ESCAPE, Industrial Forum, May 28th 2002

Challenges for Future Use of CAPE tools in Process Synthesis

Dr.-Ing. Gerhard Schembecker

The following (partly provocative) statements are based on the experience gathered by using CAPE tools in about 100 industrial process synthesis projects and by watching academic research and development in this field over the past 10 years.

In my opinion three major issues are currently limiting the use of such tools:

Fundamental physical and chemical insights are most often missing
In all of our projects information about the physical and/or chemical behavior of the components in the process are at least not completely known. However, modern CAPE tools for design and optimization need a precise description of physical properties and kinetics. There is a need for reliable estimation methods which will allow to generate information needed for the application of CAPE tools. Especially in the area of fine chemicals, solids and biochemical processes a lot of research will be needed to reach the current quality of the prediction for fluid components.

Time for optimization is far too long
After some promising alternative flowsheets have been identified usually an optimization phase should follow in order to determine in a very structured and strict way the optimal configuration of the process and the optimal operating parameters of the flowsheet elements. However, even very limited and small tasks take several weeks to solve (an industrial case: the optimization of a multi-phase, multi-stage reactor network with complex kinetics). Major reason is that at present no generic optimization tools are available. Each single optimization problem has to be implemented by building a dedicated new model (incl. physical property estimation and kinetic equations).

Too many me-too projects in academia
Computer based conceptual design seems to be a "modern" research branch in chemical engineering. One can find plenty of recent publications describing new methods, frameworks, generalized methodologies. However, the application of such academic results to realistic processes (and not only HDA!) is very limited. If a researcher wants to see his/her academic results used by the chemical industry one needs more than just a few good ideas. In order to become a competent partner for industry one has to build up a knowledge base over years and years. Too many researchers are obviously not willing to follow this "stony path". This is maybe the reason why only a few academic competence centers for computer aided process synthesis exist worldwide (CMU, DTU, UMASS, UMIST).

this file in MS-Word format