Biomedical Engineering Reference
In-Depth Information
movement outside the design boundary is considered a change and as such would
most likely require regulatory post-approval.
In order to establish a design space that will allow for maximum process
flexibility while ensuring all CPPs and CQAs are identified and maintained, a large
degree of process understanding is essential. Process analytical technology (PAT)
is a ''pillar/guiding principle'' of the cGMP initiative [
1
]. The PAT framework
published in September 2004 defines process understanding and highlights the
tools required to achieve this standard of process knowledge:
A process is generally considered well understood when (1) all critical sources of vari-
ability are identified and explained; (2) variability is managed by the process; and, (3)
product quality attributes can be accurately and reliably predicted over the design space
established [
3
].
PAT provides in-depth process understanding, but to implement PAT and
operate under the principle of quality by design the process must be well under-
stood. Many in the industry have applied these to processes to glean greater
process knowledge. However, although PAT is a relatively new concept, it has
evolved over the last decade. It has transitioned from being an analysis in the
process, to supplement quality control, to being an analysis of the process [
4
]. As a
result of PAT being embraced by industry, tools that are capable of real-time
monitoring and control must be developed. Currently, few developed tools exist
and even fewer have actually been implemented in a manufacturing environment.
This chapter explores the use of selected PAT tools which can be used in the
context of M
3
C in bioprocess applications and looks at the advantages and limi-
tations of each. Calorimetry is examined in terms of its operating principle and
signal processing methods. A description of the current state and potential future
developments is provided along with a summary of its reported use as a PAT tool.
The history of the development of dielectric spectroscopy for bioprocess purposes
is then described within the scope of PAT, and a detailed overview of the different
applications in the field of process engineering in the production of biological
products is provided. Issues relating to correlation and data pre-processing tech-
niques are also discussed as well as the potential industrial applications of
dielectric spectroscopy. The final set of PAT analysers considered are those based
on vibrational spectroscopy. The theory behind the use of MIR, NIR and Raman
spectroscopy for bioprocessing applications is outlined, and the necessity of using
multivariate data processing is explained. Reported uses of these techniques for
bioprocess monitoring and control applications are summarised, and the current
state of the different technologies are compared.
Finally, a synopsis of available control strategies for bioprocesses based on
measurements from PAT tools and data analysis is given.
Search WWH ::
Custom Search