other, such as compaction force, temperature, spraying
rate, etc. In this case, the responses are functions of
factor levels as described in equation (2).
 Change two or more things at a time
 Process parameters are intentionally and simultaneously
varied according to a predermined matrix of factor levels
combinations. Their main difference from mixture
designs is that each factor can be adjusted independently
of the others.
Responses = f(factor levels) ...........Eq. (2)
11/15/2022 15
CPS, IST, JNTUH

16.  Factors are usually represented by capital letters (A, B,
C …), while their
 Lower levels : (-) or -1
 Intermediate level : (0).
 higher levels: (+) or +1, are respectively.
 This is obviously a coded representation of the levels,
which however corresponds to actual values of the
parameters, according to equation 3. Eq. (3)
Xcoded=(Xactual –Xmean) / [(Xhigh-Xlow)/2].....Eqn (3)
11/15/2022 16
CPS, IST, JNTUH

17. 11/15/2022 CPS, IST, JNTUH 17
Factorials level Description
One Factorial
Level
 One factorial experiments look at only one factor
having an impact on output at different factor levels. The
factor can be qualitative or quantitative.
 In single factor experiments, ANOVA models are used
to compare the mean response values at different levels
of the factor
Two level
factorials
 Full 2 level factorials can support linear models, thus
they are not capable for addressing more complex
phenomena, requiring higher order models.
Three level
factorial designs
 Three-level designs are useful for investigating
quadratic effects. The three-level design is written as a
3k factorial design

18. 3. Full factorials
 Full factorials include all possible combinations of
factor levels. The number of experiments required is
provided by equation 4.
 Change many things at a time.
 For example, a full two-level factorial for three factors
requires 23= 8 experiments.
Number of experiments=Levels (Factors) .......Eqn (4)
11/15/2022 18
CPS, IST, JNTUH

19. 4. Fractional factorial designs
 Fractional factorial designs represent a part of the
relevant full design, typically ½ or ¼ … of the full
factorial.
 They are typically used when the number of factors
exceeds 4-5, for screening purposes.
 Main limitation is related to confounding or alias of
main effects and interactions.
11/15/2022 19
CPS, IST, JNTUH

20. Response Surface Analysis Method (RSM)
 RSM explores the relationships between several explanatory
variables and one or more response variables.
 The method was introduced by G. E. P. Box and K. B. Wilson in
1951. The main idea of RSM is to use a sequence of designed
experiments to obtain an optimal response.
 Response surface analysis is an off-line optimization technique.
Usually, 2 factors are studied; but 3 or more can be studied.
11/15/2022 CPS, IST, JNTUH 20
 With response surface analysis,
we run a series of full factorial
experiments and map the response to
generate mathematical equations that
describe how factors affect the
response.

21. Central Composite Designs (CCDs)
 A Box-Wilson Central Composite Design, commonly called 'a
central composite design,' contains an imbedded factorial or
fractional factorial design with center points that is augmented with
a group of 'star points' that allow estimation of curvature.
 If the distance from the center of the design space to a factorial
point is ±1 unit for each factor, the distance from the center of the
design space to a star point is |α| > 1. The precise value of α depends
on certain properties desired for the design and on the number of
factors involved.
11/15/2022 CPS, IST, JNTUH 21
 Star points represent new extreme
values (low and high) for each factor in
the design
Fig: Central Composite Design for Two
Factors

22. Examples of DoE application in medicinal product
development and pharmaceutical processes.
AREA APPLICATION Applied DoE TYPE
Oral Drug
Delivery
Dispersible tablets
development
Several factorial
experiments at 2-3
factors, 2-3 levels
Immediate release tablet
platform
fractional factorial design
Fast dissolving pellets 25– 1 fractional factorial
design, 5 factors
Gastroretentive dosage form 3-level-3-factor, box-
behnken design
Inhalation
Drug
Delivery
Powder for inhalation
(formulation and process
development)
Half-fractional factorial
design with 5 factors at
two levels
Table- 1: Examples of DoE application
11/15/2022 22
CPS, IST, JNTUH

23. Examples of DoE application in medicinal product
development and pharmaceutical processes.
Injections
Formulation for
parenteral nutrition
(development)
Mixture design
Nanopharmaceutics
Solid Lipid
Nanoparticles for
Inhalation (process
development)
Two level full
factorial design
Table- 2: Examples of DoE application
11/15/2022 23
CPS, IST, JNTUH

24. DOE Software
1. Design Expert Software
 The analysis of the designs is carried out using Design
Expert Software (Statease, version 9.0.1, Minneapolis,
US).
11/15/2022 CPS, IST, JNTUH 24
Starting from the
analysis of the full
two level factorial
with center points,
graphical tools such
as half normal and
Pareto plots are
helpful in identifying
the most influential
effects

25. 2. Analysis of Variance (ANOVA)
 The formal statistical analysis is carried using ANOVA
and the relevant data for the two responses are shown in
Tables. 8
11/15/2022 CPS, IST, JNTUH 25

26. Uses of DOE
It is a multipurpose tool that can be used in various
situations for identification of important input factors
(input variable) and outputs (response variable).
1. Comparison: This is one factor among multiple
comparisons to select the best option that uses t‒test,
Z‒test, or F‒test.
2. Variable screening: These are usually two-level
factorial designs intended to select important factors
(variables) among many that affect performances of a
system, process, or product.
11/15/2022 26
CPS, IST, JNTUH

27. Uses of DoE
3. Transfer function identification: if important input
variables are identified, the relationship between the
input variables and output variable can be used for
further performance exploration of the system, process
or product via transfer function.
4. System Optimization: the transfer function can be used
for optimization by moving the experiment to optimum
setting of the variables. On this way performances of the
system, process or product can be improved.
5. Robust design: Deals with reduction of variation in the
system, process or product without elimination of its
causes.
11/15/2022 27
CPS, IST, JNTUH

28. Advantages of DOE
 Maximize process knowledge, with the minimum use
of resources.
 Provide accurate information, in the most efficient
way possible.
 Identify factor interactions.
 Characterize the relative significance of each factor.
 Allow for the prediction of the process behavior
within the design space.
 Establish a solid cause and effect relationship between
CPPs and CQAs.
 Allow for multiple response optimization. As
pharmaceutical products exhibit several CQAs, the
latter require simultaneous optimization.
 Make the product or process more robust
11/15/2022 28
CPS, IST, JNTUH

29. Conclusion
 In conclusion, Design of experiments are statistical
thinking and knowledge management very useful tools
in pharmaceutical development.
 DoE promote operational excellence within the QbD
framework. Moreover, the evolution of the
manufacturing science in the pharmaceutical sector.
11/15/2022 29
CPS, IST, JNTUH

30. References
1. N. Politis S, Colombo P, Colombo G, M. Rekkas D. Design of
experiments (DoE) in pharmaceutical development. Drug development
and industrial pharmacy. 2017 Jun 3;43(6):889-901.
2. Durakovic B. Design of experiments application, concepts, examples:
State of the art. Periodicals of Engineering and Natural Sciences (PEN).
2017 Dec 28;5(3).
3. https://www.slideshare.net/UpendraKartik/design-of-experiments-
75405493 ( Accessed on 2022/09/26)
4. Fang KT, Lin DK. Ch. 4. Uniform experimental designs and their
applications in industry. Handbook of statistics. 2003 Jan 1;22:131-70.
5. Christensen LB, Johnson B, Turner LA, Christensen LB. Research
methods, design, and analysis.
6. N. Politis S, Colombo P, Colombo G, M. Rekkas D. Design of
experiments (DoE) in pharmaceutical development. Drug development
and industrial pharmacy. 2017 Jun 3;43(6):889-901.
11/15/2022 30
CPS, IST, JNTUH

31. Thank you
for your attention
Any question?
11/15/2022
Center of Pharmaceutical sciences, IST,
JNTUH 31