Page 61 - Design of Simple and Robust Process Plants

3.3 Design Philosophies 45
. All critical equipment from a reliability perspective need to be ordered with a
RAM specification next to its duty specification.
. The single component philosophy is achievable, and brings about high reduc-
tions in both cost and capital.

3.3.5
Optimize Design

Until now, the optimization of process design has been the result of evolution, as
engineers have been stimulated to improve their designs through external circum-
stances. Examples of this include the energy crisis, increasing environmental
requirements, product quality requirements, and market competition. More
recently, the increasing focus to achieve a sustainable world has been added to this
list. A good illustration is the process flowsheets of a batch process at different
stages of development (see Chapter 1, Figures 1.4±1.7).
The evolution of a process plant is clearly illustrated in that example. The evolu-
tion of plant designs will continue, particularly in the field of reactor technology
based on the design of catalysts. The demands on environmental requirements will
become even more severe, the emphasis coming from the drive for a sustainable
world ± in other words, higher selectivities and lower utility consumption will be
foremost. In future, communities will need to be convinced by industry of the appli-
cation of sustainable technology. The global market conditions will put pressure on
low cost, resulting in further optimization of the design and its operation, and this
in turn will lead to higher levels of integration between processes.
Although for new designs, careful evaluation of available designs and the evalua-
tion of process synthesis options will become major requirements, the application
of the current process synthesis tools are limited by:
. The tools are not yet sufficiently mature.
. The data required as input are of limited availability, notably for the evalua-
tion of different reaction/separation techniques.
. Limited applicability to existing situations ± bear in mind that most plants
have lifetimes of 30±40 years and are subject to major modification every 10
years.
. Industry is not willing to take large risks related to new technology/designs
(conservatism).

Optimization of the design will provide considerable improvements in efficiency
(Douglas, 1988; Biegler et al., 1997; Koolen et al., 1999; Seider et al., 1999). In indus-
try, the current practice of optimization is restricted to a comparison of selected
cases by the engineers; this situation must change, and is supported by ongoing
research in this field.
Research to solve the conceptual design as one discrete optimization problem will
advance, and indeed the modeling environment is evolving in that direction. The