Dr. Ciprian Zavoianu

My project deals with enhancing currently available Evolutionary Computation methods employed for the optimization of design parameters used in the development process of electrical drives.

My main task included requirements analysis and system design and development (Java/J2EE) of conceptual solutions for a multi-agent based logistics optimizer to be used by airlines and business jet operators.

My tasks focused on analysis, research and development in the field of Genetic Programming (GP) used for complex data based modeling tasks (symbolic regression). My work with the HEAL team revolved around designing a system that can enhance a GP process in order to enable it to produce parsimonious regression models. The considered modeling scenarios consisted of two, fairly complex, industrial optimization problems proposed by the steel industry.

My main tasks were related to R&D regarding the EU FP7 project DEHEMS. My job was to develop (in Java) a sensor array simulator that would enable a possible future user of DEHEMS to rapidly test and evaluate the system. The simulator was also used to test various storage engine prototypes developed internally by our research team.

During my 4 months university internship with Axxom (Romania) I’ve worked as a part time system analyst on the company's main application, an advanced production & logistics optimization platform named ORion-PI . At first I worked as a bug fixer and, at the end of my internship, I also had the chance to add new features to the application and make changes to other preexisting modules.

machine learning, numeric and metaheuristic optimization, pattern recognition, bioinformatics

parallel and distributed computing, multi-agent systems, advanced topics in artificial intelligence

algorithms, data structures, artificial intelligence, design patterns, programming languages, database modeling, statistics