Study on the Method of the Technology Forecasting Based on Conjoint Analysis

Abstract

We discuss an application of conjoint analysis in technology forecasting, summarize basic operation steps of conjoint analysis, and give a stimulant example of technology forecasting. In this example, we consider five factors that will affect the emergence of a new technology. These factors have investing demand in a new technology, potential market value of a new technology, realizable difficulty of a new technology, supporting degree of relative technology to a new technology, and the competitive power of a new technology with original technology. Technology development has a discontinuity. With discontinuity, we cannot forecast the future of technology development, based on the current trend of technology development. As using quantitative methods to make forecasting, we assumed that current trends of technology development hold a fixed law, so those quantitative methods cannot forecast the discontinuity of technology development. Some subjective forecasting methods have huge improvement in technological discontinuity forecasting. The improvement is that forecaster’s subjective judgments and capability are embodied in forecasting. But this method has two inherent defects: one is the lack of design ability, which makes this method susceptible to the influence of organizer and forecasting, and the other is that while facing numerous forecasters, the forecasting data are often difficult to explain and analyze; we also have difficultly in making a synthesized judgment. A subjective and synthesized judgment of technology development is similar to economical utility, thus we could apply the measure of colony’s utility to improve the appropriateness and reliability of subjective forecasting method. Using conjoint analysis, we can judge the colony’s utility accurately, because the datum that we use in analysis comes from the subjective judgments of forecasters to various fields of technical development, but the influence of the random error can be dispelled by using some theory model about data processing. Therefore, conjoint analysis method is one useful tool for technology forecasting. The effectiveness of this method has been testified by simulation experiment.