No abstract

Abstract. In this study, we propose a populated iterated greedy algorithm with an Inver-Over operator to solve the traveling salesman problem. The iterated greedy (IG) algorithm is mainly based on the central procedures of destruction and construction. The basic idea behind it is to remove some solution components from a current solution and reconstruct them in the partial solution to obtain the complete solution again. In this paper, we apply this idea in a populated manner (IGP) to the traveling salesman problem (TSP). Since the destruction and construction procedure is computationally expensive, we also propose an iteration jumping to an Inver-Over operator during the search process. We applied the proposed algorithm to the well-known 14 TSP instances from TSPLIB. The computational results show that the proposed algorithm is very competitive to the recent best performing algorithms from the literature.

Quite commonly, faculty performance evaluations use a weighted scheme. Individual faculty members are evaluated on a scale with respect to teaching, research, and service activities. These scores are then combined using pre-determined weights to obtain a combined score that is often used to compare different members. The presented study aimed to investigate the effects of selecting the weights on the individual scores and rankings. The interest is not on single faculty members, but rather on the systems aspects of the practice. That is, how do the weights affect the educational system as a whole? How sensitive is the evaluation system to the selection of the weights? In order to question the leverage a decision maker who determines the weights would have on the outcome of the rankings, the approach based on numerical examples and formal linear programming (LP) considerations is used.

No abstract

No abstract