Browsing by Author "Erdogan, Mehmet Serdar"

Now showing 1 - 4 of 4

An Optimization Model for Vehicle Scheduling and Routing Problem
(Springer Science and Business Media Deutschland GmbH, 2022) Tunay Tokmak; Mehmet Serdar Erdoğan; Yigit Kazancoglu; Erdogan, Mehmet Serdar; Tokmak, Tunay; Kazançoğlu, Yiğit; N.M. Durakbasa , M.G. Gençyılmaz
Vehicle scheduling has a significant impact on the logistic operations of businesses and effective scheduling can increase customer satisfaction. In this context a vehicle scheduling model is developed to enhance the distribution operations of a company. The aim of the developed mixed-integer linear programming model is to minimize the number of vehicles departing in a day in order to decrease the extreme density that the company experiences on certain days. While designing the mathematical model due date constraints have been taken into consideration. However to propose better solutions due dates are expanded one day two days and three days respectively and the model is solved for each case. As due dates extended the number of vehicles departing in a day decreased significantly. The model is solved using IBM ILOG CPLEX (OPL) software for seven days and fifteen days periods by analyzing one month’s data acquired from the company’s database. As longer periods are optimized the model generates better results. However solution time increases. © 2022 Elsevier B.V. All rights reserved.
Citation - WoS: 1
Citation - Scopus: 1
Designing a Railway Network in Cesme Izmir with Bi-objective Ring Star Problem
(SPRINGER-VERLAG SINGAPORE PTE LTD, 2022) Oya Merve Puskul; Dilara Aslan; Ceren Onay; Mehmet Serdar Erdogan; Mehmet Fatih Tasgetiren; Erdogan, Mehmet Serdar; Tasgetiren, Mehmet Fatih; Onay, Ceren; Aslan, Dilara; Puskul, Oya Merve; NM Durakbasa; MG Gencyilmaz
Transportation is a significant subject in today's world especially in terms of the environment and the needs of the community. Clearly high rates of urbanization and population growth result in high volumes of demand for public transportation at the same growing ratio. This project aims to design an optimal railway network to meet the region's public transportation needs and to reduce the region's pollution due to the high seasonal density of the population in the Cesme district. The objective functions of a project are determined by minimizing both assignment cost and routing cost. The assignment cost denotes the total cost of getting on the tram for people. The routing cost is defined as the total construction costs of the tram line's selected nodes. This problem is solved by the epsilon-constraint method as a multi-objective optimization problem. Consequently it has been determined that the two main costs do not decrease at the same time. They are in a correlation where one reduces and the other increases. This is the first study that applies a multi objective ring star problem to a real life case study.
Citation - WoS: 1
Citation - Scopus: 1
Forecasting Damaged Containers with Machine Learning Methods
(Springer Science and Business Media Deutschland GmbH, 2022) Mihra Güler; Onur Adak; Mehmet Serdar Erdoğan; Ozgur Kabadurmus; Güler, Mihra; Adak, Onur; Erdogan, Mehmet Serdar; Kabadurmus, Ozgur; N.M. Durakbasa , M.G. Gençyılmaz
Forecasting the number of damaged containers is crucial for a maritime company to effectively plan future port operations. The purpose of this study is to forecast damaged container entries and exits. In this paper we worked with a global logistics company in Turkey. Comparisons between ports were made using the company’s internal port operations data and externally available data. The external data that we used are Turkey’s GDP exchange rates (USD/EUR) import and export data TEU of Mersin port and the total TEU of Turkey’s ports (2015–2020). Our aim is to forecast the number of damaged containers at a specific port (Mersin Turkey) using different machine learning methods and find the best method. We used Linear Regression Boosted Decision Tree Regression Decision Forest Regression and Artificial Neural Network Regression algorithms. The performances of these methods were evaluated according to various metrics such as R2 MAE RMSE RAE and RSE. According to our results machine learning methods can forecast container demand effectively and the best performing method is Boosted Decision Tree regression. © 2022 Elsevier B.V. All rights reserved.
Vehicle Routing Problem with Multi Depot Heterogeneous Fleet and Multi Period: A Real Case Study
(SPRINGER-VERLAG SINGAPORE PTE LTD, 2022) Baris Karadeniz; Mehmet Serdar Erdogan; Yigit Kazancoglu; Erdogan, Mehmet Serdar; Kazancoglu, Yigit; Karadeniz, Baris; NM Durakbasa; MG Gencyilmaz
In this study we designed a vehicle routing problem model for chocolate manufacturer company to optimize their distribution system. Our model includes 3 warehouses heterogenous fleet multi period and about two dozen customers. We got our customer data directly from company. We considered distance between the customers their demands and vehicle capacities. Sensitivity analysis was made to make sure that it is effective. In the model that we designed we compared 10-15-20 customer cases with 3 depot 1 depot heterogeneous and homogeneous fleet scenarios. According to results of the study homogeneous fleet scenarios are costly. In addition using single depot is costly than using multiple depots.