Biomedical Engineering Reference
In-Depth Information
TABL E 13.4. Basic Practices for Process Design by Using PAT
Practice
Objectives
Risk assessment
Reduce variability, ensure safety
Continuous improvement
Capture benefits of increased understanding
Process fitness for purpose
Process capability and robustness
Intrinsic performance assessment
Built in monitoring systems, continuous
measurement
Manufacturing strategy
Minimize risk
Data collection and formal
experimental design
Study both simple and complex interactions
Multivariate tools
Develop sophisticated predictive models
Process analyzers
Capture process data
Process control
Based on high-level understanding and
process performance
The second practice captures the dynamic nature of PAT approaches and addresses
continuous improvement. Before starting process development, consideration will be
given to knowledge about product structure and process performance from similar
products and processes. Continued evaluation of design options is an iterative process
that continues throughout the life cycle of the product. The original design space and the
models used to establish process design should be revisited from time to time to see if
they can be improved in the light of new data. Equally, as new analyzers permit fresh
insight into process performance and product quality, the design space should be
modified in light of the new information. The degree to which this can be done will
be dependent on risk and the level of understanding.
Quality and risk assessments should continue to be applied at each step, to ensure
that the third practice, process fitness for purpose, is followed. This ensures that the
process is appropriately designed and has the required process capability to deliver the
desired outputs and relates consistently. It includes product quality, process performance
and process systems, and commercial viability.
The next practice is intrinsic performance assessment, and it differs from traditional
approaches in significant ways. The first is that the process will be designedwith intrinsic
monitors and controls operating in harmony. Measurements will be made in-line or
online rather than by sampling, monitoring will be continuous rather than on the basis of
averages and will be focused on tracking product quality and process performance rather
than meeting acceptable ranges. The process will be considered in its entirety, such that
outputs from one operation are inputs to the next. The focus is on performance and
capability rather than simply following instructions.
The manufacturing strategy should be designed to minimize risk. One aspect of this
is the selection of scale and manufacturing technology, and as described elsewhere, the
risks in transferring from one scale or even one site to another can be minimized if the
process is designed to include scaleable or even scale-independent technologies. The
process is described in dimensionless terms and controlled by process parameters and
quality attributes rather than equipment parameters.
