A methodology for economic evaluation of process technologies in the early research and development stages

A systematic methodology has been developed by the author for building, combining, and exercising a set of specially devised performance, design, and cost models in a form suitable for process economic assessments in the presence of major technological uncertainties. This document describes the development and utilization of the new methodology, i.e., one that incorporates model development and uncertainty analysis for the projection of potentially competitive, full-scale performance and costs of first-of-a-kind process technology still in the early research and development stages. Via simulation, a cohesive spectrum or distribution of the resulting performance and cost figure-of-merit values, along with their associated probabilities, is calculated.

The appropriate format for development of the user's modeling system, which includes the capability to reoptimize the proposed process for each set of process inputs considered is presented, along with the required stepwise approach for selection of values and ranges of the major uncertain process variables or inputs. The basic principles of this combined methodology can be applied to many new processes or technologies—particularly those in their early R&D stages.

Interpretation of the probabilistic output data is also discussed. Such data can be useful to the experimentalists as well as to those decision makers who must recommend or decide whether a particular process should be further developed, or which of several competing technologies should be selected for continued support.

Recent experiences with this methodology in the assessment of advanced uranium isotope separation processes and in assessment of a photochemical syngas cleanup system allow two major conclusions to be drawn. These are (1.) that disappointments in process-performance related areas rather than hardware cost issues tend to have the most deleterious effects on unit cost and (2.) that the process proponent's earliest single-point "best guess" unit cost estimates are usually found to fall in the most optimistic fringes of the computed uncertainty ranges.

A simplified analysis of a catalytic plug flow reactor system and a highly detailed analysis of a proposed photochemical syngas decontamination process are presented to illustrate application of the methodology.