The Business Case for Continuous Manufacturing of Pharmaceuticals

By
Kristopher Ray Wilburn
Bachelor of Science in Chemical Engineering, Rensselaer Polytechnic Institute, Troy, 2002
Submitted to the MIT Sloan School of Management and the Department of Chemical
Engineering in Partial Fulfillment of the Requirements for the Degrees of
Master of Business Administration
AND
Master of Science in Chemical Engineering
OFi
in conjunction with the Leaders for Global Operations Program at the
Massachusetts Institute of Technology JUP
June 2010 LIB
2010 Massachusetts Institute of Technology. All rights reserved
Signature of Author
Department of Chemical Engineering &
MIT Sloan School of Management
May 7, 2010
Certified by
Certified by
ARCHIVES
TECHNOLOGY
N 0 8 2010
RARIES
Bernhardt Trout, Thesis Supervisor
Professor of Chemical Engineering
Ernst Berndt, Thesis Supervisor
Louis E. Seley Professor in Applied Economics
Accepted by
William Deen, Graduate Committee Chairman
Department of Chemical Engineering
Accepted by
Debbie Berechman, Excuitive Director of MBA Program
MIT Sloan School of Management
This page has been intentionally left blank
Business Case for Continuous Manufacturing of Pharmaceuticals
By
Kristopher Ray Wilburn
Submitted to the MIT Sloan School of Management
and the Department of Chemical Engineering
in Partial Fulfillment of the
Requirements for the Degrees of
Master of Business Administration and
Master of Science in Chemical Engineering
May 7, 2010
Abstract
Manufacturing in the pharmaceutical industry is presently characterized as a batch production
system, which has existed in its current form for decades. This structure is the result of historical
regulatory policy as well as the conservative nature of the industry. Recent clarification by US
and European regulatory bodies has opened the possibility to new approaches to the
manufacturing process. This combined with changes in the market for the pharmaceutical
industry has accelerated the rate at which new manufacturing technologies are explored.
Continuous manufacturing is a paradigm shift in the pharmaceutical industry manufacturing
structure, encompassing several new technologies and systems. The business impact of
continuous manufacturing has not been well defined.
This assessment aims to compare a continuous manufacturing process to a batch manufacturing
process for a particular Novartis product. The product has an established batch production
process. Cost estimates and the continuous process cost is estimated using a four-step process:
defining the process flow, performing the material balance, estimating the capital costs, and
estimating the operating costs. This analysis shows that for the particular Novartis product
considered, a continuous process is an improvement over the batch process in four performance
characteristics: capital investment, operating cost, throughput time, and working capital
requirement.
Thesis Supervisor: Bernhardt Trout
Title: Professor of Chemical Engineering
Thesis Supervisor: Ernst Berndt
Title: Louis E. Seley Professor in Applied Economics
This page has been intentionally left blank
Acknowledgements
I would like to acknowledge several individuals who were invaluable in completing this
project. I would first like to thank Walter Bisson for sponsoring the project and for his support
and mentorship during the project. Additionally, I would like to thank several people at Novartis
who contributed to multiple aspects of the project. These people include Marc Achermann, Pat
Brennan, Roland Messer, Berthold Schenkel, and Hedinn Valthorsson. I would not have been
able to complete my work without your expertise and support. I would also like to thank
Sussane Mai and Luanne Shock introducing me to so many great people at Novartis and making
me comfortable in Basel.
I would also like to thank my MIT supervisors, Ernst Berndt and Bernhardt Trout. You
both provided excellent support and advice during the internship and thesis writing.
Finally, I would like to dedicate this thesis to my lovely wife Michelle, for her support
and patience throughout my two years as an MIT student. You are always there to listen to me
and give me advice when I am challenged by the day to day requirements of class work,
internship, and thesis writing. Thank you for being such a wonderful wife who makes my life
better every day.
This page has been intentionally left blank
Table of Contents
Summary of Tables .................................................................................................................. 8
Summary of Figures ................................................................................................................. 8
1 Introdu ction ....................................................................................................................... 10
1.1 Project M otivation ........................................................................................................ 11
1.2 Overview of Thesis....................................................................................................... 12
2 B ackgroun d ........................................................................................................................ 13
2.1 Trends in the Pharmaceutical Industry.......................................................................... 13
2 .2 N ov artis ...................................................................................................................... 14
2.2.1 Pharmaceutical Technical Operations.................................................................... 14
2.2.2 Toyota of Pharmaceutical Industry ........................................................................ 15
2.3 Novartis M IT Collaboration........................................................................................ 16
2.3.1 Pilot Scale Continuous Process ............................................................................. 17
3 H ypoth esis ......................................................................................................................... 18
4 Current State of the Production Process and Supply Chain.................................................... 19
4.1 Overview of the Current Structure .................................................................................. 19
4.1.1 Batch Production................................................................................................ 19
4.1.2 Chemical and Pharmaceutical Operation Split........................................................ 20
5 A pp roach ........................................................................................................................... 2 2
5.1 Previous W ork by M IT Practice School........................................................................ 24
6. Total Production Cost M odel............................................................................................ 25
7 Business Case Analysis ....................................................................................................... 26
7.1 Continuous versus Batch Process................................................................................ 27
7.2 Analysis of Operational Costs .................................................................................... 34
7.2.1 M aterial Costs........................................................................................................ 35
7.2.2 Ecology and Utilities ........................................................................................... 35
7.2.3 Labor Costs, M aterial Handling, and Overhead Costs .............................................. 36
7.2.4 QA Costs............................................................................................................... 37
7.3 Analysis of Capital Costs............................................................................................ 38
7.3 Analysis of W orking Capital...................................................................................... 42
7.4 Application of Cost M odel to Additional Process Steps.................................................. 44
8 C on clu sion s ....................................................................................................................... 4 7
8.1 Key Findings and Specific Recommendations ............................................................ 47
8.2 Questions for Further Research ................................................................................... 48
8.3 Limitations of Analysis ............................................................................................. 49
8.4 Final Comments ........................................................................................................... 50
W orks Cited ......................................................................................... 52
Summary of Tables
Table 1 - Novartis Total Product Cost Categories............................................................................. 25
Table 2 - Comparison of Batch Size and Continuous Flow-rate...................................................... 32
Table 3 - Summary of Operation Costs for Batch and Continuous Processes ................................. 34
Table 4 - Batch and Continuous Primary and Support Equipment Costs ........................................ 39
Table 5 - Additional Equipment-Related Capital Cost Factors ........................................................ 40
Table 6 - Additional Building and Other Capital Cost Factors ........................................................ 41
Table 7 - Batch and Continuous Building Footprint Summary......................................................... 41
Table 8 - Total Capital Investment Requirement and Improvement ............................................... 41
Table 9 - Batch and Continuous Throughput Time.......................................................................... 42
Table 10 - Batch and Continuous Working Capital .......................................................................... 42
Table 11 - Continuous Cost Estimation Factors............................................................................... 45
Table 12 - Batch and Continuous Total Production Costs for Full Process Flow............................. 46
Table 13 - Key Parameters for Continuous versus Batch Comparison ............................................. 47
Summary of Figures
Figure 1 - Pharma TechOps Vision for Operations Performance..........................................15
Figure 2 - Blue Sky Vision for a Continuous Process..........................................................17
Figure 3 - Comparison of Batch and Continuous Supply Chain ......................................... 27
Figure 4 - Batch Process Flow Block Diagram ........................................................................ 28
Figure 5 - Continuous Process Flow Block Diagram ............................................................ 30
Figure 6 - Annual Production Volume for Novartis Product Portfolio............................... 48
This page has been intentionally left blank
1 Introduction
Efforts to improve operational performance within the pharmaceutical manufacturing industry
include the application of Lean Manufacturing methodologies and Six Sigma quality techniques.
These efforts have resulted in significant improvements in throughput times and production
costs. Most efforts to improve the process have focused on working around the batch production
system, which has been the method of production within the industry for decades (Van Arnum,
2007). The combination of industry regulation, familiarity with batch processes, and flexibility
of batch production systems has limited efforts to adopt innovative continuous manufacturing
technologies. Equipment makers and manufacturers have published results of continuous
technologies replacing individual unit operations with a few process steps. This paper will
explore the benefits of a truly integrated continuous chemical and pharmaceutical process for a
specific Novartis product. The Novartis/MIT Center for Continuous Manufacturing has outlined
the process flow for production of a Novartis product in a continuous flow process. By
comparing this process to the actual batch production process, I will evaluate the benefits of an
integrated continuous manufacturing process.
In the final section of the analysis I provide a methodology for estimating continuous production
costs on a per kilogram of drug substance (DS) and process step level for a product similar in
production scale to this product. This methodology is based on the four steps analyzed in detail
in the first part of this thesis. These estimates are based on the integrated system approach of a
continuous process and are not intended to be applied to cost estimates of individual process
steps or unit operations run as a continuous step within an overall batch production system.
1.1 Project Motivation
The sponsor for the work contributing to this analysis is the Technical Operations group within
the Novartis AG Pharmaceuticals division. In 2007 Novartis entered into a collaborative effort
with the Massachusetts Institute of Technology (MIT) providing $65 million over ten years to
develop continuous manufacturing technologies for the pharmaceutical industry. The research
group is titled the Novartis - MIT Center for Continuous Manufacturing (CCM) and consists of
several technical fellows from Novartis as well as 7-10 MIT faculty, graduate students,
postdoctoral fellows, and staff scientists (Buderi, 2007). As research has progressed the CCM
has developed a continuous process flow for a specific commercial Novartis product. The
Novartis program manager for Continuous Manufacturing sponsored the Leaders for Global
Operations project, which generated the hypothesis and analyses in this thesis, in order to answer
the question "What is the business case of this specific Novartis product, manufactured at
commercial scale using the continuous process flow developed by the CCM?"
1.2 Overview of Thesis
This thesis is organized as described below.
Chapter 1 outlines the general motivation for the thesis and provides an overview of the thesis
contents.
Chapter 2 provides a brief discussion of the industry trends and company background as well as
the collaboration between Novartis and MIT to develop continuous manufacturing technologies
for pharmaceuticals.
Chapter 3 presents the hypothesis for the study undertaken.
Chapter 4 discusses the current state of manufacturing processes and supply chain at Novartis,
which is common within the pharmaceutical industry.
Chapter 5 describes the general approach to the analysis and previous work that supports this
approach.
Chapter 6 describes the existing total production cost (TPC) model used by Novartis.
Chapter 7 presents the results of the comparison of the continuous and batch production
processes for total production costs, capital costs, working capital, and cycle time.
Chapter 8 discusses an overview of the business case and a summary of key findings.
2 Background
2.1 Trends in the Pharmaceutical Industry
The industry for patent protected pharmaceutical products is characterized by high-risk research
activities that are rewarded by lucrative patent protection. The industry is also highly regulated
with extensive requirements to prove efficacy and safety for a new product. Even after products
are approved, changes to the formulation and manufacturing process often require new studies
for safety and efficacy. Companies with a strong patented-product portfolio enjoy healthy
margins and therefore manufacturing costs are typically a low percentage of the revenues. The
combination of regulation and healthy margins has not put pressure on the industry to make
advances in manufacturing technologies.
Two recent trends have placed more pressure on pharmaceutical companies with internal
manufacturing organizations to improve the efficiencies of manufacturing operations. The first
trend is the reduced productivity of research and development activities in generating new patent
protected products. With the increased competition of generic pharmaceutical companies,
margins of companies focused on patent protected products will face competitive pressures. The
second trend facing internal manufacturing organizations is the increased capacity and capability
of contract manufacturing organizations and forward integrating suppliers. The outside options
are typically solely focused on manufacturing operations and often have cost advantages over
internal organizations. The make/buy decision is one faced more frequently by the head of
supply chain and operations groups.
2.2 Novartis
Novartis AG is an innovative healthcare products company headquartered in Basel Switzerland.
Novartis AG consists of four business divisions: Pharmaceuticals, Vaccines and Diagnostics,
Sandoz, and Consumer Health. In 2009 Novartis had approximately 100,000 employees and
generated $44 billion in revenue (Novartis AG, 2010).
2.2.1 Pharmaceutical Technical Operations
The Pharmaceuticals business primarily develops and markets patent protected prescription
drugs covering several therapeutic areas including cardiovascular, oncology, neuroscience,
respiratory, and immunology. The Technical Operations group within the Pharmaceuticals
business, or Pharma TechOps, is responsible for manufacturing the Pharmaceutical divisions 50+
products and delivering to a global market. The Pharma TechOps group operates approximately
23 sites worldwide with the majority of the sites in Europe, North and South America, and Asia.
These sites fall into one of three operational groups: Chemical Operations, which is responsible
for the production of small molecule drug substance; Biopharmaceutical Operations, which is
responsible for the production of large molecule biologic drug substance; and Pharmaceutical
Operations, which is responsible for converting drug substance from Chemical Operations and
Pharmaceutical Operations into a dosage form suitable for the customer, called the drug product.
Novartis Pharma TechOps is backwards integrated in the respect that they are producing the
majority of the drug substance including early intermediates. With the trend toward increasing
contract manufacturing capacity and capability in the industry, Pharma TechOps must
consistently provide a cost competitive manufacturing option to the Pharmaceutical division.
2.2.2 Toyota of Pharmaceutical Industry
As the industry trends have increased the competition for low cost, highly responsive
manufacturing, Pharma TechOps has focused on improving operations along key metrics. The
organization has been successful in improving customer service level, reducing cost of goods
sold, and increasing flexibility of the last five to ten years. Pharma TechOps set the goal of
being the Toyota of the Pharmaceutical Industry by adopting many of the lean philosophies and
six-sigma techniques to eliminate waste within all aspects of the manufacturing supply chain.
These continuous improvements have taken place within the structure of the traditional batch
manufacturing process. The Pharma TechOps group believes that a step change in performance
is possible by changing from batch to continuous process technologies. Figure 1 illustrates the
operations performance vision for Pharma TechOps.
Operations
performance
peacess innadon
cuftre
brand supply modelf
Low COGS
CSL > 9%
Future 2005 2010
Figure 1 - Pharma TechOps Vision for Operations Performance.
2.3 Novartis MIT Collaboration
Novartis entered into the collaboration with MIT for several reasons. On one hand the faculty
and graduate research programs are highly their technical expertise in healthcare related science
and technology. Perhaps a bigger challenge to an organization such as Novartis is overcoming
the internal resistance to radical new technologies. It is often advantageous to engage an outside
expert when attempting to make paradigm-changing improvements within an organization.
Novartis maintains a presence in the collaboration by matching internal experts with MIT
researchers focused on specific topics. The collaboration faces certain challenges related to the
nature of the partnership. First is the geographic challenge due to the physical locations of
Novartis (Basel Switzerland) and MIT (Boston USA). The second challenge is related to the
different priorities of for profit companies and research institutions. The Novartis - MIT CCM
has addressed these challenges in part by altering how the CCM is organized and by developing
a roadmap appropriate for both organizations.
The Novartis - MIT Center for Continuous Manufacturing is organized into small teams of six to
eight individuals who are focused on a specific technical challenge. The center's roadmap is
organized along two major efforts termed Blue Line and Red Line. The Blue Line research
focuses on what Novartis describes as Blue Sky technologies. These technologies will enable
the most aggressive vision of a continuous manufacturing process, but they also have a longer
time horizon to possible commercial application. Figure 2 is an illustration of a Blue Sky vision
for a continuous process. The Red Line research focuses on developing a bench scale process
for a specific Novartis product using technologies available on a shorter horizon. The purpose of
the Red Line is to create an opportunity for learning and to deliver a tangible result with
commercial application on a shorter time scale.
ml am
Figure 2 - Blue Sky Vision for a Continuous Process
2.3.1 Pilot Scale Continuous Process
Much of the basis for this thesis comes from the work done by the Novartis - MIT Center for
Continuous Manufacturing's developing a Red Line continuous process. Although the pilot
scale process has not yet been built, the lab scale reactions and separations have been validated
to the point where a material balance is available for the continuous process. In addition to the
material balance, a preliminary equipment configuration has been defined according to
throughput and the material balances requirements at the unit operation level. With this starting
point, it is possible to forecast a commercial scale continuous manufacturing process for the
purposes of comparison to the actual batch manufacturing process.
....................................................
3 Hypothesis
A continuous manufacturing process for pharmaceutical products has the potential to reduce
overall production costs. Using the case study of a commercial Novartis product with a given
annual demand requirement, known batch production cost estimates, and a high level proposal
for a continuous production process, I will show that the following improvements can be
expected:
" Lower capital investment costs and facility footprint.
" Lower operating costs.
" Lower total product cost
" Shorter throughput times.
" Lower working capital requirements.
4 Current State of the Production Process and Supply Chain
4.1 Overview of the Current Structure
The supply chain structure of Novartis Pharma TechOps consists of multiple facilities located
across multiple geographic regions. The supply chain path for a product depends on the
production volume, the process technology, and the number of steps required. An example
product may require Drug Substance synthesis steps taking place across two or three facilities
followed by the Drug Product formulation steps at another facility. The final process step could
be a packaging operation taking place at yet another location. Most products also pass through a
safety stock location in a warehouse at the interface of the Drug Substance and the Drug Product
steps.
4.1.1 Batch Production
An integral aspect to the current production system is defining a batch of material. Process steps,
which may consist of a series of unit operations, have a defined recipe stating the quantity of
starting materials and targeted output materials. The individual unit operations follow a recipe
that controls the process conditions applied to a specific batch of material. After processing is
completed, quality control checks are performed according to specifications. By this approach,
the integrity of a batch is verified and documented. This allows for traceability of materials in
the event that a product is found to be contaminated or have adverse effects to customers. The
paradigm of batch production and post production quality control is common to the industry in
large part due to the fact that pharmaceutical products are regulated by government agencies.
The batch model does have some additional advantages. In an innovative company such as
Novartis, the flexibility in process steps that a batch mode provides is important as new drugs are
being continuously developed. New drugs often require unique processing pathways, which can
be simply accommodated by careful selection and ordering of the batch processing equipment
already in place. Thus, no large structural rearrangement is necessary, which cuts down on
equipment costs. These batch processes are also well understood, currently well optimized for
the operations they perform and readily available for use as the equipment is already in place.
Also, the scale up from lab scale is relatively straightforward conceptually, in that the batch
processing equipment resembles the equipment used in the lab.
4.1.2 Chemical and Pharmaceutical Operation Split
As described in the overview of the supply chain structure, there is a physical split in the supply
chain between the chemical and biopharmaceutical operations that produce the active
pharmaceutical ingredient or drug substance, and the pharmaceutical operations that produce the
drug product. In addition to the physical separation of facilities, these are also separate
organizations within a firm such as Novartis. One reason for the separation is the nature of the
physical processes taking place in each facility. A chemical operations facility consists of
equipment designed to perform the unit operations required in chemical synthesis including
reactions and separations. A pharmaceutical operations facility typically consists of equipment
designed to perform the unit operations required in creating a drug product formulation including
powder mixing and tableting. In addition to the different unit operation requirements, there is
also a difference in the complexity of market requirements in a chemical operations facility as
compared to a pharmaceutical operations facility. Drug substance produced in a single chemical
operations site may satisfy the requirements for multiple pharmaceutical operations because the
drug substance is essentially a raw material. Pharmaceutical operations must produce multiple
dosage sizes, multiple combinations therapies, and serve multiple markets with different
packaging requirements due to language and labeling needs. For this reason the chemical
operations facilities can typically take advantage of economies of scale in production that the
pharmaceutical operations cannot. This complexity in the pharmaceutical operation also means
that the pharmaceutical operations are typically more labor intensive from a direct labor and
overhead perspective than are the chemical operations facilities in the same supply chain.
Conversely the chemical operations facilities typically have higher fixed costs and are more
capital intensive than the pharmaceutical operations.
In addition to the internal reasons for this structure, it is important to note that the regulatory
environment also contributes to this structure. As the product flows through the supply chain,
the level of quality control requested by the regulatory bodies increases. This means that the
type of analytical equipment and support staff required onsite is often different for a
pharmaceutical operation facility than for a chemical operation facility.
5 Approach
The approach taken in this thesis to estimate the costs of both systems, and consists of four basic
steps.
1. Calculate the annual demand of drug substance required based on the forecasted annual
demand of doses. For the batch process, translate the annual demand into a daily
capacity requirement. For the continuous process, translate the annual demand into flow-
rates requirements based on 24 hours by 7 days per week operation.
2. Perform a material balance based on the unit operation yields, process inputs, and waste
streams. Combine this with the annual demand to determine batch sizes and unit
operation flow-rates.
3. Size equipment at the unit operation level based on the batch size or flow rate and the
residence time. At this point one should characterize the ideal throughput time for the
batch and/or continuous process. Estimate the price for the unit operations equipment and
support equipment. Estimate the total capital investment costs based on all costs required
to design, build, and qualify all required facilities.
4. Estimate the operating components of the Total Product Cost. Estimate the annual
material, ecology, and utility costs based on the material balance information. Estimate
the direct labor cost based on the operating requirements of the facilities. Estimate the
material handling cost based on the number of facilities required, warehousing
requirements, and incoming material inspection requirements. Estimate the quality
assurance cost based on the level of offline labor-intensive analytical requirements and
documentation requirements. Estimate the overhead costs based on the organizational
structure of the overall supply chain as well as the number of products produced in the
facility. The accuracy of these estimates is largely dependent on the level of detail
available with respect to the facility process design.
5.1 Previous Work by MIT Practice School
Two previous business case analyses for comparison of a continuous process versus a batch
process were performed by teams of students from the MIT Practice School within the Chemical
Engineering department. These groups used the same approach described above, but did not
include the Total Product Cost operating components such as overhead, quality assurance, and
material handling. The previous groups also did not have the benefit of a defined continuous
process with validated yields and a preliminary material balance. The results of this thesis are
generally consistent with the work of the MIT Practice School (Alf, Barr, & Couling, 2008).
One component explored by the Practice School reports is the comparison of a multipurpose
batch facility with multiple dedicated continuous lines (Barr, Musolino, & Xu, 2008). From a
practical perspective the multiproduct case is overly general and a better comparison requires a
detailed understanding of the process requirements for each product.
6. Total Production Cost Model
The cost categories included in this comparison are based on the standard cost categories used by
Novartis Pharma TechOps' internal cost accounting system (Brennan & Goerke, TechOps
Finance Product Costing Overview, 2008). Novartis uses a calculation called Total Product Cost
to set a cost on what they intend to produce. The Total Product Cost consists of Materials Costs
and Processing Product Costs. Material Costs consist of price of materials from vendors and toll
manufacturing fees. For this product Novartis did not charge any toll manufacturing fees; hence
this is not included in the analysis. The Processing Product Costs include several costs
categories associated with converting the materials from raw materials to finished goods. Table
1 list the Processing Product Cost categories and describes whether these are fixed or variable
and direct of overhead costs.
Category Direct or Overhead Variable or Fixed
Utilities Direct Variable
Ecology Direct Variable
Material Handling Direct Variable
Direct Labor Direct Variable
Quality Assurance Direct Variable
Production Area Overhead Indirect Fixed
General Factory Overhead Indirect Fixed
Building and Equipment Direct Fixed
Depreciation
Table 1 - Total Product Cost Categories
7 Business Case Analysis
The results of my analysis indicate that switching to a continuous process for the last four steps
of a fifteen step process results in a 36% reduction on total production costs for the steps
analyzed, which results in a 14% reduction in the total cost per tablet. Furthermore the
throughput time of the four steps decreases by 90% and working capital requirements decreases
by 90%. All absolute product costs in this analysis have been normalized in order to protect the
confidentiality of Novartis' product cost information. The percentage improvements are based
on the calculations made during the project utilizing actual Novartis product cost information.
The cost components are broken down by the operational components and the capital investment
requirements. The drivers of each cost category are discussed in more detail below.
In addition, the total production cost is estimated for the product assuming a continuous process
is developed for a full fifteen step process. A methodology is developed from the more detailed
analysis of the last four process steps and applied to the early process steps for a more complete
comparison of a continuous versus a batch production process.
7.1 Continuous versus Batch Process
The primary structural change to the supply chain is a result of combining the chemical and
pharmaceutical operations, which operate in separate facilities for the current batch process
supply chain, into a single facility. In addition all unit operations are linked to allow for a
continuous flow of material from beginning to the end of the process. Figure 3 illustrates the
change to the supply chain for an integrated continuous manufacturing process.
Batch Supply Chain
Continuous Supply Chain
Integrated Site(s) 100ton/year per Prcess Line
Integrated
Site(s)
1 .00to n/year
per
Process
u ne
-I
Integrated Site(s) 100ton/year per Process Line
Qualified Markets
Qualified Markets
Figure 3 - Comparison of Batch and Continuous Supply Chain
Figure 4 is a high-level process flow diagram for the current batch manufacturing process. The
chemical synthesis of the active pharmaceutical ingredient (API) involves three chemical
reactions. The three reactions will be referred to as Cl 0, CI1, and C13 and they take place in a
Novartis Chemical Operations site. Each reaction step also requires processing in addition to the
reaction step in order to purify the desired product streams for the subsequent step. Additionally,
the C14 step is a mechanical milling of the C13 intermediate product in order to obtain the
Starting
Materials
Starting
Materials
SiteChemical
Chemical
Site 2
official API. Once the desired API is created, it is converted to a coated tablet as the drug
product (DP) form. This takes place at one of multiple Novartis Pharmaceutical Operations sites.
Figure 4 - Batch Process Flow Block Diagram
The C10 step involves a reaction step, which takes place in an agitated vessel. This is followed
by a solid/liquid separation step, which takes place in a nutsche. The final step is a drying
process that takes place in a paddle dryer to store the intermediate product as a dry powder
(Messer, Novartis Product X Base Suspension, 2009) (Messer, Novartis Product X Base, 2009).
The C1I step consists of a gas/liquid reaction step, which again takes place in an agitated vessel.
This is followed by a separation in a nutsche. Then an additional reaction and liquid/liquid
..........
....... .... 40NOWWW"N
separation take place in an agitated vessel. The solution is then distilled and solvent is added to
store the intermediate product stream in a solution (Messer, Novartis Product X Base Sol, 2009).
The C13 step consists of a reaction step, which takes place in an agitated vessel. This is
followed by a crystallization step to purify the product stream, which takes place in a stirred
vessel. The next step is a solid/liquid separation that takes place in a centrifuge. The final step is
a drying process that takes place in a paddle dryer to store the intermediate as a dry powder
(Neville, Novartis Product X Salt, 2008).
The C14 step consists of a mechanical milling of the C13 intermediate dry powder in order to
achieve a rough powder size distribution prior to the DP step. This step takes place at a different
Novartis Chemical Operations site from the C 10, CI1 and C 13 steps (Clausen).
The DP steps consist of a series of mixing steps, which combine the API with the remaining
formulation ingredients. The primary ingredients are first mixed in a wet granulation vessel.
The mixture is then dried in a fluidized bed drier and sieved to obtain a targeted particle size
distribution. The remaining formulation ingredients are then added in a blending vessel. The
formulation is then converted to a tablet using a tablet press. The final step for the bulk drug
product is a coating step, which takes place in a pan coater (Desset-Brethes & Leboulanger,
PHAD000160A).
Figure 5 is a high-level process flow diagram for the new continuous manufacturing process.
The process flow is grouped with highlighted chemical reaction steps as they correspond to the
batch process in terms of chemical reactions. The process equipment is completely different and
the C14 step is eliminated in the new process flow. The DP is still a coated tablet, however the
core of the tablet is now based on a thin film polymer material rather than a compressed powder.
Figure 5 - Continuous Process Flow Block Diagram
The C10 step utilizes new reaction chemistry with a shortened reaction time. This allows for a
reasonable residence time and reactor volume for the flow chemistry. The reactor is a
combination of glass plate micro-reactors for efficient mixing and a static mixer tube to achieve
the required residence time for the reaction to reach completion (Foley & Heider, 2009). A
countercurrent liquid/liquid separation is used to split one of the excess starting materials. The
other excess starting material is recovered during a continuous crystallization process. The
crystallization process is carried out in a series of agitated vessels. One recycle loop is required
in this process segment in order to improve the overall material yield and process economics.
The recycle loops primary equipment is a falling film evaporator used to separate the solvent
material.
. . ......... ..
The CI1 step involves a similar chemistry as the batch process, but replaces a gas phase acid
reactant with an aqueous phase acid reactant. This short reaction can be accomplished in a tube
reactor with sufficient volume to achieve the target yield. A neutralization process follows with
an aqueous base mixed in a micro-mixer reactor with short reaction time limited by mass
transfer. The product stream is then purified by separating the salt in aqueous phase from the
product with an organic solvent. This separation occurs in a countercurrent liquid/liquid
extraction column.
The C 13 step involves the salt formation step, which takes place in an agitated vessel running in
a continuous mode. The agitated vessel is a relatively small sized based on the product stream
flow-rate and the residence time requirement. The next steps are a continuous crystallization,
filtration, and drying required to achieve a pure product stream. All of this equipment is based
on standard designs but adapted for continuous flow and lower throughput rates of a continuous
system.
The DP steps result in a coated tablet, but use a very different process and materials to achieve
the end product. The dry powder is dissolved and blended with a soluble medical grade polymer
material. This mixing takes place in extruder type equipment capable of achieving high
uniformity with a low internal volume. The liquid is then transformed into a tablet and coated
using a continuous coating process.
The analysis compares both the operational and investment costs for the Novartis standard
product costing categories. In addition the analysis examines the impact on throughput time and
working capital. In order to estimate costs I use an annual demand of 300 Tonnes of active
pharmaceutical ingredient with a 50% drug loading. This annual demand is based on actual
demand requirements for this product and the capacity for the primary batch chemical production
facility. (Brennan, Novartis Product X Demand Forecast, 2009) (Neville & Roland, ChemOps
CH - PU WSH, 2009) The batch sizes were taken from the actual production process and used
to calculate the number of batches per year at each step in the process. For the continuous
process, the annual demand is translated into an hourly volume flow-rate through each unit
operation. This calculation assumes around the clock operation with four weeks of shutdown
during the year. The last three chemical synthesis steps of a fifteen-step synthesis and the bulk
drug product formation are the steps that are considered for the continuous flow analysis. Table
2 describes how the annual demand translates to batch sizes and the number of batches as well as
continuous material flow-rates at multiple stages of the process.
Batch Size Number of Batches Continuous Flowrate
C1O 943 kg 406 19 kg/hr
C11 1115 kg 274 12.6 kg/hr
C13 707 kg 431 12.5 kg/hr
DP 257 kg 2334 25 kg/hr
Table 2 - Comparison of Batch Size and Continuous Flow-rate
The continuous process as described in the previous section was sized based on some of the
commercially available equipment and the annual demand. For this case the current batch
production capacity was replaced by three parallel continuous lines, which are capable of 100
metric tons per year of drug substance equivalent production. It is possible that further
reductions in building footprint and equipment investment costs are possible by scaling up the
equipment. Three reasons to keep the continuous process lines this size are avoiding scale-up
issues from development to production, risk mitigation with parallel production lines, and
potential for a distributed production network. Section 7.3, which discusses capital investment
costs, is based on three parallel lines for the continuous production process.
7.2 Analysis of Operational Costs
The operational costs of the continuous process are driven by the material balance of the process,
the layout of the plant, and the operating strategy of the plant. The material balance and process
conditions drive the material, ecology, and energy costs. The integrated structure of a
continuous process and the control strategy drives the labor, material handling, QA, and
overhead costs. All costs reported refer to the cost per kg of DS equivalent in a 50% drug load
coated tablet form.
The comparison of the batch and continuous process operation costs are outlined in Table 3. The
absolute costs are a normalized to a 300 mg tablet that cost CHF 1.00 to produce in the current
batch process.
Category Batch Process Continuous Process Percent Improvement
(CHF/kg DS) (CHF/kg DS) Batch to Continuous
Material 426 309 27%
Utilities 19 13 31%
Ecology 26 19 25%
Material Handling 82 36 56%
Direct Labor 178 89 50%
Overhead 171 137 20%
Quality Assurance 115 32 72%
Table 3 - Summary of Operation Costs for Batch and Continuous Processes
7.2.1 Material Costs
When comparing the material costs of a continuous process versus a batch process there are two
primary drivers of cost reductions. The first is an improved reaction pathway that is enabled by
the continuous process and the second is an improvement in reaction yields due to tighter process
control from improved heat and mass transfer (Weiler & Junkers, 2009) (Lavric & Woehl, 2009).
In this case the continuous process achieves a 24% reduction in material cost through a new
reaction pathway in the C10 through DP steps. The new pathway achieves a higher overall yield
than the batch process. This pathway reduces cost by elimination, reduction, or replacement of
solvents. Pilot scale production will be required to validate the improvement in yields due to
improved heat and mass transfer properties and better control of reaction conditions. Tighter
control of reaction conditions can yield fewer side products thereby simplifying or eliminating
separation and purification steps. Companies developing continuous flow micro-reactors have
tested several reaction types with demonstrated improvements in reaction yields (Braune, et al.,
2008).
7.2.2 Ecology and Utilities
The ecology cost is calculated based on the material balance and category of waste stream
produced. The primary drivers of ecology costs for this product are water treatment and
incinerated waste. The continuous synthesis pathway eliminates or replaces some solvent
requirements. This combined with an improvement in yield reduces the waste streams that
require treatment. For this product and process the reduction in ecology cost is 25%. The
utilities costs are driven by three main components; the unit operation process temperatures, the
unit operation mass flow-rates, and the building footprint HVAC. As the footprint requirement
is decreased by 60%, the reduction in HVAC energy requirements represents the largest decrease
in the utility costs. The improved reaction pathway does, however, require different process
temperatures during two of the process steps. In the first step the chemistry is run at a higher
temperature but with less solvent mass to heat and with a reaction time that is 1/10th of the batch
process. In the second step the chemistry is run at ambient conditions in a very short reaction
time. The overall decrease in utilities cost is 30%.
7.2.3 Labor Costs, Material Handling, and Overhead Costs
These four operational costs were grouped together because the improvements are largely due to
the structure of a fully integrated continuous process. If individual steps in a process are
transformed to continuous operation, then, with the exception of QA, these cost are not expected
to decrease, and in some cases may increase.
To model the labor costs I first determined that the primary tasks for the new plant are control
room operations, equipment repair, and preventative maintenance (Roberge, Zimmermann,
Rainone, Gottsponer, Eyholzer, & Kockmann, 2008). Based on the projected number of
technicians required to support a continuous line and the need to support 24x7 operations, I
calculated a total headcount requirement to be 50% less than batch for the continuous process.
The material handling activities between process steps and between sites will be eliminated.
Purchasing, receiving raw materials, and shipping finished product would be the only remaining
material handling steps in the process. This results in a 55% decrease in material handling costs.
Overhead costs cannot be predicted with an exact model due in large part to unknowns regarding
how much support staff and site services would be required for an integrated process. At a high
level, the reduction of plant footprint and a reduction in the number of facilities in the supply
chain should reduce overhead costs. On the other hand a company's supply chain strategy will
largely influence the potential for overhead gains. The footprint and scalability of a continuous
process gives a company the flexibility to have multiple smaller regional plants or fewer larger
centralized plants. For this process comparison I projected a 20% decrease in overhead costs.
7.2.4 QA Costs
A continuous manufacturing process running at steady state requires a process characterization
and a robust process control system utilizing PAT in order to meet all regulatory requirements.
The traditional post process testing currently in place for most products is highly labor intensive
and insufficient to control a continuous process (Swaminathan & Zega, 2005). Real time release
of a Novartis product manufactured in batch processes has been demonstrated by application of
PAT tools with a demonstrated elimination of 90% of labor costs. The total QA costs for the
continuous process are estimated to be 72% less than the current batch process.
7.3 Analysis of Capital Costs
In this example a supply chain network consisting of one chemical plant and four pharmaceutical
process lines is replaced by a single production facility. In the chemical synthesis steps, large
vessels used for reactions, separations, and crystallizations, are replaced by micro reactors,
continuous extraction, and continuous crystallization equipment. In the pharmaceutical steps,
blenders, fluidized bed dryers, and tablet presses are replaced by new equipment such as thin
film forming coaters and thin film cutting and forming equipment. This equipment is able to
meet the annual output requirements with a 60% reduction in building footprint. The continuous
process does require additional investment in process analytical technologies (PAT) in line with
Quality by Design (QbD) principles in order to achieve real time product release. The
investment costs were annualized according to Novartis cost accounting practices in which
equipment investments are depreciated with a 15 year straight line method (Brennan & Goerke,
TechOps Finance Product Costing Overview, 2008). Again, the absolute costs in this section are
normalized to a 300 mg tablet that cost CHF 1.00 to produce in the batch process.
The equipment costs are driven by three main factors: equipment sizing, number of process
operations, and process equipment complexity. The equipment is dramatically smaller than
batch equipment required to achieve the same annual output. The primary equipment
components such as micro-reactors have a relatively high cost to equipment space ratio
compared to standard batch equipment. The larger equipment cost decrease comes from a
decrease of support equipment costs due to smaller pumps, heat exchangers, and makeup tanks
(Achermann, 2009). The continuous process in this example reduces the number of process unit
operations from 19 to 14. This is achieved by elimination of some steps required for separation
and drying of intermediates. The continuous process also eliminates unit operations at the
interface between chemical synthesis and pharmaceutical formulation steps. The process does
require additional analytical and control system investments in order to fulfill all requirements
for real-time release consistent with QbD strategies (Pellek & Van Arnum, 2008). A large
number of these costs would also be required to achieve real time release goals for a batch
process; however I observe these are required for the continuous process. Table 4 outlines the
basic equipment and support equipment requirements by process step.
Step Batch Equipment and Support Costs Continuous Equipment and Support Costs
C1O CHF 6,918,000 CHF 10,393,200
Cli CHF 8,201,700 CHF 5,352,000
C13 CHF 19,561,200 CHF 11,116,500
DP CHF 22,624,808 CHF 15,060,000
Total CHF 57,305,708 CHF 41,921,700
Table 4 - Batch and Continuous Primary and Support Equipment Costs
Costs for installation, engineering design and qualification are expected to remain similar to the
batch process as a percentage of the total investment costs. The costs due to additional
complexity are offset by the smaller equipment size. In addition, I expect some economies of
scale due to parallel scale-up. The cost for central facilities and HVAC are reduced relative to
the cost of the primary equipment due to the lower peak demand requirements and small
footprint of the facility. The operational and performance qualification is expected to double as a
percentage of the total investment costs due to the increased complexity of control systems and
testing of integrated equipment.
Table 5 and Table 6 show the additional equipment related and building related costs. Table 5
details the equipment installation and qualification costs as a percentage of the basic equipment
and support equipment costs. For example, to calculate the piping cost for the C1O process one
would multiply the C10 Batch Equipment and Support Cost (CHF 6,918,000) by the Piping
Factor in Table 5 (50%) for CHF 3,459,000. Table 6 details the building and facilities related
costs as a percentage of the Equipment and Support Equipment costs. The same method is used
to calculate the Additional Building Related Capital Costs. The total Capital Costs for the
process are calculated by summing the Additional Building Related Capital Costs and the
Additional Equipment Related Capital Costs.
Additional Equipment Related Capital Cost Factors
Cost Area Batch
1. Erection 5%
2. Piping 50%
3. Electrical 15%
4. Automation HW + Instruments 25%
5. Automation SW + Tests 40%
6. Engineering 87%
Total 222%
Table 5 - Additional Equipment-Related Capital Cost Factors
Continuous
5%
50%
15%
40%
60%
87%
257%
Additional Building Related Capital Cost Factors
Cost Area
1. Building
2. HVAC
3. Process Infrastructure
4. Other
Total
Batch
150%
60%
45%
50%
305%
Continuous
50%
30%
45%
50%
175%
Table 6 - Additional Building and Other Capital Cost Factors
The most significant cost difference from the batch to the continuous process from Table 6 is the
building related space costs. The building footprint requirements of a continuous plant are 60%
less than that of the batch process. This is driven by a reduction in the space required by the
primary process equipment and the support equipment as well as an elimination of warehouse
space required for intermediate material storage. The total office space requirement remained
constant for this analysis. Table 7 outlines the building space requirements by category.
Process Equipment
Infrastructure Space
Walk Areas
Warehousing/Storage
Office
Total
Combined Batch
(m2)
5,940
2,454
1,360
1,282
1,188
12,224
Footprint
49%
20%
11%
10%
10%
Continuous Footprint
(m2
1,000
1,000
400
422
1,188
3,950
25%
25%
10%
11%
23%
Table 7 - Batch and Continuous Building Footprint Summary
Table 8 shows the overall comparison of the capital investment requirements for the batch
manufacturing process and the continuous manufacturing process.
Total Capital Investment Total Capital Investment Percent Improvement
Batch Process Continuous Process
CHF 572,541,329 CHF 338,203,315 41%
Table 8 - Total Capital Investment Requirement and Improvement
7.3 Analysis of Working Capital
The continuous process achieves a dramatic reduction in throughput time from intermediate
starting material to bulk drug product. This directly reduces the level of working capital
required. In addition a safety stock inventory location can be eliminated further reducing overall
working capital. The Throughput Time and Working capital improvements are highlighted in
Table 9 and Table 10.
Throughput Time Residence Time QA + Storage Time Transport Time
Batch 17 Days 30 Days 5 Days
Continuous 3 Days 2 Days 1 Days
Table 9 - Batch and Continuous Throughput Time
Working Capital Pipeline Inventory Safety Stock Total
(Days of Inventory) (Days of Inventory)
Batch 52 49 101
Continuous 6 6 12
Table 10 - Batch and Continuous Working Capital
The continuous process developed for this product achieves shorter throughput times for three
reasons. The first is a decrease in reaction times and separation step residence times. This is in
part due to the reaction chemistry that is chosen and in part due to the improved heat and mass-
transfer characteristics, which allows for shorter process times within unit operations. The
second reason is the elimination of process steps. One example is the elimination of a drying
step required to store an intermediate in a stable condition. The final factor is the elimination of
material holdups between process steps and transportation requirements between sites. Each
process step has a buffer to hold batches between process steps. For the product considered here,
this buffer time is low because the production takes place in a plant dedicated to this product.
Two site-to-site shipping steps as well as a warehousing operation are eliminated by integrating
the chemical synthesis and drug product formation,. The throughput time improvement for this
process is a decrease from 52 days (Hafeli, 2008) to 6 days.
The throughput time improvement has a corresponding impact on the working capital required at
a steady state production level. With the steady state production assumption one can use Little's
Law (1= R*T, I-Inventory in the System R-Steady State Demand T-Throughput Time) to
calculate the inventory in the system. Reducing the throughput time from 52 to 6 days would
lead to a decrease in working inventory of 90%. In addition to the decrease in working
inventory, a benefit of the continuous process is the elimination of a safety stock step at the
interface of chemical synthesis and drug product formation. In order to achieve desired customer
service levels, Novartis includes safety stock in the manufacturing supply chain. The customer
service level goal, process lead-time, and demand variability determine how much safety stock is
held at certain steps in the process. For the batch process there is a safety stock target of 49 days
of inventory at the interface between chemical synthesis and drug product formation. This safety
stock inventory is shifted to finished product inventory. Because the lead-time is reduced by the
continuous process, the level of the safety stock added to finished product inventory would be
reduced to 6 days of inventory.
7.4 Application of Cost Model to Additional Process Steps
The in depth analysis in the previous sections provides a framework for estimating the total
production costs of future products. This methodology applies to an integrated continuous
process as outlined in the introduction. The following stepwise approach is not significantly
different from cost estimation for a batch process, however general cost estimation factors
derived from the detailed case example are applied to the stepwise approach.
1. Perform a mass balance to derive the material costs. The continuous process may allow for
improved synthesis paths or better yields, but these must be validated at the lab or pilot scale.
2. Estimate the equipment and building depreciation costs. This should be done in detail based
on a process flow diagram when possible. If the detailed equipment list for a continuous process
in unknown, one can use a general approximation method to calculate the equipment cost. A
simple approach is taking the number of process steps (defined as chemical conversion steps)
and multiplying by a general factor.
3. Determine the labor headcount required to run a continuous process line. The labor model
utilized in this approach is having a dedicated control room headcount and headcount used for
equipment maintenance and rapid equipment failure response. The total headcount per shift can
be estimated at a minimum of 4 per process line and adding an additional headcount for every
two process steps worth of equipment. The plant is expected to run in a 4 shift operation model
for 24x7 production. The local labor rate is applied as appropriate.
4. Material handling, QA, utilities, and ecology were each largely derived from the material
balance and factory footprint. Based on the case example, these costs did not change
significantly as a percentage of the total costs versus a batch process with the exception of QA
costs, which are reduced by a half. Table 11 gives the percentage of total TPC for each
component. Also included here are the overheads. These are difficult to estimate and are
dependent on overall operational strategy of the organization.
Category Percentage of TPC
Material Handling 1.7%
Quality Assurance 2.0%
Utilities 2.5%
Ecology 4.8%
Production Area Overhead 4.2%
General Factory Overhead 4.0%
Table 11 - Continuous Cost Estimation Factors
Summing each component can give a high level approximation of the product TPC when using a
continuous manufacturing process. This estimate is at best in the 30% accuracy range as
determined by a sensitivity analysis of key factors that drive cost. Chemical yield accuracy,
equipment cost accuracy, and equipment installation cost accuracy are the factors that contribute
to the overall cost estimate accuracy. The overall benefits come from the opportunity to pursue a
more efficient overall chemical synthesis, elimination of intermediate steps and material
handling, a less labor-intensive process, better QA systems, and a smaller factory footprint.
By using this approach I have estimated the total product cost by extending a continuous process
to the previous 10 production steps. The material costs are estimated to decrease by 5%, which
is a conservative estimate based on the detailed analysis in the previous section. The additional
total production cost components are estimated using the table of factors described above. For
this product material costs are higher than average, even for the early synthesis steps and
therefore the overall gains are limited by the improvement in material costs due to yield
improvement or alternate synthesis pathways.
Table 12 provides a summary of the total product cost per kg of DS in a 50% drug load of a
coated table assuming a continuous production system for the 14 step chemical synthesis and
drug product formation.
Category
Material
Direct Labor
Bldg & Equip
Material Handling
Quality Assurance
Utilities
Ecology
Production Area Overhead
General Factory Overhead
Total
Improvement
21%
50%
33%
56%
64%
21%
29%
20%
20%
29%
Table 12 - Batch and Continuous Total Production Costs for Full Process Flow
8 Conclusions
8.1 Key Findings and Specific Recommendations
When considering a product that has an annual production level of 300 tons of drug substance
per year and a 50% drug load, continuous manufacturing presents an advantage over batch
manufacturing in several areas. Table 13 highlights the comparison results for the conditions
presented in this analysis. The total production cost incorporates an annualized capital
investment as well as all operating costs. The working capital improvement also reflects the
throughput time of the process. These results support the hypothesis that there is a positive
business case for continuous manufacturing processes based on an improvement in capital
investment costs and facility footprint, operating costs, Total Product Cost, throughput times, and
working capital requirements.
Percent Improvement Percent Improvement
Cl0 - DP Full Process
Total Production Cost 21% 30%
150mg Tablet
Total Production Cost 19% 28%
300mg Tablet
Capital Investment 50% 42%
Working Capital 91% 90%
Table 13 - Key Parameters for Continuous versus Batch Comparison
8.2 Questions for Further Research
Additional research should consider the case for other products with different annual production
volumes and number of products. Error Reference source not found.highlights the fact that
for a company such as Novartis, product volumes requirements vary widely. The advantage
often cited for batch manufacturing is the flexibility to handle low volume products by running
campaigns for many products using the same equipment. Further research could examine the
business case of a multipurpose batch facility with a facility producing multiple products on low
volume dedicated continuous process lines.
100
80 -Natural Products
&
Highly Actives
40
Multipurpose
20-
High Volume
0
0 500 1000 1500 2000
Production Volume [tly]
Figure 6 - Annual Production Volume for Novartis Product Portfolio
8.3 Limitations of Analysis
The conclusions of this thesis are limited to volume range and greater accuracy of estimates can
be obtained by a refinement of the cost estimations. This analysis covers a range of active
pharmaceutical ingredient volume of 100 - 300 tons per year. Furthermore, both the batch and
continuous processes were defined as dedicated to one product. As the volume of production
scales to less than 100 tons per year, the conclusions of this thesis may not apply. Dedicated
batch equipment is expected to lose economies of scale more quickly than continuous equipment.
On the other hand, products with lower annual production volumes are typically produced in a
plant with mixed use equipment.
The accuracy of cost estimates for continuous equipment and a plant designed for continuous
will improve as equipment design requirements are refined and plant automation and
infrastructure is designed in detail. Collaboration with equipment vendors is required to refine
equipment designs and obtain better cost estimates. A detailed engineering plant design is
required with piping requirements, automation and instrumentation requirements, and support
infrastructure.
8.4 Final Comments
The outlook for continuous manufacturing of pharmaceuticals is quite positive despite some
barriers to implementation. Many of the process equipment technologies described in this
business case are available today. Close partnerships between pharmaceutical manufacturers and
equipment suppliers should be able to close the gap on remaining technologies. Replacing
existing capital investments with new technologies is perhaps the largest barrier to
implementation. This requires more than a simple business case for the new investment. It
requires a change to the skills and structure of the current manufacturing organization.
Furthermore, for wider transition to continuous manufacturing as the standard for the supply
chain, a toolbox of process technologies and reaction chemistries must be developed to support a
wide variety of active pharmaceutical ingredients and drug product formulations.
It is likely that a pilot line process could be running in three to five years based on one to two
years of additional technology development for the process described in this report. Most capital
projects require two years from start of design work to process qualification. This line should
provide valuable learning for additional process lines. Novartis should focus on developing a
continuous process for new, higher volume products first and then converting existing high
volume products where the business case makes sense. A realistic time horizon for
implementation of new of products and processes to continuous production in five to ten years is
envisaged. The business case supports this transition as technology, regulatory understanding,
and organizational skills evolve.
This page has been intentionally left blank
Works Cited
Achermann, M. (2009, July). Equipment Cost and Installation Cost. Novartis Internal
Communications .
Alf, M., Barr, M., & Couling, D. (2008, February). Economic Impact of Continuous
Manufacturing Technologies. MIT-NPC-08SP1 . MIT David H. Koch School of Chemical
Engineering Practice.
Barr, M., Musolino, N., & Xu, J. (2008, March). Economic Impact of Continuous Manufacturing
Technologies: Part II. MIT-NPC-08S4 . David H. Koch School of Chemical Engineering
Practice.
Braune, S., Pochlauer, P., Reintjens, R., Steinhofer, S., Winter, M., Lobet, 0., et al. (2008).
Selective nitration ina microreactor for pharmaceutical production under cGMP conditions.
Chemistry Today, 26, 1-4.
Brennan, P. (2009, July). Novartis Product X Demand Forecast. Internal Novartis
Communication .
Brennan, P., & Goerke, A. (2008, February 1). TechOps Finance Product Costing Overview.
Novartis Internal Document.
Buderi, R. (2007, September 28). Novartis-MIT Center Aims To Transform Drug-Manufacturing
Processl. Retrieved January 15, 2010, from xconomy Boston:
http://www.xconomy.com/boston/2007/09/28/novartis-mit-center-aims-to-transform-drug-
manufacturing-process/
Chang, E. (2009, August). Thin Film Process. Novartis/MIT CCM Internal Communications.
Chirkot, T. S. (2009). The Relevance of Continuous Solid Oral Dosage Processing and NIR
Spectroscopy In Meeting the Needs of QbD and PAT. Pharmaceutical Technology, 33 (10),
112-120.
Clausen, C. (n.d.). Novartis Product X Salt Formation/DS. Novartis Internal Document.
Desset-Brethes, S., & Leboulanger, B. (PHADOOO60A). Extract from Pre-validation report -
150mg Film coated tablet. Novartis internal document.
Foley, M., & Heider, P. (2009, Oct). Continuous Chemistry and Process Flow. Novartis/MIT
CCM Internal Documents .
Gehrhardt, H. (2009, June 17). Novartis Product X Synthesis. Novartis Internal Document.
Hafeli, R. (2008, September 23). Novartis Product X Lean Link Practical Vision. Novartis
Internal Communication.
IBISWorld. (2009). Pharmaceutical & Medicine Manufacturing in the US. IBIS World.
LaPorte, T. L., Hamedi, M., DePue, J. S., Shen, L., Watson, D., & Hsieh, D. (2008).
Development and Scale-Up of Three Consecutive Continuous Reactions for Productinos of 6-
Hydroxybuspirone. Organic Process Research & Development, 12, 956-966.
Lavric, E. D., & Woehl, P. (2009). Advanced-Flow glass reactors for seamless scale-up.
Chemistry Today, 23, 45-49.
Maribel, R. (2007, April 2). Continuous Processing-Finally. Pharmaceutical Technology.
Martin, C. (2008, October). Continuous Mixing of Solid Dosage Forms via Hot-Melt Extrusion.
Pharmaceutical Technology, 76-86.
Messer, R. (2009, 1 12). Novartis Product X Base. Novartis Internal Document.
Messer, R. (2009, 1 29). Novartis Product X Base Sol. Novartis Internal Document.
Messer, R. (2009, 1 12). Novartis Product X Base Suspension. Novartis Internal Document.
Neville, D. (2008, 10 30). Novartis Product X Salt. Novartis Internal Document.
Neville, D., & Roland, M. (2009, June 26). ChemOps CH - PU WSH. Internal Novartis
Communication .
Novartis AG. (2010, January 31). Welcome to Novartis. Retrieved from Novartis Corporate Web
Site: http://www.novartis.com/about-novartis/our-businesses/index.shtml
Pavlou, F. (2009). Microreactor technology: Is the industry ready for it yet? Pharmaceutical
Technology, 21 (10).
Pellek, A., & Van Arnum, P. (2008). Continuous Processing: Moving with or against the
Manufactruing Flow. Pharmacuetical Technology, 52-58.
Roberge, D. M., Zimmermann, B., Rainone, F., Gottsponer, M., Eyholzer, M., & Kockmann, N.
(2008). Microreactor Technology and Continuous Processes in the Fine Chemical and
Pharmaceutical Industry: Is the Revolution Underway. Organic Process Research &
Developemnt (12), 905-910.
Swaminathan, V., & Zega, J. (2005). Continuous Pharmaceutical Processes and Science-Based
Manufacturing. American Association of Pharmaceutical Scientists Annual Meeting and
Exposition.
Van Arnum, P. (2007, April 24). Outlining Implementation Strategies for Lean Manufacturing.
Retrieved from Pharmtech.com: http://pharmtech.findpharma.com/pharmtech
Weiler, A., & Junkers, M. (2009, September 1). Using Microreactors in Chemical Synthesis:
Batch Process versus Continuous Flow. Pharmaceutical Technology.