Browsing by Author "Ozcan, Sel"

Now showing 1 - 4 of 4

Citation - WoS: 8
Citation - Scopus: 9
A differential evolution algorithm with a variable neighborhood search for constrained function optimization
(Springer Verlag service@springer.de, 2015) M. Fatih Tasgetiren; Ponnuthurai Nagaratnam Suganthan; Sel Ozcan; Damla Kizilay; Tasgetiren, M. Fatih; Suganthan, P.N.; Kizilay, Damla; Ozcan, Sel
In this paper a differential evolution algorithm based on a variable neighborhood search algorithm (DE_VNS) is proposed in order to solve the constrained real-parameter optimization problems. The performance of DE algorithm depends on the mutation strategies crossover operators and control parameters. As a result a DE_VNS algorithm that can employ multiple mutation operators in its VNS loops is proposed in order to further enhance the solution quality. We also present an idea of injecting some good dimensional values to the trial individual through the injection procedure. In addition we also present a diversification procedure that is based on the inversion of the target individuals and injection of some good dimensional values from promising areas in the population by tournament selection. The computational results show that the simple DE_VNS algorithm was very competitive to some of the best performing algorithms from the literature. © 2015 Elsevier B.V. All rights reserved.
Citation - WoS: 9
Citation - Scopus: 9
A reward-based algorithm for the stacking of outbound containers
(Elsevier B.V., 2017) Sel Ozcan; D. T. Eliiyi; Ozcan, Sel; Eliiyi, Deniz Türsel
As global trade increases container transshipment activities increase rapidly. Therefore high competition among container terminals has emerged. The decision for good stacking positions for containers plays a critical role on the performance of a container terminal since it influences the productivity of the terminal in a strong sense. In this study we focus on the minimization of the berthing time of vessels by improving the stacking operations through minimization of non-value-added handling operations of containers which is called reshuffling as well as the traveling time of the cranes operating at the storage yard. Most of the containers in the container terminal we focus are outbound which are transported into the container terminal via external trucks and stored in the stacking yard until they are loaded onto vessels. We propose a reward-based algorithm for the stacking of outbound containers by taking the following four components into consideration, container's distance to the closest RTGC RTGC's workload the number of stacked containers at the neighborhood bays and the current height of the stacks at the storage yard. The inputs of the algorithm include the relevant information of the container to be stacked that are just entered into the terminal gate current usage information of the storage yard and the current positions of the yard cranes. The proposed stacking strategy is in implementation phase to one of the container terminals in Izmir. The results seem to be promising when compared to the current randomized stacking strategy in the container terminal. © 2017 Elsevier B.V. All rights reserved.
Citation - WoS: 2
Citation - Scopus: 2
An alternative MILP model for makespan minimization on assembly lines
(SPRINGER HEIDELBERG, 2017) Sel Ozcan; Deniz Tursel Eliiyi; Levent Kandiller; Kandiller, Levent; Ozcan, Sel; Eliiyi, Deniz Türsel
The Simple Assembly Line Balancing Problem-2 (SABLP-2) is defined as partitioning the tasks among stations in order to minimize the cycle time given the number of stations. SALBP-2 reduces to the identical parallel machine scheduling problem with makespan minimization (P-m parallel to C-max) when precedence relations are ignored providing a lower bound. In a certain layout setting tasks revisiting the same station over consecutive tours might be preferable when the sole objective is to minimize the makespan of producing the order quantity. In this study the tradeoff between the makespans obtained from SALBP-2 and (P-m parallel to C-max) as a function of order quantity is analyzed. A piecewise linear concave makespan function is observed. We developed an alternative model formulation and an iterative solution scheme for makespan minimization for all possible order quantities. The results of our computational experiment indicate that SALBP-2 outperforms for small order quantities whereas (P-m parallel to C-max) yields the best results for larger order quantities. However there is a certain range of order quantity for which the proposed model dominates the other two. Our results are validated in benchmark instances.
Citation - WoS: 10
Citation - Scopus: 17
Optimization of university course scheduling problem with a hybrid artificial bee colony algorithm
(IEEE, 2011) Adalet Oner; Sel Ozcan; Derya Dengi; Dengi, Derya; Oner, Adalet; Ozcan, Sel
Course scheduling problem (CSP) is concerned with developing a timetable that illustrates a number of courses assigned to the classrooms. In this study a hybrid algorithm composed of a heuristic graph node coloring (GNC) algorithm and artificial bee colony (ABC) algorithm is proposed to solve CSP. The study is one of the few applications of ABC on discrete optimization problems and to our best knowledge it is the first application on CSP. A basic heuristic algorithm of node coloring problem takes part initially to develop some feasible solutions of CSP. Those feasible solutions correspond to the food sources in ABC algorithm. The ABC is then is used to improve the feasible solutions. The employed and onlooker bees are directed or controlled in a specific manner in order to avoid the conflicts in the course timetable. Proposed solution procedure is tested using real data from a university in Turkey. The experimental results demonstrate that the proposed hybrid algorithm yields efficient solutions. © 2011 IEEE. © 2011 Elsevier B.V. All rights reserved.