Dariusz Wagner

Computer-assisted constrained approval voting

Approval voting, proposed independently by several analysts in the 1970s, is a voting system in which voters can vote for as many candidates as they like in multicandidate elections. Recently, S.J. Brams modified this system, introducing so-called constrained approval voting. It is designed for a professional association to ensure equitable representation of different interests. In his new system approval voting is combined with the constraints on the number of persons that can be elected from different categories of members. In the present paper the problem of constrained approval voting is formulated (following the work of R.F. Potthoff) as an integer programming problem. Some computational aspects of this problem are discussed. The paper presents a numerical example illustrating a possibility to apply the discussed voting procedure in the election of members of the Committee for Organization and Management Sciences of the Polish Academy of Sciences.

Forecasting voting behaviour using machine learning – Poland in transition

The aim of the paper is to apply some inductive learning method from examples (which gives explicit decision rules of “if-then” type) to forecast the voting behaviour of individual members of the Polish Parliament. Results obtained are both interesting and promising.

Identification of Ideological Dimensions under Fuzziness: The Case of Poland

Under the new political conditions (parliamentary democracy) the results of elections to the parliament have to be considered as an unprecise representation of the real political situation. Consensus reached under such conditions is not stable.

Interactive system for negotiations on allocation of research funds

Abstract The-paper is concerned with an interactive system making it possible to apply some auction procedures for the purpose of assigning research teams to given projects necessary for the realization of a research program. At each stage of this procedure the head of every research team has to make a decision on the assignment of his workers. It is assumed that the strategy of research teams heads consists in maximization of the average time-discounted income per worker. The proposed system of interactions among the research teams heads provides a possibility of reaching the consensus in the matter of assignment of research workers to particular projects. Simultanously, it makes possible to solve the problem of research funds allocation. Such a system can be considered as a multiperson game of Nash type with the constant sum of the players payments. The presented interactive procedure makes use of some results obtained in the course of experiments in making social choice.

Application of Kemeny’s Median for Group Decision Support

Some general problems concerned with the choice of a method of group judgement for expert opinions given in the form of preference orders are discussed. A special attention is paid to Kemeny’s median and its modification proposed by B.G. Litvak, because of their important properties. Basic notions related to those methods are given. Heuristic algorithms for computing a median for both cases under consideration are proposed. Numerical examples illustrating the application of these algorithms are presented. A real-life example is also described.

Determining Weights of Research Topics on the Basis of Expert Judgements. The Case of Systems Research Institute

In the paper a problem of determining weights of research topics on the basis of expert judgements is considered. Such weights can be used to facilitate the process of fund allocation.

Allocation of research funds in competitive environment. A computerized negotiation system

The paper is concerned with the problem of financing of complex research programs. One of tasks to be solved consists in assigning research teams, willing to participate in a given program, to research projects being its elements, under conditions of constrained budget. It is assumed that the strategy of every research team head is to maximize the average time-discounted income per person. In the previous paper of the authors a special negotiation procedure has been proposed to solve this problem. This paper presents some possible extensions and modifications of the procedure. At each stage of this procedure the heads of research teams involved have to make decisions on the assignment of their workers to particular projects. The proposed system of interactions among the research teams heads provides a possibility of reaching the consensus in the matter of this assignment. Simultaneously, it makes possible to solve the problem of research funds allocation. Such a system is considered as a multiperson game of Nash type with the non-zero sum of the players payments.