Discovering Complex System Dynamics with Intelligent Data Retrieval Tools

This paper presents the theoretical foundations of an intelligent online modelling tool capable of processing heterogeneous information on complex techno-economical systems. Its main functionality is to investigate, elicit, and apply rules and principles that govern the development processes of technologies and related markets. Specifically, we will focus on applications of the tool to model the evolution of information technology (IT). We will distinguish several relevant subsystems of the system under study, which describe the demographic, education, global economic trends, as well as specific market factors that determine the demand for and use of IT. The group modelling techniques are implemented in the new tool to enable the collaborative and distributed model building with intelligent verification of entries called ‘model wiki’. Based on the information elicited from experts, gathered from the web and professional databases, a discrete-time control model of technological evolution emerges, coupled with a controlled discreteevent system. The latter processes qualitative information and models the influence of external events and trends on the discrete-time control system parameters. We propose novel uncertainty handling techniques capable of processing and combining different types of uncertain information, coming i.a. from Delphi research and forecasts. The quantitative information is dynamically updated by autonomous webcrawlers, following an adaptive intelligent strategy. The resulting model can be used to simulate long-term future trends and scenarios. Its ultimate goal is to perform an optimization process and derive recommendations for decision makers, for example when selecting IT investment strategies in an innovative enterprise.