If a decision problem is not multiple criteria one, it is not a real-world problem

Who we are

We are an R&D group originating from Russia and Poland.
The core of the group is two professors of Technical University of Czestochowa, Ludmila Dimova and Pavel Sevastianov .
Our scientific representative in USA is Dmitry Sevastianov.

It must be emphasized that we are developers oriented on problems, not methods. For example, we found out that fancy genetic algorithms are ineffective for solving of portfolio selection optimization problems in a fuzzy setting when compared with other methods for adaptive random search of extremum, which bewilders a lot of young scientists unfamiliar with a history of optimization techniques. However, it does not mean that we not intend to apply genetic algorithms to solve other problems.

Our major scientific directions are modeling, identification and multicriteria optimization under conditions of different nature in applications for economics, technology and ecology.

They include:

• Development of interval and fuzzy mathematics and their synthesis with methods of the possibility theory and the Dempster-Shafer Theory of Evidence
• The synthesis of mathematical statistics methods and fuzzy regression analysis in data processing (processing of real-time market data)
• The problem of local criteria and alternatives' comparative importance estimation on a basis of objective data and subjective opinions
• The problem of aggregation of local criteria in multicriteria, hierarchical and uncertain setting (for example, when assessing investment projects, stocks, possibility of bankruptcy, production quality, commercial contracts, etc.)
• Multicriteria optimization in a fuzzy setting (portfolio problems, logistics, optimization of technological processes quality, quality of cities air under conditions of budget restrictions, etc)
• Simulation modeling of economical and technological processes in a fuzzy setting
• Methods for fuzzy scheduling
• Methods for search of bottlenecks in ramified technological schemes in a fuzzy setting
• Original methods for solving systems of linear and non-linear equations with interval and fuzzy parameters using parallel programming and transputers and their application for macro-economical problems (Leontiev Model)
• Development of the methods of fuzzy logic and their synthesis with the theory of rough sets and their application for problems of technical analysis on stock market
• Software implementation of these methods

Following is systemized list of applications, major part of which is in use and described in scientific papers (available upon request).

We have descriptions of all applications in a form of block schemes and technical documentation. These materials may be adapted and re-used for real-life problem solving.

Currently only scientific and methodological results are published in English.

Applications and software developed

1.  Methods for multicriteria hierarchical estimation of alternatives in uncertain setting
1.  Investment projects estimation
2.  Assessment of product quality
3.  Technical projects valuation
4.  Estimation of commercial contracts
5.  Real estate valuation
6.  Estimation of life quality in different regions of a country
7.  Stocks valuation
2. Methods for multicriteria optimization in fuzzy setting
1.  Multicriteria optimization of product quality (optimization of technological processes)
2.  Fuzzy multicriteria optimization of stock portfolio
3.  Multicriteria optimization of ecological investments aimed to urban population life conditions improvement
4.  Multicriteria optimization of distributor decision-making (transport and trade logistics) in fuzzy and probabilistic setting
5.  Optimization of decision-making on stock market on a basis of synthesis of technical analysis, fuzzy logic and Dempster-Shafer Theory of Evidence
3.  Methods for analysis, modeling and optimization in production management field
1.  Methods for search of bottlenecks in ramified technological networks in interval and fuzzy-interval setting
2. Optimization of enterprise management in interval and fuzzy-interval setting (fuzzy scheduling)
3. Simulation modeling of production processes in probabilistic and fuzzy setting
4. Modeling of economics in a framework of the basic Leontiev model (World Input-Output Model) under conditions of fuzzy matrix of technological parameters (software realization using parallel programming for transputer)

Applied tools

• mathematical statistics
• fuzzy sets theory
• numerical methods for solving systems of equations with partial derivatives describing technological processes
• simulation methods
• methods for search of extremum, etc.

Work style

Immersion in client problems, detailed (sometimes for 3-6 months) examination of technology and organization of production, which often led to complete re-formalization of problems initially proposed by engineers. Further, the problem was solved with a choice of optimal synthesis of methods. Some of these methods we developed ourselves, which was a basis for our scientific careers.

Our history

Our research group (at this moment already international) started in 1976 in former USSR.

We developed information systems on a basis of contracts with major metallurgical, engineering and aeronautics enterprises and research centers. The information systems we built were meant to solve problems in following order:

• modeling
• identification on a basis of laboratory and industrial experiments
• multicriteria optimization of technology and production processes.

In the early nineties we actively used this approach for production quality optimization.

Scientific and applied results of this work were the basis for three books we published in Russian between 1989 and 1993.

In years of industrial crisis and collapse of USSR we switched on problems of medicine, ecology and economics.

Two our systems for solving of ecological and medical problems after Chernobyl disaster were applied on a level of Ministry of Health (Belarus). At the same time we had contracts with the pharmaceutical firms Sanofi and Bayer. We conduct multicriteria drug efficiency estimation for medicines proposed by these firms to solve medical problems after Chernobyl disaster.

Success of our systems was conditioned by optimal synthesis of the methods of mathematical statistics, the methods of fuzzy sets theory and the methods for solving of multicriteria problems.

From 1994 our research group was also engaged in solving economical problems in a fuzzy setting. Original methods of fuzzy budgeting as well as multicriteria evaluation of investment projects were designed and applied for the development of business plans for major enterprises of Belarus on a basis of credit lines provided by the European Bank of Reconstruction and Development .

The systems of multicriteria hierarchical estimation of level of the social economical development of regions were developed and applied for the government of Republic of Belarus.

In 1999 two of our colleagues went in USA. The scientific core of our group consisting of two professors moved in Poland.

Now our Polish research group consists of two Professors, some specialists with PhD, post-graduate students. Some years ago we have founded the chair of Informational Systems and Economic Informatics in Technical university of Czestochowa, where our research group is now located. Our chair is a part of the Institute of Theoretical and Applied Informatics , which possesses the original CLUSTER (the kind of transputer), now consisting of 32 parallel linked processors. This makes it possible for us to overcome some restrictions of common computer systems.

Peculiarity of the scientific work in Poland, when academic results are valued more then applied ones, and the necessity of revival of group under new conditions were the reasons for our recent concentration on methodological problems and manpower training in Poland.

As of today our human resource problems are virtually solved. There are two doctoral dissertations ready for defense, four of our best graduating students of informatics have attended special course under our supervision starting from the second year of study. Each of them has 4-5 publications on English and Polish and they are good specialists able to solve problems from mathematical formalization to software development by themselves.

Our efforts in education allowed us to form a group of young specialists involved in real-life work. There are three contracts with Polish municipalities for elaboration of the system for multicriteria estimation of investment projects proposed in a framework of European Union programs.

Also, there are propositions for cooperation from Russian colleagues. For example, one of them is from the company SeverStal experiencing difficulty with an estimation of reconstruction projects on dozens of their steel plants. There is also an interest in Russia in our methods for multicriteria optimization of economical and financial processes under uncertainty of various nature.