Yazar "Kabakulak, Banu" seçeneğine göre listele

Listeleniyor 1 - 5 / 5
  • Küçük Resim
    Öğe
    A lot-sizing problem in deliberated and controlled co-production systems
    (Taylor and Francis Ltd., 2022) Kabakulak, Banu
    Abstract: We consider an uncapacitated lot-sizing problem in co-production systems, in which it is possible to produce multiple items simultaneously in a single production run. Each product has a deterministic demand to be satisfied on time. The decision is to choose which items to co-produce and the amount of production throughout a predetermined planning horizon. We show that the lot-sizing problem with co-production is strongly NP-Hard. Then, we develop various Mixed-Integer Linear Programming (MILP) formulations of the problem and show that LP relaxations of all MILPs are equal. We develop a separation algorithm based on a set of valid inequalities, lower bounds based on a dynamic lot-sizing relaxation of our problem and a constructive heuristic that is used to obtain an initial solution for the solver, which form the basis of our proposed Branch & Cut algorithm for the problem. We test our models and algorithms on different data sets and provide the results. © Copyright © 2022 “IISE”.
  • Küçük Resim Yok
    Öğe
    A MACHINE SPEED OPTIMIZATION PROBLEM IN A CAPACITATED FELT PRODUCTION SYSTEM
    (2025) Kabakulak, Banu
    In this study, we analyze a felt production system with unique requirements, such as maintaining machine speeds within specific limits to facilitate successful chemical reactions. By incorporating machine speed constraints and restrictions on both work-in-process (WIP) and end-product inventories, we aim to determine the optimal production quantities for each felt type and the corresponding machine speeds over a defined planning horizon. The objective is to minimize total costs, including machine setup and production costs as well as WIP and end-product inventory holding costs. To achieve this, we introduce the Machine Speed Optimization (MSO) problem and adapt it to the specific requirements of a felt manufacturing company operating in Istanbul, Türkiye. The MSO problem is formulated as a mixed-integer linear programming (MILP) model, which is NP-hard to solve for optimal production decisions. We validate the MSO model using the felt manufacturing company's case over five periods, demonstrating its effectiveness in automating production planning and machine speed decisions. The simulations for a 5-day planning horizon demonstrate a cost reduction of 3853 TL, a 24% decrease in WIP inventory, and up to a 15% improvement in machine utilization compared to the current practices of the felt manufacturing company. Additionally, the optimized machine speeds achieved through the MSO model enable the system to increase throughput by 11%. Experimental analysis of computational complexity reveals that the MSO model can generate an optimal 6-month production plan, including machine speeds, in under one hour.
  • Küçük Resim Yok
    Öğe
    A Mathematical Programming Approach for IoT-Enabled, Energy-Efficient Heterogeneous Wireless Sensor Network Design and Implementation
    (Mdpi, 2024) Taparci, Ertugrul; Olcay, Kardelen; Akmandor, Melike Ozlem; Kabakulak, Banu; Sarioglu, Baykal; Gokdel, Yigit Daghan
    The Internet of Things (IoT) is playing a pivotal role in transforming various industries, and Wireless Sensor Networks (WSNs) are emerging as the key drivers of this innovation. This research explores the utilization of a heterogeneous network model to optimize the deployment of sensors in agricultural settings. The primary objective is to strategically position sensor nodes for efficient energy consumption, prolonged network lifetime, and dependable data transmission. The proposed strategy incorporates an offline model for placing sensor nodes within the target region, taking into account the coverage requirements and network connectivity. We propose a two-stage centralized control model that ensures cohesive decision making, grouping sensor nodes into protective boxes. This grouping facilitates shared resource utilization, including batteries and bandwidth, while minimizing box number for cost-effectiveness. Noteworthy contributions of this research encompass addressing connectivity and coverage challenges through an offline deployment model in the first stage, and resolving real-time adaptability concerns using an online energy optimization model in the second stage. Emphasis is placed on the energy efficiency, achieved through the sensor consolidation within boxes, minimizing data transmission hops, and considering energy expenditures in sensing, transmitting, and active/sleep modes. Our simulations on an agricultural farmland highlights its practicality, particularly focusing on the sensor placement for measuring soil temperature and humidity. Hardware tests validate the proposed model, incorporating parameters from the real-world implementation to enhance calculation accuracy. This study provides not only theoretical insights but also extends its relevance to smart farming practices, illustrating the potential of WSNs in revolutionizing sustainable agriculture.
  • Küçük Resim Yok
    Öğe
    A branch-cut-and-price algorithm for optimal decoding in digital communication systems
    (Springer, 2021) Kabakulak, Banu; Taskin, Z. Caner; Pusane, Ali Emre
    Channel coding aims to minimize the errors that occur during the transmission of digital information from one place to another. Low-density parity-check codes can detect and correct transmission errors if one encodes the original information by adding redundant bits. In practice, heuristic iterative decoding algorithms are used to decode the received vector. However, these algorithms may fail to decode if the received vector contains multiple errors. We consider decoding the received vector with minimum error as an integer programming (IP) problem and propose a branch-and-price method for its solution. We improve the performance of our method by introducing heuristic feasible solutions and adding valid cuts to the mathematical formulation. Our computational experiments reveal that our branch-cut-and-price algorithm significantly improves solvability of the problem compared to a state-of-the-art IP decoder in the literature and has superior error performance than the conventional sum-product algorithm.
  • Küçük Resim Yok
    Öğe
    Energy Efficient Mission Control of Unmanned Intelligent Swarm Systems
    (Springer International Publishing, 2023) Kabakulak, Banu
    The unmanned vehicles capture increasing attendance in the last few decades since they can be used in a wide variety of applications on the ground, air, or sea. The unmanned systems, which avoid dangerous cases for humans, are especially preferred for high-risk missions such as battlefield surveillance. Evolving technologies on the swarms of unmanned systems allow low cost and quick execution of various missions easily by cooperation. As an example, a large agricultural area can be scanned by a swarm of drones in a shorter time to detect abnormalities. However, unmanned systems have limited energy resources such as battery or fuel. Hence, it is of critical importance for swarm missions of unmanned systems to be planned optimally in terms of scarce energy resources. In this chapter, we give some background technical information about the unmanned systems and explain swarm mission problems with their solution methods. We also mention some useful softwares for implementation of the swarm systems with some concluding remarks on the future research tracks on the swarm missions. © Springer Nature Switzerland AG 2023.