Browsing by Author "Yüksel, Damla"

Now showing 1 - 12 of 12

Citation - WoS: 35
Citation - Scopus: 25
A circular food supply chain network model to reduce food waste
(Springer, 2022) Ozgur Kabadurmus; Yigit Kazancoglu; Damla Yüksel; Melisa Ozbiltekin-Pala; Yüksel, Damla; Pala, Melisa Özbiltekin; Kabadurmus, Ozgur; Kazançoğlu, Yiğit
Food loss and waste (FLW) is a growing global problem throughout the world. The rapid increase in food waste and deficiencies in treatment processes have led to greater harm to the environment. A circular food supply chain (FSC) is now an essential means of encouraging circular economy. Proper food waste treatment and recycling operations can not only benefit the environment but these wastes can also be used as raw material for production in a circular economy. In this study a circular food supply chain network model is designed to reduce the food waste generated in the circular food supply chain systems of municipalities. Then a mixed-integer linear programming model is generated to model the proposed circular food supply chain network model. The MILP model is a network model aimed at reducing the food waste generated. To do so two objectives are considered: the overall cost of the network is minimized and the amount of distributed food waste from the generation nodes to the end nodes is maximized. Due to the bi-objective nature of the proposed mathematical model the Improved Augmented Epsilon Constraint method (AUGMECON2) is implemented to solve the problem optimally. To illustrate the applicability and effectiveness of the proposed mathematical model two real-life case studies were carried out in Izmir the third largest city in Turkey. The computational results demonstrate that the proposed model is beneficial for both small and large municipalities since it provides the Pareto-optimal set where the total amount of distributed food waste is maximized and the total cost is minimized. © 2022 Elsevier B.V. All rights reserved.
Citation - WoS: 43
Citation - Scopus: 53
An energy-efficient bi-objective no-wait permutation flowshop scheduling problem to minimize total tardiness and total energy consumption
(PERGAMON-ELSEVIER SCIENCE LTD, 2020) Damla Yuksel; M. Fatih Tasgetiren; Levent Kandiller; Liang Gao; Yüksel, Damla; Taşgetiren, M. Fatih; Gao, Liang; Kandiller, Levent
In manufacturing scheduling sustainability concerns that raise from the service-oriented performance criteria have seldom been studied in the literature. This study aims to fill this gap in the literature by integrating the different energy consumption levels at the operational level. Since energy-efficient scheduling ideas have recently been increasing its popularity in industry due to the need for sustainable production this study will be a good resource for future energy-efficient scheduling problems. Energy consumption in high volume manufacturing is a significant cost item in most industries. Potential energy saving mechanisms are needed to be integrated into manufacturing facilities for cost minimization at the operational level. A leading energy-saving mechanism in manufacturing is to be able to adapt/change the machine speed levels which exactly determines the energy consumption of the machines. Hence in this study the afore-mentioned framework is applied to the no-wait permutation flowshop scheduling problem (NWPFSP) which is a variant of classical permutation flowshop scheduling problems. However it has various critical applications in industries such as chemical pharmaceutical food-processing etc. This study proposes both mixed-integer linear programming (MILP) and constraint programming (CP) model formulations for the energy-efficient bi-objective no-wait permutation flowshop scheduling problems (NWPFSPs) considering the total tardiness and the total energy consumption minimization simultaneously. This problem treats total energy consumption as a second objective. Thus the trade-off between the total tardiness - a service level measurement indicator - and the total energy consumption - a sustainability level indicator - is analyzed in this study. Furthermore due to the NP-hardness nature of the first objective of the problem a novel multi-objective discrete artificial bee colony algorithm (MO-DABC) a traditional multi-objective genetic algorithm (MO-GA) and a variant of multi-objective genetic algorithm with a local search (MO-GALS) are proposed for the bi-objective no-wait permutation flowshop scheduling problem. Besides the proposed algorithms are compared with the multi-objective energy-efficient algorithms from the literature. Consequently a comprehensive comparative metaheuristic analysis is carried out. The computational results indicate that the proposed MO-DABC algorithm outperforms MILP CP MO-GA MO-GALS and algorithms from the literature in terms of both cardinality and quality of the solutions. The powerful results of this study show that the proposed models and algorithms can be adapted to other energy-efficient scheduling problems such as no-idle flowshop blocking flowshop and job-shop scheduling problems or to other higher-level integrated manufacturing problems.
Beklemesiz permütasyon akış tipi çizelgeleme problemleri için yeni çözüm teknikleri
(2024) Yüksel, Damla; Kandiller, Levent
No-Wait Permutation Flowshop Scheduling Problem (NWPFSP) is a scheduling problem variant where jobs must proceed through machines in a fixed order without waiting times between operations. This thesis explores innovative solution techniques for the NWPFSPs. The primary contributions of this thesis are twofold: single-objective optimization and bi-criteria optimization. For single-objective optimization, this thesis examines five mathematical model formulations — three using Mixed-Integer Linear Programming (MILP) and two using Constraint Programming (CP) — focused on separately minimizing makespan, total flow time, and total tardiness. One MILP model is enhanced with valid inequalities to address these objectives. A new Lower Bound (LB) mechanism based on the Shortest Path (SP) algorithm is developed to optimize makespan, total flow time, total tardiness, and the number of tardy jobs separately. Following that, two mathematical models, one belonging to the MILP class and the other to the CP class, have been studied for the number of tardy job minimization in NWPFSPs. A novel upper bound, the Sacrifice and Rearrange Heuristic (SRH), is introduced to minimize the number of tardy jobs. Optimizing the number of tardy jobs in NWPFSPs requires high-quality due dates, as they are crucial for improving performance metrics related to lateness. A new mechanism for generating high-quality due dates has been developed to address this. Incorporating the Sacrifice and Rearrange Heuristic (SRH), this mechanism ensures practical and effective due dates. For bi-criteria optimization, the NWPFSP is approached as a combinatorial optimization problem with two objectives, aiming to minimize total flow time and makespan simultaneously: Bi-Criteria No-Wait Permutation Flowshop Scheduling Problems (BI-CRI NWPFSPs). Initially, an MILP model formulation is explored to address BI-CRI NWPFSPs. Following this, Q-learning-guided algorithms are developed for Bi-CRI NWPFSPs. Q-learning, a well-known reinforcement learning technique, is employed to direct action selection, thereby reducing the need for random exploration during the iterative metaheuristic process. The developed Q-learning guided metaheuristics are Bi-Criteria Iterated Greedy Algorithm with Q-Learning (BC-IGQL) and Bi-Criteria Block Insertion Heuristic Algorithm with Q-Learning (BC-BIHQL). The performance of these algorithms is compared with other state-of-the-art approaches. Thus, this thesis advances the literature on the NWPFSPs by developing new solution techniques for both single-objective and bi-criteria scenarios.
Drought Modelling Using Artificial Intelligence Algorithms in Izmir District
(SPRINGER-VERLAG SINGAPORE PTE LTD, 2022) Zeynep Irem Ozen; Berk Sadettin Tengerlek; Damla Yuksel; Efthymia Staiou; Mir Jafar Sadegh Safari; Yüksel, Damla; Safari, Mir Jafar Sadegh; Özen, Zeynep İrem; Tengerlek, Berk Sadettin; Staiou, Efthymia; NM Durakbasa; MG Gencyilmaz
The world's water resources are decreasing day by day due to factors such as climate change drought inefficient pricing policies implemented by the government population growth uncontrolled water consumption technological developments and industrialization. A decrease in water resources causes water scarcity in the long-term period. This study is conducted to analysis the meteorological drought in Izmir district Turkey. Inspired by the real-life problem drought estimation models are developed through artificial neural network-based artificial intelligence techniques incorporating a decision support system. The Z-score index (ZSI) values are computed using precipitation data collected from five meteorological station in Kucuk Menderes basin and several developed models are compared according to the variety of statistical performance metrics.
İki amaçlı beklemesiz permutasyon akış tipi çizelgeleme problemleri
(2019) Yüksel, Damla; Kandiller, Levent; Taşgetiren, Mehmet Fatih
In the field of permutation flowshop scheduling problems, there is a vast literature covering mathematical models and heuristics approaches. However, less work has been reported in the field of no-wait permutation flowshop scheduling problems, a variant of permutation flow shop scheduling problem where the waiting time for the jobs between the machines is not allowed. This thesis proposes both mixed-integer linear programming and constraint programming model formulations for no-wait permutation flowshop scheduling problem under various objectives such as (i) makespan, (ii) total flow time and (iii) total tardiness. Moreover, energy-efficient scheduling has become very popular recently since energy consumption in high volume manufacturing is the leading essential difficulty in most industries. Both mixed-integer programming and constraint programming model formulations are developed in this thesis on the energy-efficient (bi-objective) no-wait permutation flowshop scheduling problems with the objective of minimizing (i) makespan, (ii) total flow time and (iii) total tardiness, separately. The bi-objective no-wait permutation flowshop scheduling problems treat the total energy consumption as a second objective in this study. Furthermore, due to the NP-hardness nature of the first objective of the problem, a novel multi-objective discrete artificial bee colony algorithm (MO-DABC), a traditional multi-objective genetic algorithm (MO-GA) and a variant of multi-objective genetic algorithm (MO-GALS) are proposed for the bi-objective no-wait permutation flowshop scheduling problems. Consequently, a comprehensive comparative metaheuristic analysis is carried out. Hence, this thesis contributes to the literature of no-wait permutation flowshop scheduling problem for not only single-objective problems but also the bi-objective problems which consider energy efficient scheduling by ensuring various new mathematical models and metaheuristics.
Citation - Scopus: 1
Mathematical Models for No-Wait Permutation Flowshop Scheduling Problems
(Springer Science and Business Media Deutschland GmbH, 2023) Damla Yüksel; Levent Kandiller; M. Fatih Tasgetiren; Yüksel, Damla; Taşgetiren, Mehmet Fatih; Kandiller, Levent; N.M. Durakbasa , M.G. Gençyılmaz
In today's complex manufacturing industry no-wait permutation flowshop scheduling problems are one of the most encountered types of scheduling problems. The floor configuration of this type is the one with the restriction over the jobs that cannot wait between successive machines. The problem can be modeled from different points of view on the floor configuration. Hence in this study three mixed-integer programming models and two constraint programming models are studied on the no-wait permutation flowshop scheduling problems for the three objective functions: makespan total flow time and total tardiness. Among five mathematical models two are newly proposed to the literature on no-wait flowshop scheduling problems and three are previously studied for the variants of the no-wait flowshop scheduling problems. Complete experimentation is accomplished on the well-known benchmark set of Taillard. Regarding the computational experiments Model 4 performs best regarding the objective function value and the gap percentage reported in makespan and total flow time minimization. However for total tardiness although Model 3 performs best in terms of the gap percentage reported Model 4 still performs best in terms of the objective value reported. © 2023 Elsevier B.V. All rights reserved.
Citation - WoS: 2
Citation - Scopus: 2
Parallel Workforce Assignment Problem for Battery Production
(SPRINGER-VERLAG SINGAPORE PTE LTD, 2022) Aslihan Erdosan; Aysenur Tali; Berfu Kircali; Buket Gunal; Ecem Nazli Dedecengiz; Damla Yuksel; Banu Yetkin Ekren; Yüksel, Damla; Erdosan, Aslihan; Günal, Buket; Dedecengiz, Ecem Nazlı; Ekren, Banu Yetkin; Erdoğan, Aslıhan; Talı, Ayşenur; Kırcalı, Berfu; NM Durakbasa; MG Gencyilmaz
This paper studies workforce assignment problem for battery production in a company in Turkey. Several types of batteries are produced in the studied company. Mostly the operations are semi-automated. In the production process the workers are assigned to multiple operations irregularly based on the priority of productions. In the company average utilization of worker is low and average cycle time of a product is high due to inefficient allocation of the workforce within the operations. In order to analyze the main system problem we simulate the system and observe the queue lengths to identify the bottlenecks. By dynamic assignment of workers at stations based on real time queue conditions the workloads can be balanced throughout the production lines. In this project a simulation-based system improvement is completed by applying: (i) dynamic utilization of workforce to reduce average cycle time of a battery (ii) assignment of parallel workforce where they can work for the same operation simultaneously and (iii) observation of real-time queue lengths of stations. Three dynamic assignment policies are developed and compared with each other. The best policy providing minimum cycle time for a battery production is selected to be the best.
Citation - WoS: 15
Citation - Scopus: 14
Q-learning guided algorithms for bi-criteria minimization of total flow time and makespan in no-wait permutation flowshops
(ELSEVIER, 2024) Damla Yuksel; Levent Kandiller; Mehmet Fatih Tasgetiren; Yüksel, Damla; Taşgetiren, Mehmet Fatih; Kandiller, Levent
Combining Deep Reinforcement Learning and meta-heuristic techniques represents a new research direction for enhancing the search capabilities of meta-heuristic methods in the context of production scheduling. Q-learning is a prominent reinforcement learning in which its utilization aims to direct the selection of actions thus preventing the necessity for a random exploration in the iterative process of the metaheuristics. In this study we provide Q-learning guided algorithms for the Bi-Criteria No-Wait Flowshop Scheduling Problem (NWFSP). The problem is treated as a bi-criteria combinatorial optimization problem where total flow time and makespan are optimized simultaneously. Firstly a deterministic mixed-integer linear programming (MILP) model is provided. Then Q-learning guided algorithms are developed: Bi-Criteria Iterated Greedy Algorithm with Q-Learning (BCIGQL). Bi-Criteria Block Insertion Heuristic Algorithm with Q-Learning (BC-BIHQL). Moreover the performance of the proposed Q-learning guided algorithms is compared over a collection of Bi-Criteria Genetic Local Search Algorithms (BC-GLS) Bi-Criteria Iterated Greedy Algorithm (BC-IG) Bi-Criteria Iterated Greedy Algorithm with a Local Search (BC-IGALL) and Bi-Criteria Variable Block Insertion Heuristic Algorithm (BC-VBIH). The complete computational experiment performed on 480 problem instances of Vallada et al. (2015) which is known as the VRF benchmark set indicates that the BC-BIHQL and the BC-IGQL algorithms outperform the BC-GLS BC-IG BCIGALL and BC-VBIH algorithms in comparative performance metrics. More specifically the proposed BC-BIHQL and BC-IGQL algorithms can yield more non-dominated bi-criteria solutions with the most substantial competitiveness than the remaining algorithms. At the same time both are competitive with each other on the benchmark problems. Moreover the BC-IGQL algorithm dominates almost 97% and 99% of the solutions reached by the BC-IG BC-IGALL and BC-VBIH algorithms in small and large datasets. Similarly The BC-BIHQL algorithm dominates almost 98% and 99% of the solutions reached by the BC-IG BC-IGALL and BC-VBIH algorithms in small and large datasets respectively. This means that among all the features that have been compared the Qlearning-guided algorithms demonstrate the highest level of competitiveness. The outcomes of this study encourage us to discover many more bi-criteria NWFSPs to reveal the trade-off between other conflicting objectives such as makespan & the number of early jobs to overcome various industries' problems.
Citation - Scopus: 1
Standardization of Packaging Materials for Various Products and Cost Optimization in Packaging
(Springer Science and Business Media Deutschland GmbH, 2020) Özge Ceyhan; Merve Türk; Hafize Helvacı; Damla Yüksel; Nejat Kutup; Yüksel, Damla; Türk, Merve; Ceyhan, Özge; Helvacı, Hafize; Kutup, Nejat; M.N. Osman Zahid , R. Abd. Aziz , A.R. Yusoff , N. Mat Yahya , F. Abdul Aziz , M. Yazid Abu , N.M. Durakbasa , M.G. Gençyilmaz
This project is related to the standardization of the packaging of variable-size products which belongs to base headboard marquise and cupboard (nightstand) product categories for a bed products company. The company cares importance to special size demands of their customers and produces products in a wide range of sizes. This causes complexity in the stock-tracking of cardboard plates which are the most important packaging materials in the packaging process and high waste of cardboards. Therefore this study aims to improve the packaging system of the company and to reduce the packaging costs by decreasing the amount of waste related to the materials used for packaging. The main objective of the study is to create a standardized packaging plan by considering all product sizes the company has produced in the last three years in the product categories of base headboard marquise and cupboard. To achieve this goal after the completion of literature search forecasts for the next year’s sales were done and a mathematical model that has two conflicting objectives was formulated. Then the mathematical model was solved by using IBM ILOG CPLEX program for each product category and optimal package sizes were determined. A useful Decision Support System (DSS) which enables to prepare weekly packaging plans that reduce the extra packaging costs and the amount of waste has been created at the end of the project. © 2022 Elsevier B.V. All rights reserved.
Süt Sevkiyat Problemi için Matematiksel Modeller
(2021) Sinem Özkan; Damla Kizilay; Hande Öztop; DAMLA YÜKSEL; Yüksel, Damla; Kizilay, Damla; Özkan, Sinem; Öztop, Hande
Bu çalışma küçük ölçekli bir dağıtım şirketi için şişelenmiş süt sevkiyat problemini ele almaktadır. Şişelenmiş süt sevkiyat problemi ile birçok gerçek\rhayat uygulamasında karşılaşabilmektedir. Sosyal sorumluluk projesi kapsamında şirketler ve belediyeler dahil olmak üzere birçok organizasyon \rçocuklar için ilkokullara ve yoksul ailelere ücretsiz olarak süt dağıtmaktadır. Bu şirketler genellikle belli bir kapasiteye sahip araçlar kullanarak ve\rmüşterilerin (okullar ve aileler) uygun oldukları saat aralıklarını dikkate alarak dağıtım yapmaktadır. Planlama sürecinde genellikle akaryakıt ve\rdepolama maliyetleri gibi masraflar en aza indirilmek istenmektedir. Bu kısıtlar altında problemin teslim zaman aralığı ve araç kapasitesi kısıtlı\raraç rotalama problemi (CVRPTW) olduğu ortaya çıkmaktadır. Temel hedeflerden biri araçların yakıt tüketimini azaltmak için araç türünü dikkate\ralarak toplam seyahat mesafesini en aza indirmektir. Diğer bir amaç ise zamanında teslim edilemeyen sütlerin depolama maliyetini azaltmak için \rmüşterilere uygun zaman aralıklarına göre hizmet vermek ve geç ürün teslimlerini azaltmaktır. Bu hedeflere ulaşmak amacıyla probleme özgü\rkarmaşık tam sayılı doğrusal programlama (MILP) ve kısıt programlama (CP) modelleri geliştirilmiştir. Matematiksel modelleri doğrulamak ve\rkarşılaştırmak için literatürde oldukça bilinen veri setleri üzerinde probleme özgü parametreler dikkate alınarak bazı değişiklikler yapılmıştır. Detaylı\ranalizlere ve sonuçlara göre her iki modelin de çalışılan problem için oldukça rekabetçi olduğu gözlemlenmiştir. Ancak uzun planlama vadesine\rsahip veri örnekleri için MILP modelinin CP modelinden çözüm kalitesi ve çözüm süresi açısından daha iyi performans gösterdiği görülmüştür.
Citation - Scopus: 1
The Modification and the Line Balancing of an Assembly Line for the Inclusion of a New Product Model
(Springer Science and Business Media Deutschland GmbH, 2020) Dilruba Telli; Yalcin Berberoglu; Gizem Karadeniz; Doğangün Yılmaz Uzun; Elif Baykal; Damla Yüksel; Yigit Kazancoglu; Yüksel, Damla; Karadeniz, Gizem; Telli, Dilruba; Berberoğlu, Yalçın; Kazançoğlu, Yiğit; Uzun, Doğangün Yılmaz; Baykal, Elif; N.M. Durakbasa , M.N. Osman Zahid , R. Abd. Aziz , A.R. Yusoff , N. Mat Yahya , F. Abdul Aziz , M. Yazid Abu , M.G. Gençyilmaz
This study aims to modify the 5th lean assembly line in order to include a new product model in a combi boiler producer company. After reviewing customers’ needs company decides to produce a new product. New product model will be produced in the one of the existing lines in the plant. Most suitable line determined according to other product models job sequences’ similarity with new product models’. After the analysis of the production process the changes and possible constraints for the 5th assembly line are determined. After that solution procedures are developed the required calculations are made and implemented. The solutions are developed by various techniques and approaches such as Bubble diagram Value Stream Mapping Lean Production Line Balancing. Particularly mathematical model which is solved by IBM ILOG Optimization Studio programmed comprehensive analysis with a simulation model which is conducted by Arena Simulation software are presented in this study. The outcomes of the solutions are simulated compared and automatized. At the end three options are provided for the 5th lean line to include a new product type while continuing the production process with the current products. © 2022 Elsevier B.V. All rights reserved.
Water Resource Management Using a Multiperiod Water Pricing Model in Izmir District
(SPRINGER-VERLAG SINGAPORE PTE LTD, 2022) Zeynep Irem Ozen; Berk Sadettin Tengerlek; Damla Yuksel; Efthymia Staiou; Levent Kandiller; Yüksel, Damla; Özen, Zeynep İrem; Tengerlek, Berk Sadettin; Staiou, Efthymia; Kandiller, Levent; NM Durakbasa; MG Gencyilmaz
Water is a product that cannot be substituted for living creatures to survive. The availability of water resources is crucial. The balance of water availability can change with environmental factors and as a result living creatures struggle with water scarcity. This study includes the policy of water pricing in the Izmir district. A multiperiod water pricing model based on transportation and inventory carrying problems is developed for encouraging water consumption reduction. Different pricing policy scenarios are proposed using the developed model to provide sustainable water management and penalize high water use. The aim is to maximize consumer welfare and ensure fair distribution among groups.