Investment Decision-Making about Portfolio of Technology Development Projects; Based on the Analysis of Success Criteria using Fuzzy Neural Network and MADM

Technology development project is a type of investment project and it is important to identify the performance indicators and planning for the correct investment. The purpose of this research is the development of indicators of portfolio success, accurate analysis of the effects of indicators on each other and the achievement of a proper investment model. In this research, the success criteria of technology development projects with a new approach from the point of view of project, plan and portfolio differentiation have been identified and categorized, their importance has been measured through statistical tests and the relationship between success criteria in each The project, plan and portfolio levels are analyzed using the DEMATEL method. The ANFIS algorithm is used to achieve optimal decision making for investment. For this purpose, investment scenarios based on their criteria and weights are considered as network inputs. Finally, based on 9 samples of 7 technology development projects, using MATLAB software and with the method Grid Partitioning, FCM, Subtractive Clustering The utility function is drawn up. From the results of this research, we can identify and categorize the success criteria of technology development projects and identify the most influential and most important criteria in each of the project, plan, and portfolio groups. Also, the effect and relationship of success metrics with each other are presented through "network mapping" charts and the results of the comparative neuro-fuzzy inference system provide the utility function of investment as a criterion for decision-making in the projects. This methodology can be generalized to the plan and the portfolio.