Article
Energy & Fuels
Sudlop Ratanakuakangwan, Hiroshi Morita
Summary: This study proposes a combination of multi-objective optimization and efficiency measurement for determining an efficient energy mix in energy planning. It considers various dimensions of energy planning and associated uncertainties. The proposed model includes multiple objective functions related to energy need, cost, environmental impact, security, social impact, and social benefit. The slacks-based measure methodology is applied to identify the best energy mix. The results show significant improvements in reducing emissions and dependence on certain power plant types, increasing employment and the proportion of electricity generated from renewable sources, with slight tradeoffs in costs. The quantitative results from the model can assist policymakers in efficiently determining an energy policy that optimizes various aspects under given constraints and scenarios of uncertainty.
Article
Computer Science, Information Systems
Salvador Botello-Aceves, S. Ivvan Valdez, Arturo Hernandez-Aguirre
Summary: This article presents a mapping method for improvement directions in multi-objective optimization and proposes an iterative updating solution based on the Broyden method. Compared to other transformations reported in specialized literature, the proposed method shows better conditioning and provides more accurate search directions.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Huizhen Zhang, Kun Zhang, Yuting Chen, Liang Ma
Summary: This paper studies a two-level medical facility location problem with multiple patient flows and proposes a solution approach that is validated using a real case.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Weixiong Huang, Juan Zou, Yuan Liu, Shengxiang Yang, Jinhua Zheng
Summary: This paper proposes a constrained multi-objective evolutionary algorithm framework based on global and local feasible solutions search to address the complexity of feasible regions caused by constraints. The framework is divided into three stages and an adaptive method is used to decide when to switch the search state. The experimental results show that the proposed framework is highly competitive for solving CMOPs.
INFORMATION SCIENCES
(2023)
Article
Engineering, Civil
Di Weng, Ran Chen, Jianhui Zhang, Jie Bao, Yu Zheng, Yingcai Wu
Summary: Planning ideal transit routes in complex urban environments can enhance the performance of public transportation systems, but is challenging due to the huge number of possible routes. This study introduces the definition of pareto-optimal transit routes based on multiple criteria and develops an efficient search framework to address route extraction challenges.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2021)
Article
Thermodynamics
Sinvaldo Rodrigues Moreno, Juliano Pierezan, Leandro dos Santos Coelho, Viviana Cocco Mariani
Summary: This study proposes a novel multiobjective algorithm based on the lightning search algorithm (MO-LSA) for designing wind farm layouts, aiming to minimize the cost of annual energy production, the overall wind farm area, and wake effect losses. By applying concepts of the convex hull and evaluating different wind speed scenarios, the performance of MO-LSA is compared with three other multiobjective optimization algorithms from the literature. The results show that MO-LSA provides the best Pareto front for the analyzed scenarios, with well-distributed solutions across the search space and alternative wind park layouts with improved efficiency in terms of power output and investment costs.
Article
Computer Science, Artificial Intelligence
Biyue Li, Tong Guo, Yi Mei, Yumeng Li, Jun Chen, Yu Zhang, Ke Tang, Wenbo Du
Summary: This paper proposes a multi-objective airspace complexity mitigation model to optimize flight trajectories and ensure the safety and efficiency of air transport. The proposed Memetic Algorithm with Adaptive Local Search effectively solves the multi-objective and non-linear optimization problem. Comprehensive comparisons with other algorithms on Chinese air traffic dataset show the superiority of the proposed algorithm in reducing airspace complexity.
SWARM AND EVOLUTIONARY COMPUTATION
(2023)
Article
Green & Sustainable Science & Technology
Mohamed Abdallah, Sadeque Hamdan, Ahmad Shabib
Summary: The study proposed a systematic optimization framework to identify the best waste management strategies, including maximizing material and energy recovery in facilities such as MRFs and AD plants. The optimal strategy reduced the carbon footprint from landfilling by 97.6%, increased profitability by approximately 288%, and could cover around 4.2% of the UAE's total energy demand.
JOURNAL OF CLEANER PRODUCTION
(2021)
Article
Energy & Fuels
Madathodika Asna, Hussain Shareef, Munir Azam Muhammad, Leila Ismail, Achikkulath Prasanthi
Summary: This paper introduces an effective planning methodology for electric vehicle fast-charging stations using a multi-objective binary version of the atom search optimization algorithm with quantum operations. By incorporating non-dominated sorting and Pareto concepts, the algorithm shows improved search capability and efficiency, successfully solving multi-objective optimization problems for fast-charging station location planning.
INTERNATIONAL JOURNAL OF ENERGY RESEARCH
(2022)
Article
Computer Science, Artificial Intelligence
Lourdes Uribe, Adriana Lara, Kalyanmoy Deb, Oliver Schutze
Summary: The study proposes an effective method for computing multi-objective descent directions without the need for explicit gradient computation, obtaining the direction by extracting information from the current population of the MOEA. Two hybrid methods focused on specific types of problems are demonstrated, with numerical results on benchmark problems supporting the benefits of the novel approach.
SWARM AND EVOLUTIONARY COMPUTATION
(2021)
Article
Computer Science, Artificial Intelligence
Quan Minh Phan, Ngoc Hoang Luong
Summary: Neural Architecture Search (NAS) is a multi-objective optimization problem that automates the design process of high-performing neural network architectures. This article introduces a local search method, PSI, to enhance the performance of MOEAs. Experimental results confirm the effectiveness of the proposed method, reducing computational costs significantly.
APPLIED INTELLIGENCE
(2023)
Article
Green & Sustainable Science & Technology
Chengzhou Li, Ligang Wang, Yumeng Zhang, Hangyu Yu, Zhuo Wang, Liang Li, Ningling Wang, Zhiping Yang, Francois Marechal, Yongping Yang
Summary: This study proposes a multi-objective optimization methodology for planning distributed energy systems considering process synergy and thermal integration. The system design and dispatch strategy are optimized to achieve economic benefit, reduce carbon emission, and decrease fossil fuel consumption. A new multi-energy complementary distributed energy system is developed, which includes solar energy utilization and hybrid energy storage technologies. The results show significant cost reduction and environmental benefits compared to traditional energy systems. The optimal design of the system provides a reference for decision-making and flexible operation.
JOURNAL OF CLEANER PRODUCTION
(2023)
Article
Computer Science, Artificial Intelligence
Sumit Kumar, Ghanshyam G. Tejani, Nantiwat Pholdee, Sujin Bureerat
Summary: A novel Multi-Objective Passing Vehicle Search (MOPVS) algorithm is proposed for structural design optimization, and it outperforms other algorithms in terms of efficiency and diversity in solving challenging design problems.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Hongtao Tang, Senli Ren, Weiguang Jiang, Qingfeng Chen
Summary: This paper proposes a method combining improved lowest horizontal line method and particle swarm optimization algorithm to solve UA-FLP, achieving multi-objective optimization of material handling cost, the adjacent value, and the utilization rate of floor shop. The algorithm simplifies the legalization of facility layout, overcomes the shortcomings of previous facility layout methods, and shows promising results in experiments.
Article
Energy & Fuels
Jonas Finke, Valentin Bertsch
Summary: In order to address climate change, the energy sector is transitioning towards a climate-neutral future based on renewable energy sources. Energy system models are important tools for generating insights and supporting decision-making in this transformation. However, there is a need for highly adaptable energy system models that can incorporate multiple objectives. This paper presents an implementation of the augmented epsilon-constraint method using the energy system optimization framework Backbone, which allows for the simultaneous optimization of multiple objectives in different energy systems.
Article
Computer Science, Artificial Intelligence
Jacqueline Schmitt, Jochen Boenig, Thorbjoern Borggraefe, Gunter Beitinger, Jochen Deuse
ADVANCED ENGINEERING INFORMATICS
(2020)
Article
Computer Science, Artificial Intelligence
Thomas Schlegl, Stefan Schlegl, Domenico Tomaselli, Nikolai West, Jochen Deuse
Summary: Improving the recall of information retrieval systems for similarity search in time series databases is of great practical importance. In this paper, a novel adaptive search algorithm is proposed that refines the query based on user feedback and adapts to new patterns without user input. Experimental results show that the algorithm achieves considerably higher recall for fault pattern retrieval compared to other state-of-the-art adaptive search algorithms, and these results are transferable to other domains.
ADVANCED ENGINEERING INFORMATICS
(2022)
Article
Engineering, Industrial
Ruba Al-Zqebah, Florian Hoffmann, Nick Bennett, Jochen Deuse, Lee Clemon
Summary: This study compares the performance of curved and linear layouts in woolsheds using computer-aided production engineering simulation. The results show that the curved layout improves production efficiency compared to the linear layout. The study also highlights the impact of equipment capacity and shearer speed on production efficiency, rather than the number of wool handlers. This is the first application of discrete event simulation to evaluate woolshed operations and demonstrates the potential gains from layout, equipment, and worker changes.
JOURNAL OF INDUSTRIAL ENGINEERING AND MANAGEMENT-JIEM
(2022)
Article
Engineering, Industrial
Ralph Richter, Marius Syberg, Jochen Deuse, Peter Willats, David Lenze
Summary: This paper introduces a PDCA cycle to analyze and reduce variability in value streams. It divides the value stream into stable and unstable zones, and applies measures to reduce variability in the unstable zones, with the goal of gradually turning them into stable zones and extending the sustainable implementation of lean practices. An IT system is developed to acquire, process, and visualize the data, providing structured information for experts and management to reduce production variability.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2023)
Article
Automation & Control Systems
Thomas Schlegl, Domenico Tomaselli, Stefan Schlegl, Nikolai West, Jochen Deuse
Summary: This paper presents an algorithm that automates the manual process of defining control limits for fault detection. The algorithm can efficiently search for control limits in annotated time series data and achieves state-of-the-art performance on traditional time series classification problems.
JOURNAL OF PROCESS CONTROL
(2022)
Article
Engineering, Industrial
Frederik Ferid Ostermeier, Jochen Deuse
Summary: Forgetting effects occur when production is interrupted and workers lose routine. Existing models ignored the impact of intermittent production in mixed-model environments, leading to ineffective scheduling. This study extends existing forgetting models to consider intermittent production as interruptions and introduces a routine loss factor. Simulation studies show that non-optimal schedules can be avoided by appropriately modeling forgetting.
FLEXIBLE SERVICES AND MANUFACTURING JOURNAL
(2023)
Article
Computer Science, Artificial Intelligence
Michael Mario Wocker, Frederik Ferid Ostermeier, Tobias Wanninger, Ronny Zwinkau, Jochen Deuse
Summary: In highly automated manufacturing systems, preventive maintenance activities need to be executed during production times, even in 24/7 operation. This research introduces a mixed-integer program that models both job scheduling and maintenance activity assignment in flexible job shops. A local search algorithm is developed to solve both problems in an integrated way. Numerical studies based on real data show that joint job scheduling and maintenance activity assignment is essential for minimizing the makespan and only a limited amount of maintenance activities can be compensated.
JOURNAL OF INTELLIGENT MANUFACTURING
(2023)
Article
Engineering, Industrial
Frederik Ferid Ostermeier, Jens Jaehnert, Jochen Deuse
Summary: Several parts supply strategies, both production-order-independent and -dependent, can be used to supply parts from warehouses to assembly lines. The sequencing, kitting, and batch supply strategies share similarities that allow joint modelling, while line stocking, just-in-time, just-in-sequence, and just-in-sequence kit supply require separate modelling. Joint modelling is essential for setting up an efficient software system that covers multiple parts supply strategies. This study provides insights from a software system implemented in the automotive industry for sequencing, kitting, and batch supply.
PRODUCTION AND MANUFACTURING RESEARCH-AN OPEN ACCESS JOURNAL
(2023)
Proceedings Paper
Automation & Control Systems
Marius Syberg, Nikolai West, Jorn Schwenken, Rebekka Adams, Jochen Deuse
Summary: The ongoing digitization of online learning resources has led to the proliferation of collaboration platforms for specific areas of application and disciplines. This paper focuses on deriving collaborative and competency-based requirements for implementing a collaboration platform for industrial data analytics. The defined requirements are transformed into features and applied in an online platform, with validation in a dynamic value network system.
IOT AND DATA SCIENCE IN ENGINEERING MANAGEMENT
(2023)
Proceedings Paper
Automation & Control Systems
Mikhail Polikarpov, Georgii Emelianov, Fabian Huebner, Aqib Farooq, Rekha Prasad, Jochen Deuse, Jochen Schiemann
Summary: The recycling of refrigerating appliances is crucial for protecting the Earth's atmosphere from ozone depletion and greenhouse gas emissions. However, the current manual and error-prone data collection process in recycling plants is inefficient. This paper proposes an automated data collection system that utilizes pre-trained vision models and laser scanning technology to extract attributes of individual refrigerators and estimate the material content. This system enables continuous performance monitoring and efficient control of refrigerator recycling plants.
2022 IEEE 27TH INTERNATIONAL CONFERENCE ON EMERGING TECHNOLOGIES AND FACTORY AUTOMATION (ETFA)
(2022)
Article
Engineering, Multidisciplinary
Jochen Deuse, Nikolai West, Marius Syberg
Summary: Industrial Engineering has been crucial for the success of manufacturing companies and its job description is now expanding to include Industrial Data Science. This paper reviews the origins of Industrial Engineering and considers possibilities for future applications of Industrial Data Science.
INTERNATIONAL JOURNAL OF PRODUCTION MANAGEMENT AND ENGINEERING
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Nikolai West, Jorn Schwenken, Jochen Deuse
Summary: This paper presents and compares three methods for detecting shifting bottlenecks in manufacturing systems. The comparative study conducted on a specific production line shows that the Active Period Method (APM) and an adaptation of Interdeparture Time Variances (ITV) achieve the highest agreement, with an average of 80.10%.
ADVANCES AND TRENDS IN ARTIFICIAL INTELLIGENCE: THEORY AND PRACTICES IN ARTIFICIAL INTELLIGENCE
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Thomas Schlegl, Stefan Schlegl, Amelie Sciberras, Nikolai West, Jochen Deuse
Summary: Real-world big data applications require detecting rare patterns, but existing algorithms lack robustness and generalization. We propose enhancements to shapelet algorithm to improve feature diversity and accuracy, suitable for highly imbalanced data.
2021 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Nikolai West, Jonas Gries, Carina Brockmeier, Jens C. Goebel, Jochen Deuse
Summary: The application of Industrial Data Science in connected Smart Products requires data modeling and structuring, and ensuring data quality to meet user requirements. Data preparation aims to provide high-quality data for users, and the developed model provides an approach for assessing and ensuring Data Analysis Quality.
2021 IEEE 22ND INTERNATIONAL CONFERENCE ON INFORMATION REUSE AND INTEGRATION FOR DATA SCIENCE (IRI 2021)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Rene Woestmann, Philipp Schlunder, Fabian Temme, Ralf Klinkenberg, Josef Kimberger, Andrea Spichtinger, Markus Goldhacker, Jochen Deuse
2020 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA)
(2020)
