Browsing by Author "Arslan, Bartu"

Now showing 1 - 8 of 8

Citation - WoS: 27
Citation - Scopus: 27
A reinforcement learning approach for transaction scheduling in a shuttle-based storage and retrieval system
(WILEY, 2024) Banu Y. Ekren; Bartu Arslan; Arslan, Bartu; Ekren, Banu Y.
With recent Industry 4.0 developments companies tend to automate their industries. Warehousing companies also take part in this trend. A shuttle-based storage and retrieval system (SBS/RS) is an automated storage and retrieval system technology experiencing recent drastic market growth. This technology is mostly utilized in large distribution centers processing mini-loads. With the recent increase in e-commerce practices fast delivery requirements with low volume orders have increased. SBS/RS provides ultrahigh-speed load handling due to having an excess amount of shuttles in the system. However not only the physical design of an automated warehousing technology but also the design of operational system policies would help with fast handling targets. In this work in an effort to increase the performance of an SBS/RS we apply a machine learning (ML) (i.e. Q-learning) approach on a newly proposed tier-to-tier SBS/RS design redesigned from a traditional tier-captive SBS/RS. The novelty of this paper is twofold: First we propose a novel SBS/RS design where shuttles can travel between tiers in the system, second due to the complexity of operation of shuttles in that newly proposed design we implement an ML-based algorithm for transaction selection in that system. The ML-based solution is compared with traditional scheduling approaches: first-in-first-out and shortest process time (i.e. travel) scheduling rules. The results indicate that in most cases the Q-learning approach performs better than the two static scheduling approaches.
Akıllı Bir Otomatik Araçlı Depolama ve Çekme Sistem Tasarımı/ Banu Ekren
(2021) Arslan, Bartu; Eroğlu, Ecem; Küçükyaşar, Melis; Ekren, Banu Yetkin
Bu proje kapsamında gerçekleştirilen çalışmaların amacı, Otomatik Araçlı Depolama ve Çekme Sistemleri?nde (AVS/RS?lerinde), gerçek zamanlı bilgiyi kullanarak, sistemin kontrolünü, dinamik kararlar vererek yapabilen, bir ?akıllı depolama ve çekme sistem? tasarımı geliştirmektir. Sistemin önemli performans kriterleri olarak, bir depolama/çekme işleminin ortalama enerji tüketim miktarı ve bir depolama/çekme işleminin ortalama akış süresi esas alınmıştır. Akıllı depo sisteminin tasarımı için, ajan (etmen) tabanlı modelleme ve pekiştirmeli öğrenme yöntemlerinden yararlanılmıştır. Yapılan çalışmalar sonucunda geliştirilen dinamik modellerde, statik modellere kıyasla en az %20?lik bir performans iyileşmesinin gerçekleşmesi amaçlanmıştır. Bu doğrultuda, katlar arası dolaşabilen mekik-tabanlı SBS/RS?nde geliştirilen derin öğrenmeye dayalı sistem %40-%68 daha iyi sonuç üretmekte iken aynı zamanda geliştirilen koridorlar arası araç seyahatine izin veren esnek sistemde ise %55?e varan iyileşmeler söz konusu olmuştur.
An Open Vehicle Routing Problem for Daily Shipment Plan of a Local Bedding Company in Turkey
(Springer Science and Business Media Deutschland GmbH, 2020) Bartu Arslan; Özlem Erdin; Deniz Gülce Dağıstanlıoğlu; Ozan Bolel; Çağlar Su; Banu Yetkin Yetkin Ekren; Cansu Yurtseven; Yurtseven, Cansu; Su, Çağlar; Ekren, Banu Y.; Arslan, Bartu; Erdin, Özlem; Dağıstanlıoğlu, Deniz Gülce; Bolel, Ozan; 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 study deals with a daily shipment plan for a bedding company in Turkey. We develop a decision support system optimizing total travel distance of vehicles completing daily shipment plans. In detail the developed models both assign the shipment goods to capacitated vehicles as well as minimize the total travel distance of these vehicles by considering an open vehicle routing algorithm. The decision support system is coded by using the MS excel-macro and a user friendly tool is developed for users’ convenience. In the developed tool there are two different solution approaches: a mixed integer mathematical programming and a heuristic solution approaches. For the small size plans (i.e. less than 25 demand points) mathematical modelling guaranteeing the optimal solution can be utilized. For the large size cases the heuristic algorithm providing a near-optimal result promptly can be utilized. The macro-based tool is designed such a flexible way that the user can make any revisions on the provided plan whenever necessary. © 2022 Elsevier B.V. All rights reserved.
Kattan kata yolculuk eden SBS RS'te işlem seçimi için bir makine öğrenmesi uygulaması
(2021) Arslan, Bartu; Ekren, Banu Yetkin
With the recent growth of e-commerce, the order profiles have shifted towards smaller quantities with faster delivery time requests of customers. This change has led to companies seek for fast transaction processing automation technologies in operations of warehouses. Shuttle-based storage and retrieval system (SBS/RS) is an automated warehousing technology mostly utilized in large distribution centers because of its capability of processing high transaction rate. While the advantage of this system is its capability of processing high transaction rate by the excess numbers of shuttles in the system, a disadvantage is that the average utilization of shuttles is very low, compared to the lifting mechanisms in the system. Since a dedicated shuttle is assigned at each tier of an aisle, this system is also referred as tier-captive SBS/RS in literature. In an effort to balance the utilization levels of shuttles and lifts, a novel design referred as tier-to-tier SBS/RS is introduced. In that design, there is decreased number of shuttles in the system so that they are allowed to travel between tiers by using a separate lifting mechanism specifically dedicated for travel of them. This novel design not only balances the service lifts and shuttles, but also decreases the initial investment cost for the system by the decreased number of shuttles. However, those advantages cause a disadvantage, that is increased average cycle time per transaction performance metric in the system. In this thesis, in an effort to contribute on decreasing average cycle time per transaction performance metric, we apply a machine learning methodology for smart transaction processing in the system. Specifically, we apply Reinforcement Learning and Deep Reinforcement Learning methods for transaction selection of shuttles. The proposed approaches are compared with well-known First-in-First-out (FIFO) and Shortest Process Time (SPT) selection rules. The results show that the proposed approaches outperform both FIFO and SPT rules, significantly.
Simulation-based lateral transshipment policy optimization for s S inventory control problem in a single-echelon supply chain network
(2020) Banu Yetkin Ekren; Bartu Arslan; Arslan, Bartu; Ekren, Banu Yetkin
Since it affects the performance of whole supply chain significantly definition ofcorrect inventory control policy in a supply chain is critical. Recent technologicaldevelopment enabled real time visibility of a supply network by horizontalintegration of each node in a supply network. By this opportunity inventorysharing among stocking locations is also possible in the effort of cost minimizationin supply chain management. Hence lateral transshipment gained popularity andstudies seeking the best lateral-transshipment policy is still under research. In thisstudy we aim to compare different lateral-transshipment policies for an s Sinventory control problem for a single-echelon supply chain network system. Inthis work we consider a supply network with three stocking locations which mayperform lateral transshipment among them when backorder takes place. Wedevelop the simulation models of the systems in ARENA 14.5 commercialsoftware and compare the performance of the policies by minimizing the total costunder a pre-defined fill rate constraint by using an optimization tool OptQuest integrated in that software. The results show that lateral transshipment works wellcompared to the scenario when there is no lateral transshipment policy in thenetwork.
Citation - WoS: 14
Citation - Scopus: 17
Simulation-based lateral transshipment policy optimization for s S inventory control problem in a single-echelon supply chain network
(Balikesir University, 2020) Banu Yetkin Yetkin Ekren; Bartu Arslan; Arslan, Bartu; Ekren, Banu Y.
Since it affects the performance of whole supply chain significantly definition of correct inventory control policy in a supply chain is critical. Recent technological development enabled real time visibility of a supply network by horizontal integration of each node in a supply network. By this opportunity inventory sharing among stocking locations is also possible in the effort of cost minimization in supply chain management. Hence lateral transshipment gained popularity and studies seeking the best lateral-transshipment policy is still under research. In this study we aim to compare different lateral-transshipment policies for an s S inventory control problem for a single-echelon supply chain network system. In this work we consider a supply network with three stocking locations which may perform lateral transshipment among them when backorder takes place. We develop the simulation models of the systems in ARENA 14.5 commercial software and compare the performance of the policies by minimizing the total cost under a pre-defined fill rate constraint by using an optimization tool OptQuest integrated in that software. The results show that lateral transshipment works well compared to the scenario when there is no lateral transshipment policy in the network. © 2023 Elsevier B.V. All rights reserved.
Citation - Scopus: 2
Smart transaction picking in tier-to-tier SBS/RS by deep Q-learning
(IEOM Society, 2021) Bartu Arslan; Banu Yetkin Yetkin Ekren; Arslan, Bartu; Ekren, Banu Y.
By the rapid growth of e-commerce the intralogistics sector is facing new challenges. Intralogistics sector requires more flexible scalable processes with maximum reliability and availability. They are complicated and interconnected systems whose all components are required to be perfectly coordinated with each other for optimal functionality. In this work we study an intralogistics technology shuttle-based storage and retrieval system (SBS/RS) where shuttles are tier-to-tier. In this novel system design in an effort to increase shuttle utilization as well as decrease initial investment cost shuttles are designed in a more flexible travel manner so that they can change their tiers within an aisle by using a separate lifting mechanism. Due to the complexity of such system design as well as aiming to obtain fast transaction process time by the decreased number of shuttles in the system we implement a Deep Q-Learning (DQL) approach to let shuttles select the best transaction to process based on its targets. We compare the performance of the DQL by the average cycle time per transaction performance metric with the other well-known selection rules First-in-First-Out (FIFO) and Shortest Process Time (SPT). Results show that Deep Q-Learning approach produces better results than those FIFO and SPT. © 2021 Elsevier B.V. All rights reserved.
Citation - WoS: 16
Citation - Scopus: 19
Transaction selection policy in tier-to-tier SBSRS by using Deep Q-Learning
(Taylor and Francis Ltd., 2023) Bartu Arslan; Banu Yetkin Yetkin Ekren; Arslan, Bartu; Ekren, Banu Yetkin
This paper studies a Deep Q-Learning (DQL) method for transaction sequencing problems in an automated warehousing system Shuttle-based Storage and Retrieval System (SBSRS) in which shuttles can move between tiers flexibly. Here the system is referred to as tier-to-tier SBSRS (t-SBSRS) developed as an alternative design to tier-captive SBSRS (c-SBSRS). By the flexible travel of shuttles between tiers in t-SBSRS the number of shuttles in the system may be reduced compared to its simulant c-SBSRS design. The flexible travel of shuttles makes the operation decisions more complex in that system motivating us to explore whether integration of a machine learning approach would help to improve the system performance. We apply the DQL method for the transaction selection of shuttles in the system to attain process time advantage. The outcomes of the DQN are confronted with the well-applied heuristic approaches: first-come-first-serve (FIFO) and shortest process time (SPT) rules under different racking and numbers of shuttles scenarios. The results show that DQL outperforms the FIFO and SPT rules promising for the future of smart industry applications. Especially compared to the well-applied SPT rule in industries DQL improves the average cycle time per transaction by roughly 43% on average. © 2023 Elsevier B.V. All rights reserved.