Discrete Manufacturing Systems Research Articles

A new redundancy strategy for enabling graceful degradation in resilient robotic flexible assembly cells

The development of resilience in manufacturing systems has drawn more attention than ever. Using redundant components is one of the key strategies for building and enhancing the resilience of a manufacturing system. However, current redundancy strategies require duplicated machinery employed either in active or in standby status. This in turn causes extra costs in designing and achieving resilience. Achieving an efficient deployment of the redundant component in the face of failures is also challenging. In this paper, we introduce a novel redundancy strategy, called adaptive standby redundancy (ASR), to achieve resilient performance for discrete manufacturing systems while reducing the cost of employing the duplicated components that are typically used in traditional systems. This novel strategy permits achievement of high levels of utilisation of the system and graceful degradation in case of failure, keeping the system functional. The strategy is then validated in a developed robotic flexible assembly cell (RFAC), which is tested and results on its efficacy and performance enhancement are discussed.

Taxonomy-Driven Graph-Theoretic Framework for Manufacturing Cybersecurity Risk Modeling and Assessment

Abstract Identifying, analyzing, and evaluating cybersecurity risks are essential to devise effective decision-making strategies to secure critical manufacturing against potential cyberattacks. However, a manufacturing-specific quantitative approach is lacking to effectively model threat events and evaluate the unique cybersecurity risks in discrete manufacturing systems. In response, this paper introduces the first taxonomy-driven graph-theoretic model and framework to formally represent this unique cybersecurity threat landscape and identify vulnerable manufacturing assets requiring prioritized control. First, the proposed framework characterizes threat actors’ techniques, tactics, and procedures using taxonomical classifications of manufacturing-specific threat attributes and integrates these attributes into cybersecurity risk modeling. This facilitates the systematic generation of comprehensive and generalizable cyber-physical attack graphs for discrete manufacturing systems. Second, using the attack graph formalism, the proposed framework enables concurrent modeling and analysis of a wide variety of cybersecurity threats comprising varying attack vectors, locations, vulnerabilities, and consequences. The risk model captures the cascading attack impact of varying attack methods through different cyber and physical entities in manufacturing systems, leading to specific consequences. Then, the constructed cyber-physical attack graphs are analyzed to comprehend threat propagation through the discrete manufacturing value chain and identify potential attack paths. Third, a quantitative risk assessment approach is presented to evaluate the cybersecurity risk associated with potential attack paths. It also identifies the attack path with the maximum likelihood of success, pointing out critical manufacturing assets requiring prioritized control. Finally, the proposed risk modeling and assessment framework is demonstrated using an illustrative example.

A new description model for enabling more general manufacturing systems representation in digital twin

The demand for high efficiency and flexible production has led to more complex and flexible production organizations in discrete manufacturing systems. How to rapidly construct/reconfigure a workshop digital twin in response to a configurable complex physical production organization becomes a challenging problem. To address this challenge, based on the existing SevenElement description model, we propose an improved Material Nodes-oriented SevenElements (MNOSE) description model. This model is not only capable of describing/expressing inter-device relationships among manufacturing equipment/devices but also intra-device internal relationships between the device and materials handled by it. Its potential impact on shop-floor digital twin is to make an updating of shop-floor digital twin easier and faster because it can easily express and rapidly (re)-construct a shop-floor digital shadow in response to a rapidly physically reconfigured workshop for a different production organization. In addition, the model can drive the logical unification of multiple digital twin services in a plant and improve the efficiency of integration between several information systems. The utility of the description model in describing shop floor is demonstrated by a case study and validated by modeling representations of typical production organizations.

A DRL-Based Reactive Scheduling Policy for Flexible Job Shops With Random Job Arrivals

In real-life production systems, arrivals of jobs are usually unpredictable, which makes it necessary to develop solid reactive scheduling policies to meet delivery requirements. Deep reinforcement learning (DRL) based scheduling methods are capable of quickly responding to dynamic events by learning from the training data. However, most of policy networks in DRL algorithms are trained to choose priority dispatching rules (PDR), thus, to some extent, the efficiency of obtained scheduling plans is limited by the performance of PDRs. This paper investigates a dynamic flexible job shop scheduling problem with random job arrivals for the total tardiness minimization. A DRL-based reactive scheduling method, proximal policy optimization with attention-based policy network (PPO-APN), is proposed to make real-time decisions for the dynamic scheduling environment, where the attention-based policy network (APN) is able to directly select pending jobs distinguished from the action space that consists of PDRs. Additionally, a global/local reward function (GLRF) is designed to address the reward sparsity issue during training processes. The proposed PPO-APN is tested on randomly generated instances with different production configurations, and is compared with frequently-used PDRs and DRL-based methods. Numerical experimental results indicate that APN and GLRF components significantly improve the training efficiency, and the PPO-APN shows better overall performance compared with other methods. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This work is motivated by a typical production scenario in discrete manufacturing systems, where orders randomly arrive at the shop floor and require to be scheduled in a short time to ensure the on-time delivery. Previous research work tends to apply DRL algorithms to choose suitable dispatching rules for the ease of implementation. Nevertheless, the jobs that can be selected by dispatching rules are rather limited, thus many possible high-quality scheduling plans are ignored. This work first sorts all the unscheduled jobs by a heuristic algorithm, and puts some of top-ranked jobs to a pool. When a machine becomes available, it will directly choose a job from the pool as the next processing task. The job selection policy is represented by a novel attention-based network, and is trained by a powerful DRL algorithm. The aforementioned process is repeatedly executed in a simulation environment to collect the training data. Therefore, after being trained for a certain period of time, the policy will become smarter and can be applied to make right decisions in real-time. The proposed reactive scheduling method has been proved to be more efficient than dispatching rules and DRL-based approaches, and is effective in the production scheduling for a wide variety of discrete manufacturing scenarios, such as automobile and electronics industries. Moreover, the proposed method can be further extended to address dynamic scheduling problems with some production characteristics via adding constraints for the job selection or re-defining calculations for the completion time of operations accordingly.

Dynamic production bottleneck prediction using a data-driven method in discrete manufacturing system

In the complex discrete manufacturing system (DMS), the production bottleneck shifts in space as time goes on and constrains operational efficiency. Accurate proactive production bottleneck prediction provides a reliable basis for dynamic production decisions and helps to improve management timeliness and production efficiency. According to the production characteristics of DMS and the relationship between supply and demand, the production bottleneck is given a new quantification. A long and short-term memory network (LSTM) with dual attention mechanism and a dynamic updating method for the source model are proposed to predict production bottlenecks accurately. Firstly, feature and state attention mechanisms are designed to improve the feature extraction and prediction ability of LSTM. Secondly, as the applicability of the prediction model gradually declines over time, sliding time windows and fast Hoeffding concept detection are combined to trigger the update of model parameters. Then a competitive strategy is explored to choose the source model that is the most suitable for the current data distribution in the model library. Model-based transfer learning is adopted to update the source model parameters, making the prediction model highly adaptive. Subsequently, an elimination strategy is set to update the model library to ensure its timeliness. Finally, experiments demonstrate that the proposed method is effective in bottleneck prediction and superior to other methods.

Two-stage approach to causality analysis-based quality problem solving for discrete manufacturing systems

In discrete manufacturing systems, the efficiency of quality problem solving always matters, but it can decrease continually due to increasing systematic complexity, uncertainty, and fuzzy internal mechanism. These factors often hinder situation awareness, confuse decision-making, and delay response times in quality problem solving. In such context, Bayesian network performs poorly in precisely locating objects to-be-adjusted in complex causal relationship networks and predicting the impact of interventions to-be-implemented. To address this challenge, a two-stage approach to causality analysis-based quality problem solving is proposed. In the first stage, an improved Bayesian network is proposed to identify the likely root causes that have a direct causality on the quality indicator. In the second stage, causal inference is used to estimate the effect of likely root causes on the quality indicator. The proposed approach compensates for Bayesian network in precisely identifying root causes that should have causation rather than correlation with quality problems, and facilitating a quantitative tuning process to create suitable solutions. To assess the effectiveness of the proposed approach, a quality problem solving case in an aerospace shell parts spinning process was conducted. The results demonstrate that the approach can accurately identify likely root causes and determine the appropriate intervention degree.

Multistage Quality Prediction Using Neural Networks in Discrete Manufacturing Systems

The deployment of a manufacturing execution system (MES) holds promising potential in facilitating the accumulation of a substantial amount of inspection data. Low quality levels in discrete manufacturing environments are the result of multi-factor coupling and the failure to detect quality issues promptly in accordance with manufacturing settings, which may trigger the propagation of downstream defects. Currently, most inspection quality methods consist of direct measurements followed by manual judgment. The integration of deep learning methods provides a feasible way in which to identify defects on time, thus improving the acceptance rate of factories. This paper focuses on the design of a data-driven quality prediction and control model, built around discrete manufacturing characteristics, and uses fuzzy theory to evaluate the quality levels of production stages. Furthermore, a multivariate long- and short-term memory sequence model is proposed in order to explore the qualitative information from time domain features. The data regarding the produced water dispensers are validated using three evaluation indices, namely, RMSE, MAE, and MAPE. The results indicate that the multivariate long- and short-term memory model exhibits stronger prediction performance.

Customized quality inspection cycles for achieving sustainable manufacturing in the era of Zero Defect Manufacturing

Manufacturing businesses aim to provide high-quality products at the lowest cost possible. Quality management is crucial for improving productivity, sustainability, competitiveness, and profitability. Quality assurance (QA) methods are used to produce high-quality parts with minimal performance loss. However, traditional QA methods may not meet the needs of contemporary manufacturing standards and market needs. To address this issue, Zero Defect Manufacturing (ZDM) is an approach that can provide greater industrial sustainability. ZDM requires 100% product inspection to ensure that the desired product quality is achieved. However, increasing quality inspection to 100% can be time, cost, or resource-consuming, significantly increasing production costs. The current research work introduces a framework that considers customized product inspection cycles for minimizing the product inspection time and at the same time assuring product quality for discrete manufacturing systems. The proposed framework identifies the parameters, factors and Industry 4.0 technologies that manufacturers need to pay attention at and use to implement the customized quality inspection cycles. The proposed methodology is tested in a real industrial use case coming from the semiconductor domain. A dynamic scheduling tool was utilized for the conduction of the simulations. The results of the simulations revealed the benefits of such a radical method and also the implementation challenges were identified and discussed.

Real-Time Scheduling for Dynamic Partial-No-Wait Multiobjective Flexible Job Shop by Deep Reinforcement Learning

In modern discrete flexible manufacturing systems, dynamic disturbances frequently occur in real time and each job may contain several special operations in partial-no-wait constraint due to technological requirements. In this regard, a hierarchical multiagent deep reinforcement learning (DRL)-based real-time scheduling method named hierarchical multi-agent proximal policy optimization (HMAPPO) is developed to address the dynamic partial-no-wait multiobjective flexible job shop scheduling problem (DMOFJSP-PNW) with new job insertions and machine breakdowns. The proposed HMAPPO contains three proximal policy optimization (PPO)-based agents operating in different spatiotemporal scales, namely, objective agent, job agent, and machine agent. The objective agent acts as a higher controller periodically determining the temporary objectives to be optimized. The job agent and machine agent are lower actuators, respectively, choosing a job selection rule and machine assignment rule to achieve the temporary objective at each rescheduling point. Five job selection rules and six machine assignment rules are designed to select an uncompleted job and assign the next operation of which together with its successors in no-wait constraint on the corresponding processing machines. A hierarchical PPO-based training algorithm is developed. Extensive numerical experiments have confirmed the effectiveness and superiority of the proposed HMAPPO compared with other well-known dynamic scheduling methods. Note to Practitioners—The motivation of this article stems from the need to develop real-time scheduling methods for modern discrete flexible manufacturing factories, such as aerospace product manufacturing and steel manufacturing, where dynamic events frequently occur, and each job may contain several operations subjected to the no-wait constraint. Traditional dynamic scheduling methods, such as metaheuristics or dispatching rules, either suffer from poor time efficiency or fail to ensure good solution quality for multiple objectives in the long-term run. Meanwhile, few of the previous studies have considered the partial-no-wait constraint among several operations from the same job, which widely exists in many industries. In this article, we propose a hierarchical multiagent deep reinforcement learning (DRL)-based real-time scheduling method named HMAPPO to address the dynamic partial-no-wait multiobjective flexible job shop scheduling problem (DMOFJSP-PNW) with new job insertions and machine breakdowns. The proposed HMAPPO uses three DRL-based agents to adaptively select the temporary objectives and choose the most feasible dispatching rules to achieve them at different rescheduling points, through which the rescheduling can be made in real time and a good compromise among different objectives can be obtained in the long-term schedule. Extensive experimental results have demonstrated the effectiveness and superiority of the proposed HMAPPO. For industrial applications, this method can be extended to many other production scheduling problems, such as hybrid flow shops and open shop with different uncertainties and objectives.

EveSyncIAI: Event Synchronization Industrial Augmented Intelligence for Fault Diagnosis

The fault diagnosis of complex timed discrete event systems remains a relevant and challenging problem in industry. As systems degrade, the synchronization required for normal operation associated with timed event sequences deteriorate. However, existing discrete event approaches cannot handle complexity well as there may be potentially hundreds of event sequences that still result in normal system behavior, and they do not scale well due to the state explosion problem. To circumvent these issues, a new data-driven augmented intelligence framework called EveSyncIAI is proposed that uses a Timed Petri Net of the normal process to enable selective extraction of discriminating time delay features, which are then used alongside machine learning for fault diagnosis. The developed methodology is applied to a case study of an operational discrete manufacturing system in the semiconductor industry. Multiclass machine learning approaches as well as binary anomaly detection algorithms with varying levels of supervision are directly compared with and without EveSyncIAI-extracted features. With precision, recall, and F1-score values above 0.90 being achieved for multi-fault classification and AUC scores exceeding 0.98 for semi-supervised and unsupervised anomaly detection approaches, the framework presented is promising for the fault diagnosis of timed sequence-sensitive systems.

Fractional Differential Equations in the Standard Construction Model of the Educational Application of the Internet of Things

Abstract This article proposes a software fusion processing method for multi-sensor monitoring the Internet of Things based on the fractional differential filtering algorithm. The purpose is to solve the difference between the detection data caused by the difference in each production information sampling node’s equipment performance and working environment. The article applies the above algorithm to the remote detection data fusion processing experiment of the discrete manufacturing system of the Internet of Things. The research found that the algorithm realizes the remote high-precision measurement of the Internet of Things sensor information. The fractional integral operator has high fusion accuracy in remote detection data fusion processing.

Calculus Logic Function in Integrated Manufacturing Automation of Single Chip Microcomputer

Abstract First, this paper uses the Lyapunov stability function to construct the time-space implicit calculus equation model of the integrated manufacturing of single-chip microcomputers for the abnormal diffusion phenomenon. Secondly, we use the differential data fusion algorithm modeling of the discrete manufacturing system and introduce the implicit calculus equation modeling method. Finally, the article uses the singular boundary method to give the simulation results. The study found that the implicit calculus equation modeling method is simpler than the existing model, and the physical and statistical concepts are clear.

Energy-Based Prognostics for Gradual Loss of Conveyor Belt Tension in Discrete Manufacturing Systems

This paper presents a data-driven approach for the prognosis of the gradual behavioural deterioration of conveyor belts used for the transportation of pallets between processing workstations of discrete manufacturing systems. The approach relies on the knowledge of the power consumption of a conveyor belt motor driver. Data are collected for two separate cases: the static case and dynamic case. In the static case, power consumption data are collected under different loads and belt tension. These data are used by a prognostic model (artificial neural network (ANN)) to learn the conveyor belt motor driver’s power consumption pattern under different belt tensions and load conditions. The data collected during the dynamic case are used to investigate how the belt tension affects the movement of pallets between conveyor zones. During the run time, the trained prognostic model takes real-time power consumption measurements and load information from a testbench (a discrete multirobot mobile assembling line) and predicts a belt tension class. A consecutive mismatch between the predicted belt tension class and optimal belt tension class is an indication of failure, i.e., a gradual loss of belt tension. Hence, maintenance steps must be taken to avoid further catastrophic situations such as belt slippages on head pulleys, material slippages and belt wear and tear.

A dual energy benchmarking methodology for energy-efficient production planning and operation of discrete manufacturing systems using data mining techniques

Adequate energy awareness is an essential prerequisite for energy-efficient production planning and operation in manufacturing systems. Nonetheless, a lack of awareness of the impact of managerial and operational energy-influencing factors on energy consumption can be observed in practice due to irrational and inexplicable energy benchmarks. To this end, this paper presents a dual energy benchmarking methodology that specifically promotes energy awareness in terms of managerial and operational factors. The presented methodology uses data mining techniques to excavate energy data as well as managerial and operational data, consisting of three steps: (i) data preparation to provide reliable multi-source heterogeneous data; (ii) dynamic energy benchmarking that uses the decision tree algorithm to classify different energy consumption patterns; (iii) load shape benchmarking that interprets energy use behaviors based on load shape features. To demonstrate the effectiveness and practicality, the methodology was implemented in a die-casting workshop. The results showed that a total of nine dynamic energy benchmarks were created primality under the impact of capacity utilization, two of which generated five representative load curves to characterize the workshop's daily energy use behaviors. It was concluded that the methodology could provide practical recommendations on energy-efficient production planning and operation in practical applications.

Dynamic Bottleneck Identification of Manufacturing Resources in Complex Manufacturing System

Bottleneck identification is of great interest in discrete manufacturing fields, as they limit the system’s throughput. However, the bottlenecks are difficult to accurately identify due to the instability and complexity of discrete manufacturing systems. This paper proposes a dynamic bottleneck identification method (DBI-BS) that is based on effective buffers and fine-grained machine states to identify bottlenecks accurately. First, the complex manufacturing system (CMS) with strong coupling between elements is decoupled into several independent parts under the guidance of the effective buffer theory. Then, the machine activity duration method is improved through further fine-grained division, and the machine states are described by the timing flow model. The method to quantify the degree of bottleneck that restricts the system throughput (TH) is proposed on the basis of the turning point theory, and the one-to-one mapping relationship between the simulated and authentic complex manufacturing systems is also studied. Simulation results show that the DBI-BS can effectively identify dynamic bottlenecks in complex manufacturing processes, and the decoupling of complex systems can effectively improve the accuracy of dynamic bottleneck identification.

Streamline 3D simulation model development for virtual commissioning with IEC61499

Current discrete manufacturing systems are characterized by an ever-increasing complexity, demanding for innovative solutions, capable to optimize performances while increasing the resilience and the capability to adapt to production modifications. With such a background, changing perspective to deal with distributed modular architectures of Cyber Physical Systems is mandatory, and the IEC 61499 standard, its object oriented, and event-based approaches promote this paradigm shift. The multi-disciplinary nature of the CPS entities and the possibility to exploit their digital counterparts, paves the way for the development of enhanced decision-support systems like the ones dedicated to Virtual Commissioning (VC). VC supports the automation developer in evaluating the impact of different management strategies, increasing the reliability of the final control applications, while reducing the amount of time to carry on physical tests on the real mechatronic system. However, creating a virtual commissioning model is still a complex and potentially expensive process that needs to be carried out by different professionals who must tightly cooperate to generate an effective playground for the automation testing. We propose a new approach to the design and develop virtual commissioning models, that, leveraging the synergies between modular simulation and IEC 61499 automation technologies, aims at improving the efficiency of the overall process of implementing 3D simulation digital twins for complex automated discrete manufacturing systems. The paper describes an open architecture, composed of reference data models and software API, and presents a proof-of-concept implementation of an integrated engineering platform of VC models.

Bottleneck detection for discrete manufacturing system based on object-oriented colored petri nets and cloud simulation

To identify the bottleneck unit of discrete manufacturing system dynamically, an intelligent detection method of bottleneck (IDMOB) is proposed by combining object-oriented colored Petri nets (OOCPN) and cloud simulation. With the proposed method, the production cells of manufacturing system are abstracted as several Petri net sub-modules. And the messages are passed through mapping functions and message passing gates among Petri net modules. Then, the cloud simulation is created to simulate the real processing, and the production data and capacity function are plugged into the cloud to detect the bottleneck unit intelligently. Finally, the effectiveness of IDMOB method is applied and validated in a practical case study from the power transformer industry.

An advanced model for solving industrial scheduling problems

This paper summarizes an advanced model and a practiceoriented approach to solve scheduling problems of discrete manufacturing systems. An advanced scheduling approach is presented which is able to adapt to the requirements of real-life situations by taking into consideration the specific characteristics of modern manufacturing and assembly systems. These detailed constraints and capabilities of the actual resource environment include the alternative technological routes, the limited available machines, the unrelated processing time, the sequence dependent setup time and the jobs with due dates. An extended flexible job shop scheduling model is defined to solve the resource allocation problems and to create the fine schedule of the execution of jobs, tasks and operations. The paper describes the most important characteristics of the analyzed problem class and shows the main approach of the developed heuristic solving method. Our approach combines a special searching technique based on fast execution-driven simulation with a multi-objective optimization model.

Machine Learning for Diagnosis of Event Synchronization Faults in Discrete Manufacturing Systems

Abstract Common in discrete manufacturing, timed event systems often have strict synchronization requirements for healthy operation. Discrete event system methods have been used as mathematical tools to detect known faults, but do not scale well for problems with extensive variability in the normal class. A hybridized discrete event and data-driven method is suggested to supplement fault diagnosis in the case where failure patterns are not known in advance. A unique fault diagnosis framework consisting of signal data from programmable logic controllers, a Timed Petri Net of the normal process behavior, and machine learning algorithms is presented to improve fault diagnosis of timed event systems. Various supervised and unsupervised machine learning algorithms are explored as the methodology is implemented in a case study in semiconductor manufacturing. State-of-the-art classifiers such as artificial neural networks, support vector machines, and random forests are implemented and compared for handling multi-fault diagnosis using programmable logic controller signal data. For unsupervised learning, classifiers based on principal component analysis utilizing major and minor principal components are compared for anomaly detection. The rule-based random forest and extreme random forest classifiers achieve excellent performance with a precision and recall score of 0.96 for multi-fault classification. Additionally, the unsupervised learning approach yields anomaly detection rates of 98% with false alarms under 3% with a training set 99% smaller than the supervised learning classifiers. These results obtained on a real use case are promising to enable prognostic tools in industrial automation systems in the future.

Real-Time Fault Detection in Discrete Manufacturing Systems Via LSTM Model based on PLC Digital Control Signals

A lot of sensor and control signals is generated by an industrial controller and related internet-of-things in discrete manufacturing system. The acquired signals are such records indicating whether several process operations have been correctly conducted or not in the system, therefore they are usually composed of binary numbers. For example, once a certain sensor turns on, the corresponding value is changed from 0 to 1, and it means the process is finished the previous operation and ready to conduct next operation. If an actuator starts to move, the corresponding value is changed from 0 to 1 and it indicates the corresponding operation is been conducting. Because traditional fault detection approaches are generally conducted with analog sensor signals and the signals show stationary during normal operation states, it is not simple to identify whether the manufacturing process works properly via conventional fault detection methods. However, digital control signals collected from a programmable logic controller continuously vary during normal process operation in order to show inherent sequence information which indicates the conducting operation tasks. Therefore, in this research, it is proposed to a recurrent neural network-based fault detection approach for considering sequential patterns in normal states of the manufacturing process. Using the constructed long short-term memory based fault detection, it is possible to predict the next control signals and detect faulty states by compared the predicted and real control signals in real-time. We validated and verified the proposed fault detection methods using digital control signals which are collected from a laser marking process, and the method provide good detection performance only using binary values.