Forecasting Success of Commercialization of Innovative Ideas Using Artificial Neural Networks; The Case of Inventors and Innovators in Yazd Province

Numerous of evidence from around the world show a large number of studies was technically successful, but only a small percentage of these be commercialized. This reflects the complexity of the commercialization process. In this regard, the identification of factors forecasting the probability of successful commercialization of these ideas can help inventors and innovators in the commercialization. Accordingly, This study investigated factors contributing to successful commercialization of inventions in Yazd province, Iran. The variables affecting commercialization of inventions and innovations have been identified and the best Artificial Neural Networks (ANN) model for forecasting the probability of success of the inventions are presented. We used demographic, personal, technological, market, financial and administrative, and legal variables as contributing factors to commercialization. Also, the multilayered perceptron (MLP) with error back propagation (EBP) algorithm with two hidden layers with sigmoid Activation (transfer) function in the hidden layer and the linear in outer layer have the best result in performance measures.