MEASURING INNOVATION POTENTIAL AT SME LEVEL WITH A NEUROFUZZY HYBRID MODEL
MEASURING INNOVATION POTENTIAL AT SME LEVEL WITH A NEUROFUZZY HYBRID MODEL

Summary/Abstract: Measuring innovation has become a crucial issue of today’s economical and political decision makers. In a remarkably short time, economic globalisation has changed the world's economic order, bringing new challenges and opportunities to SMEs. Companies cannot compete in this new environment unless it becomes more innovative and responds more effectively to consumers' needs and preferences – says the EU’s innovation strategy. Decision makers cannot make right and efficient decisions without knowing the capability for innovation of companies of a sector or a region. This need is forcing economists to develop an integrated, unified and complete method of measuring, approximating and even forecast the innovation performance not only on macro level but also on micro level. In this article I intended to show that the recent methods of measuring innovation potential are obsolete, marginally used and have weak statistical performance and effectiveness. Why? Because the world has changed! There are new requirements for social and economical modelling and building expert systems, we have enormous amount of data in a stochastic reality and even the nature of data has been changed. This is especially true in the field of management. Innovation has a so plastic and ductile concept system that it cannot be measured and described (ad absurdum forecasted) by classical crisp methods. It requires soft and intelligent methods. In the article I will show my alternative for measuring innovation potential with a new method which is accurate, strict and significant at the same time, plastic and stable at the same time and simultaneously can handle linguistic variables and blurred (fuzzy) variables. This model possesses efficient studying, adaptive responding, right decision making, information granulation and lingual communication. Via these issues problem solving, pattern recognition, linguistic procession, system design and effective forecasting and estimating can be reached.

• Details
• Contents