EFFECTIVE ANALYSIS OF FLEXIBLE COLLABORATION PROCESSES BY WAY OF ABSTRACTION AND MINING TECHNIQUES

A knowledge-based framework for supporting and analyzing loosely-structured collaborative processes (LSCPs) is presented in this paper. The framework takes advantages from a number of knowledge representation, management and processing capabilities, including recent process mining techniques. In order to support the enactment, analysis and optimization of LSCPs in an Internet-worked virtual scenario, we illustrate a flexible integration architecture, coupled with a knowledge representation and discovery environment, and enhanced by ontology-based knowledge processing capabilities. In particular, an approach for restructuring logs of LSCPs is proposed, which allows to effectively analyze LSCPs at varying abstraction levels with process mining techniques (originally devised to analyze well-specified and well structured workflow processes). The capabilities of the proposed framework were experimentally tested on several application contexts. Interesting results that concern the experimental analysis of collaborative manufacturing processes across a distributed CAD platform are shown.