Article
Automation & Control Systems
Tsung-Liang Wu, Yu-Chun Hwang, Wei-Xun Zhang
Summary: This study aimed to modify processing parameters for hot forging tools in real time to improve workpiece quality and prolong tool life. Accelerometers were used to capture machine motion, wear and thermal experiments were conducted, and feature extraction methods were applied in supervised learning models. The methodologies proposed were evaluated in two case studies.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2022)
Article
Automation & Control Systems
Jules Le Lay, Vincent Augusto, Edgar Alfonso-Lizarazo, Malek Masmoudi, Baptiste Gramont, Xiaolan Xie, Bienvenu Bongue, Thomas Celarier
Summary: This paper proposes a two-step approach based on process mining and discrete-event simulation to determine the size of a recovery unit for COVID-19 patients in a hospital. A decision aid framework is introduced to assist hospital managers in making important decisions, taking into account all units of the hospital. The impact of bed transfers on hospital admission pathways is evaluated using process mining and discrete-event simulation. The robustness of the approach is assessed through a practical case study in collaboration with a local hospital.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Article
Quantum Science & Technology
Francesco Scala, Andrea Ceschini, Massimo Panella, Dario Gerace
Summary: This article presents a generalized approach to applying the dropout technique in quantum neural network models, with different quantum dropout strategies analyzed to avoid overfitting and achieve a high level of generalization. The study highlights that quantum dropout does not impact the expressibility and entanglement of QNN models.
ADVANCED QUANTUM TECHNOLOGIES
(2023)
Article
Computer Science, Information Systems
Shinho Lee, Wookhyun Jung, Wonrak Lee, Hyung Geun Oh, Eui Tak Kim
Summary: Research on categorization and detection of Android malware has focused on proposing different learned models and machine learning algorithms. This study examines dataset construction and proposes methods to determine bias and variance, as well as reduce labeling noise. The proposed method goes beyond existing methods by using unified labels and opcode morphology for constructing new types of datasets.
Review
Immunology
Sikemi Ibikunle, Dolores Grosso, Usama Gergis
Summary: Allogeneic hematopoietic stem cell transplantation (HSCT) is the only potentially curative option for multiple hematological conditions. The two-step approach to HSCT, described here, aims to achieve an optimal balance in immune recovery, graft-versus-host disease control, and disease control. This review highlights key results from studies over the past two decades and discusses current strategies to optimize graft-versus-tumor effect and limit transplant-related toxicities.
FRONTIERS IN IMMUNOLOGY
(2023)
Article
Computer Science, Information Systems
Kazuki Tajima, Bojian Du, Yoshiaki Narusue, Shinobu Saito, Yukako Iimura, Hiroyuki Morikawa
Summary: Cross-organizational process mining aims to discover an entire process model across multiple organizations with independent ID systems. This paper proposes an accurate technique that utilizes common items in event logs and can handle cyclic orchestrations.
Article
Automation & Control Systems
Niladri S. Chatterji, Philip M. Long, Peter L. Bartlett
Summary: This paper investigates the phenomenon of benign overfitting in neural network models, deriving bounds on excess risk for two-layer linear neural networks trained with gradient flow on the squared loss. The study highlights the importance of both initialization quality and data covariance matrix properties in achieving low excess risk.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Energy & Fuels
Zhenxiao Shang, Zhen Yang, Yanfei Ma
Summary: Using waste biomass to produce high-value activated carbon is crucial for biomass resource utilization. The energy and mass balance during carbonization and steam activation is important for the system operation. Control of activation temperature and steam/bio-char ratio is essential for the activation process.
BIOMASS CONVERSION AND BIOREFINERY
(2021)
Article
Chemistry, Physical
A. R. Finney, M. Salvalaglio
Summary: Molecule- and particle-based simulations are used to test classical nucleation theory. The variational approach to Markov processes is applied to determine the suitability of different reaction coordinates for studying crystallization from supersaturated colloid suspensions. The results show that collective variables that correlate with the number of particles in the condensed phase, the system potential energy, and approximate configurational entropy are the most appropriate order parameters.
JOURNAL OF CHEMICAL PHYSICS
(2023)
Article
Engineering, Mechanical
Hu-Shuang Hou, Cheng Luo, Hua Zhang, Guo-Cheng Wu
Summary: In this paper, a discrete-time recurrent neural network is presented using the Euler scheme. The time step size is set as a bifurcation parameter and the frequency domain approach is used for Hopf bifurcation analysis. Periodic solutions are obtained using the harmonic balance method, and stability conditions are presented. The critical step size is determined for the discrete-time recurrent neural network to inherit the stable state from the continuous-time one. A numerical example is provided to support the theoretical analysis.
NONLINEAR DYNAMICS
(2023)
Article
Engineering, Environmental
Karan Chabhadiya, Rajiv Ranjan Srivastava, Pankaj Pathak
Summary: The study demonstrated the effectiveness of a two-step leaching process using organic and mineral acids for recycling exhausted Li-ion batteries, offering a sustainable solution for protecting the urban environment and fulfilling the demand for critical metals. The process optimized experimental parameters and successfully achieved high efficiency in leaching lithium, copper, cobalt, nickel, and manganese from the cathode material. This approach not only secures a secondary supply of critical metals but also provides an environmentally sustainable route for waste valorization in a circular economy.
JOURNAL OF ENVIRONMENTAL CHEMICAL ENGINEERING
(2021)
Article
Computer Science, Artificial Intelligence
Heidy Marisol Marin-Castro, Edgar Tello-Leal
Summary: This paper presents an approach to discover business process models based on BPMN standard from event logs. Experimental results show that the process models derived by this method exhibit better results in fitness and precision.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Chemistry, Multidisciplinary
Lou Gao, Ke Du, Tianlan Yan, He Li, Di Pan, Yahong Zhang, Yi Tang
Summary: This study proposes a strategy for the efficient production of lactide from biomass-derived carbohydrate, and presents the structural characteristics of the resulting lactide and its extensive applicability to various substrates.
CHEMICAL COMMUNICATIONS
(2022)
Article
Chemistry, Physical
Zhiyuan Zhang, Shuai Wang, Xianjuan Pang, Juan Liu, Lin Zhu, Zhen Feng, Sanming Du, Jun Yang, Yongzhen Zhang
Summary: This study investigates the effect of sintering temperature on the microstructures and mechanical properties of Mo2BC ceramics. The two-step sintering process is found to significantly improve the relative density, Vickers hardness, compressive strength, flexural strength, and fracture toughness of the Mo2BC ceramics.
JOURNAL OF ALLOYS AND COMPOUNDS
(2023)
Article
Green & Sustainable Science & Technology
Mohammad Javad Mirzaei, Ahad Kazemi
Summary: This paper presents a two-step approach for optimal energy management in electric vehicle parking lots, utilizing a new scheduling method for charge and discharge of electric vehicles and a modified approach to ensure maximum profit for EV owners. By implementing innovative management strategies based on equations and policies, the approach aims to accurately determine charge/discharge times and increase profitability.
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2021)
Article
Computer Science, Software Engineering
Alessandro Berti, Wil M. P. van der Aalst
Summary: Object-centric process mining is a branch of process mining that aims to analyze event data without forming exclusive event groups. This paper provides an overview of exploring and processing object-centric event logs to discover information related to the lifecycle of different objects. The paper also introduces a comprehensive tool support for implementing the proposed techniques.
INTERNATIONAL JOURNAL ON SOFTWARE TOOLS FOR TECHNOLOGY TRANSFER
(2023)
Editorial Material
Computer Science, Information Systems
Wil M. P. van der Aalst, Oliver Hinz, Christof Weinhardt
BUSINESS & INFORMATION SYSTEMS ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Luciana Barbieri, Edmundo Madeira, Kleber Stroeh, Wil van der Aalst
Summary: Despite recent advances in process mining, making it accessible to non-technical users remains a challenge. To democratize this technology, a natural language querying interface is proposed to allow users to retrieve relevant information and insights about their processes by asking questions in plain English. A reference architecture is proposed to support natural language questions and provide the right answers by integrating with existing process mining tools.
JOURNAL OF INTELLIGENT INFORMATION SYSTEMS
(2023)
Editorial Material
Computer Science, Information Systems
Timm Teubner, Christoph M. Flath, Christof Weinhardt, Wil van der Aalst, Oliver Hinz
BUSINESS & INFORMATION SYSTEMS ENGINEERING
(2023)
Article
Computer Science, Information Systems
Jan Niklas Adams, Sebastiaan J. van Zelst, Thomas Rose, Wil M. P. van der Aalst
Summary: This paper introduces a framework to extract concept drifts and their potential root causes from event data. It extracts time series describing process measures, detects concept drifts, and tests these drifts for correlation. The framework supports object-centric event data with multiple case notions, non-linear relationships, and an arbitrary number of process measures.
INFORMATION SYSTEMS
(2023)
Article
Computer Science, Artificial Intelligence
Mohammadreza Fani Sani, Mozhgan Vazifehdoostirani, Gyunam Park, Marco Pegoraro, Sebastiaan J. van Zelst, Wil M. P. van der Aalst
Summary: Predictive process monitoring aims to estimate case or event features for running process instances in process mining. However, current methods for predictive monitoring require training complex machine learning models, which is inefficient. This paper proposes an instance selection procedure that improves training speed for prediction models while maintaining prediction accuracy.
JOURNAL OF INTELLIGENT INFORMATION SYSTEMS
(2023)
Article
Mathematics
Wil M. P. van der Aalst
Summary: Traditional process modeling and analysis methods focus on one type of object per event, but in reality, events often involve multiple objects of different types. Object-centric process mining takes a more holistic approach by considering multiple object types and events. This paper introduces object-centric event data and demonstrates their use in discovering, analyzing, and improving complex processes.
Article
Automation & Control Systems
Jan Niklas Adams, Gyunam Park, Wil M. P. van der Aalst
Summary: This paper presents a method to preserve the graph structure of object-centric event data in process mining using machine learning techniques. Two different techniques are provided to preserve these structures, and their performances are evaluated in predictive process monitoring tasks. Different graph embedding techniques for object-centric event logs are compared for future research and deployment.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Azadeh Sadat Mozafari Mehr, Renata M. de Carvalho, Boudewijn van Dongen
Summary: In recent years, organizations have been paying more attention to anomaly detection as anomalies in business processes can indicate system faults, inefficiencies, or even fraudulent activities. This paper introduces an approach that considers different perspectives of a business process simultaneously to detect complex anomalies such as spurious data processing and misusage of authorizations. The approach has been implemented in the ProM framework and evaluated using real-life data from a financial organization, showing its ability to detect anomalies related to multiple aspects of a business process.
PROCESS MINING WORKSHOPS, ICPM 2022
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Gyunam Park, Wil M. P. van der Aalst
Summary: Constraint monitoring aims to detect constraint violations in business processes by analyzing event data. However, existing techniques assume a single case notion and fail to consider the interaction of multiple objects in the processes. In this work, we propose an approach using Object-Centric Constraint Graphs (OCCGs) to represent and evaluate constraints in object-centric business processes. We conducted case studies with a real-life SAP ERP system to validate the proposed approach.
PROCESS MINING WORKSHOPS, ICPM 2022
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Harry H. Beyel, Wil M. P. van der Aalst
Summary: Event logs capture information about executed activities, but not about enabled but not executed activities. Translucent event logs contain information on enabled activities. We propose two techniques to increase the availability of translucent event logs. The first technique records system states as snapshots and links them to events, adding information about enabled activities. The second technique uses a process model to add information about enabled activities to existing event logs, improving process discovery.
PROCESS MINING WORKSHOPS, ICPM 2022
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Alexandre Goossens, Johannes De Smedt, Jan Vanthienen, Wil M. P. van der Aalst
Summary: When multiple objects are involved in a process, object-centric event logs are an interesting development in process mining as they support the presence of multiple types of objects. However, the current object-centric event log formats do not fully support dynamic object attributes. This paper introduces the Data-aware OCEL format, along with an algorithm that automatically translates XES logs to this format, addressing the issues present in the existing formats.
PROCESS MINING WORKSHOPS, ICPM 2022
(2023)
Proceedings Paper
Computer Science, Interdisciplinary Applications
Tsung-Hao Huang, Wil M. P. van der Aalst
Summary: Process discovery aims to learn process models from observed behaviors in information systems. The discovered models serve as the starting point for process mining techniques to address performance and compliance problems. The algorithm applying synthesis rules from the free-choice net theory can discover process models with flexible structures while ensuring desirable properties. The proposed ordering strategy improves the quality of resulting process models and reduces computation time compared to frequency-based ordering strategy.
SERVICE-ORIENTED COMPUTING - ICSOC 2022 WORKSHOPS
(2023)
Review
Computer Science, Information Systems
Majid Rafiei, Wil M. P. van der Aalst
Summary: Process awareness is crucial for business success, and process mining is a powerful tool to discover and analyze actual business processes using event data. However, privacy concerns in inter-organizational settings hinder the sharing of event data and processes between partner organizations. This paper proposes an abstraction-based approach to enable privacy-aware process mining in such settings, addressing the challenges of collaboration and confidentiality.
Article
Computer Science, Interdisciplinary Applications
Gyunam Park, Daniel Schuster, Wil M. P. van der Aalst
Summary: As business environments become more competitive, organizations strive to improve their business processes to reduce costs and increase quality and productivity. In this study, we propose an action engine that automatically generates action plans by analyzing monitoring results to address operational problems in business processes.
COMPUTERS IN INDUSTRY
(2023)
