Multistage Manufacturing System Research Articles

Performance evaluation of multi-stage manufacturing systems operating under feedback and feedforward quality control loops

In manufacturing, the essential product characteristics are often created through multiple stages. Coupling product data obtained through inspection and controllers based on decision models with prediction capabilities enables quality control loops, enhancing both feedback and feedforward mechanisms. This paper proposes a methodology to merge the formulation of feedback and feedforward quality control loops into a performance evaluation model for multi-stage manufacturing systems. This approach evaluates quality control loop impacts system-wide, aiding in configuring and reconfiguring quality gates. A case study illustrates how allocating inspection technologies and efficient decision models improves overall system performance through effective feedback and feedforward control loops.

Resilience-oriented approach of dynamic production and maintenance scheduling optimisation considering operational uncertainty

ABSTRACT One of the main challenges in operating multistate manufacturing systems (MMSs) is maintaining stable and robust production against various disruptions. Therefore, an urgent need exists for an operation and maintenance (O&M) method that optimises MMS resilience, i.e. the capability of withstanding or recover from disruptions of various sources. Consequently, this study proposes an integrated resilience-oriented production and maintenance scheduling approach for MMSs. This approach enhances MMS resilience by reinforcing its adaptivity to the variation in production requirements. Based on a conceptual investigation of operational uncertainty and its mechanism of disruption, this study devotes to (i) formulating a performance loss-based resilience measurement with consideration of operational uncertainties and (ii) proposing a reinforcement learning-based approach to schedule MMS operation and maintenance activities for MMS components of different performance states. An industrial case study of a ferrite phase shifter manufacturing system is subsequently conducted to validate the proposed approach. Results demonstrate the effectiveness of the proposed approach in the resilience optimisation of MMSs.

SysML4GDPSim: A SysML Profile for Modeling Geometric Deviation Propagation in Multistage Manufacturing Systems Simulation

In recent years, paradigms like production quality or zero-defect manufacturing have emerged, highlighting the need to improve quality and reduce waste in manufacturing systems. Although quality can be analyzed from various points of view during different stages of a manufacturing system’s lifecycle, this research focuses on a multidomain simulation model definition oriented toward the analysis of productivity and geometric quality during early design stages. To avoid inconsistencies, the authors explored the definition of descriptive models using system modeling language (SysML) profiles that capture domain-specific semantics defining object constraint language (OCL) rules, facilitating the assurance of model completeness and consistency regarding this specific knowledge. This paper presents a SysML profile for the simulation of geometric deviation propagation in multistage manufacturing systems (SysML4GDPSim), containing the concepts for the analysis of two data flows: (a) coupled discrete behavior simulation characteristic of manufacturing systems defined using discrete events simulation (DEVS) formalism; and (b) geometric deviation propagation through the system based on the geometrical modeling of artifacts using concepts from the topologically and technologically related surfaces (TTRS) theory. Consistency checking for this type of multidomain simulation model and the adoption of TTRS for the mathematical analysis of geometric deviations are the main contributions of this work, oriented towards facilitating the collaboration between design and analysis experts in the manufacturing domain. Finally, a case study shows the application of the proposed profile for the simulation model of an assembling line, including the model’s transformation to Modelica and some experimental results of this type of analysis.

Functional health prognosis approach for multistate manufacturing systems based on quality stochastic flow manufacturing network

Ensuring the predictability of the reliability of the final product is the functional goal of multistate manufacturing system health prognosis. However, the existing studies of basic reliability-oriented health state evaluation ignore the functional input–output characteristics of the dynamic operational manufacturing system. Therefore, a novel functional health prognosis approach considering functional fitness quality based on a quality stochastic flow manufacturing network (QSFMN) is proposed. First, to achieve the specified functional goal, the functional degradation mechanism is expounded from the system’s operational perspective, and based on this, the connotation of functional health considering product fitness quality is further provided. Second, from the perspective of functional fitness quality, through introducing a buffer to reflect dynamic production rhythm, a QSFMN model is proposed based operational quality data to expound the functional relationship between functional requirements, processing machines performance, and output quality. Third, a functional health prognosis method of multistate manufacturing systems based on QSFMN is proposed. Finally, a case study of subway flow receiver manufacturing systems is proposed to verify the effectiveness and advantage of the proposed approach.

Mission reliability-centered opportunistic maintenance approach for multistate manufacturing systems

Mission reliability-centered opportunistic maintenance approach for multistate manufacturing systems

Allocating Robots/Cobots to Production Systems for Productivity and Ergonomics Optimization

Collaboration between humans and robots has great promise in manufacturing systems. The utilization of cobots in a manufacturing system can improve both productivity and ergonomics. In this paper, we study the problem of how to allocate limited cobot/robots to manufacturing systems with multiple workstations so that an integrated performance measure, considering both productivity and ergonomics is optimized. Previous work on cobot/robot allocation in manufacturing systems focus on the decomposition of tasks for a single workstation into multiple work elements, and then split them between human and robots, rather than studying multi-machine systems. To bridge this gap, we consider the allocation of cobots/robots to a multi-stage manufacturing system. Specifically, we establish an integrated performance measure and formulate cobot/robot allocation into a constraint integer programming problem. With this formulation, we obtain the optimal allocation of one available cobot/robot in simulated production systems, based on the integrated performance measure of productivity and ergonomics. Furthermore, the allocation problems of production systems with multiple cobots/robots is considered and solved with a scalable algorithm. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —Collaborative robots and independent robots are increasingly applied to manufacturing production systems. However, how to optimally allocate both types of robots considering both productivity and ergonomics influence has not been well studied. In this article, we established a practical optimization method to allocate cobots/robots to different workstations and split the work between cobot and human in one workstation when there are multiple workstations and a limited number of available cobots/robots in the manufacturing systems. Based on various real-world scenarios, we inferred useful insights for the robot/cobot allocation problem. To deal with the computational load when the number of workstations is large, a scalable optimization algorithm is also adopted. The case study results demonstrated the effectiveness of the proposed approach.

GLHAD: A Group Lasso-Based Hybrid Attack Detection and Localization Framework for Multistage Manufacturing Systems

Abstract As Industry 4.0 and digitization continue to advance, the reliance on information technology increases, making the world more vulnerable to cyberattacks, especially cyber-physical attacks that can manipulate physical systems and compromise sensor data integrity. Detecting cyberattacks in multistage manufacturing systems (MMS) is crucial due to the growing sophistication of attacks and the complexity of MMS. Attacks can propagate throughout the system, affecting subsequent stages and making detection more challenging than in single-stage systems. Localization is also critical due to the complex interactions in MMS. To address these challenges, a group lasso regression-based framework is proposed to detect and localize attacks in MMS. The proposed algorithm outperforms traditional hypothesis testing-based methods in expected detection delay and localization accuracy, as demonstrated in a simple linear multistage manufacturing system.

Mission Reliability Modeling of Multistation Manufacturing Systems Considering Cascading Functional Failure

The core function of the manufacturing system is to output high-quality products at a stable production rhythm. However, there are complex interactions between physical and functional failures of machines in the station. Functional failures can be propagated to other stations through material flow, resulting in the increasing risk of system performance degradation, which deserves further study. Therefore, in this article, a mission reliability modeling method considering the cascading functional failure of multistation manufacturing systems is presented. First, from the perspective of functional output, the cascading failure mechanism of a manufacturing system is discussed, and the connotation of functional failure and mission reliability is proposed. Second, the interaction model between machine performance and product quality is established to analyze the production rhythm considering the imperfect inspection. Third, the cascading functional failure model is constructed based on the mechanism of failure propagation. The integrated mission reliability modeling method of a multistation manufacturing system is proposed. Finally, a case study of a ferrite phase-shifting unit manufacturing system is conducted to illustrate the effectiveness and advancement of the proposed method.

Integrated mission reliability modeling for multistate manufacturing systems considering heterogeneous feedstocks based on extended stochastic flow manufacturing network

Mission reliability is a crucial health indicator for systematically characterizing the operational state and degradation process of manufacturing systems. However, the impact of heterogeneous feedstocks (qualified feedstocks and unqualified feedstocks) on processing machine degradation and the reliability of buffers have not received much attention in the existing research on mission reliability modeling of manufacturing systems. Accordingly, this study proposes a novel integrated mission reliability modeling method for multistate manufacturing systems considering heterogeneous feedstocks. Specifically, based on the operational mechanism of a multistate manufacturing system considering heterogeneous feedstocks and the connotation of the integrated mission reliability of the manufacturing system, an extended stochastic flow manufacturing network (ESFMN) model is constructed to quantify the interaction among processing machines, inspection machines, buffers, and heterogeneous feedstocks. By analyzing the mixture degradation of the processing machine under the impact of heterogeneous feedstock, an integrated mission reliability model for a multistate manufacturing system considering heterogeneous feedstocks based on ESFMN is established. Finally, an industrial example of the valve spool manufacturing system is given to illustrate the effectiveness and superiority of the proposed method.

Stochastic deep Koopman model for quality propagation analysis in multistage manufacturing systems

The modeling of multistage manufacturing systems (MMSs) has attracted increased attention from both academia and industry. Recent advancements in deep learning methods provide an opportunity to accomplish this task with reduced cost and expertise. This study introduces a stochastic deep Koopman (SDK) framework to model the complex behavior of MMSs. Specifically, we present a novel application of Koopman operators to propagate critical quality information extracted by variational autoencoders. Through this framework, we can effectively capture the general nonlinear evolution of product quality using a transferred linear representation, thus enhancing the interpretability of the data-driven model. To evaluate the performance of the SDK framework, we carried out a comparative study on an open-source dataset. The main findings of this paper are as follows. Our results indicate that SDK surpasses other popular data-driven models in accuracy when predicting stagewise product quality within the MMS. Furthermore, the unique linear propagation property in the stochastic latent space of SDK enables traceability for quality evolution throughout the process, thereby facilitating the design of root cause analysis schemes. Notably, the proposed framework requires minimal knowledge of the underlying physics of production lines. It serves as a virtual metrology tool that can be applied to various MMSs, contributing to the ultimate goal of Zero Defect Manufacturing.

A Model Review on Joint Optimization of Part Quality Inspection Planning, Buffer Allocation, and Preventive Maintenance in SMMS

In serial multi-stage manufacturing systems (SMMS), optimization of part quality inspection planning (PQIP), buffer allocation problem (BAP), and preventive maintenance (PM), individually and jointly, is attracting researchers’ attention. The model formulation for complicated manufacturing systems and the previously mentioned joint decisions is very beneficial given the interdependencies between the various manufacturing functions. As a result, this paper evaluates the literature on joint optimization of the multi-stage serial production system. The literature is classified based on the decision variables basis to represent each manufacturing function [inspection sample size and allocation (PQIP), buffer sizing and allocation (BAP), and preventive maintenance scheduling (PM)], and a general example model is presented in each classification, with a summary of recent studies, solution methods, research gaps, and future research recommendations. In the integrated models, almost all the studies considered only two functions, with that it is worth noting that research into the optimization of over two functions is still in its beginning. Furthermore, most studies neglected many of the real industrial settings that should also be integrated into the model. And finally, there was no specific solution technique recommended in the literature, yet a general simulation optimization method was used to generate and evaluate the combinatorial complex joint models.

Mission reliability-driven risk-based predictive maintenance approach of multistate manufacturing system

Mission reliability-driven risk-based predictive maintenance approach of multistate manufacturing system

Hybrid opportunistic maintenance policy for serial-parallel multi-station manufacturing systems with spare part overlap

Hybrid opportunistic maintenance policy for serial-parallel multi-station manufacturing systems with spare part overlap

Hierarchical-clustering-based joint optimization of spare part provision and maintenance scheduling for serial-parallel multi-station manufacturing systems

Both the spare part provision and the maintenance scheduling are important in ensuring the production of the Serial-Parallel Multi-Station System (SPMMS). In the multi-specification and small-batch production, the production plan, which renews batch by batch, leads to variable and uncertain working conditions. The existing Dynamic Opportunistic Maintenance (DOM), which separately optimizes each cycle, ignores the aftereffects of both the spare part provision and the Preventive Maintenance (PM) on the following PM scheduling. This paper studies the dynamic joint optimization of spare part provision and maintenance scheduling for the SPMMS, where the aftereffects of the information are especially investigated. During the failure rate modelling, the aftereffect of the historical working conditions is described with the accelerated failure-time model. In the system-level maintenance modelling, a Hierarchical-Clustering-Based Opportunistic Maintenance (HCBOM) policy, which uses a non-parametrical clustering, is introduced to incorporate the aftereffects of the PM. The case study and the policy comparisons verify the above-mentioned aftereffects. The proposed HCBOM is more cost-effective than the tradition Opportunistic Maintenance (OM) frame, the DOM, the group maintenance and the station-level maintenance. From the perspective of clustering, this work provides an innovative OM policy, which can guide practitioners to formulate cost-effective PM scheme without time-expensive calculations.

Production quality prediction of multistage manufacturing systems using multi-task joint deep learning

A multistage manufacturing system with multiple manufacturing stages is the key and main production mode for enterprises to achieve lean production. Due to the variation propagation between stages and multiple related quality prediction tasks, it is difficult to accurately predict the quality of multistage manufacturing systems with multiple tasks. Traditional single stage and single task quality prediction methods permit the multi stages and multiple tasks separately, which ignore multi-stage effects or the quality-related relationship between multiple quality output indicators and reduce the efficiency of quality prediction. In this paper, a production quality prediction framework based on multi-task joint deep learning is proposed to simultaneously evaluate the multi-task quality of all stages in a multistage manufacturing system. To be specific, variation propagation cumulative impact between multiple manufacturing stages is innovatively expressed by designing a multi-scale convolutional neural network with control gates (MCNN-CG) to extract and propagate data features. Production quality with multi-tasks at all stages is jointly predicted by designing a multi-layer multi-gate mixture-of-experts multi-task (ML-MMoE) model with reducing multi-task predictive loss simultaneously. The soft parameter-sharing strategy and multi-gate attention strategy are separately designed to ensure information sharing while learning personalized features of tasks to improve quality prediction accuracy of each task. In addition, a loss function based on homoscedastic uncertainty and regularization is designed to automatically learn the weight between multi-stage and multi-task losses. Experiments on multistage assembly test data of an inertial navigation manufacturing system show that the proposed method performs better than traditional models. Compared to the single-stage model, the proposed multistage model has an average improvement of 8.5%, 20.0% and 23.3% in R2, MAE and RMSE respectively in the second stage. Compared with the traditional multi-stage model, the proposed model has an average improvement of 1.7%, 6.2% and 9.8% in R2, MAE and RMSE respectively in the second stage.

Digitalization platform for data-driven quality management in multi-stage manufacturing systems

AbstractDigital transformation is driving the current technological trends in manufacturing. An integral constituent is a communication between machines, between machines and humans, or between machines and products. This extensive communication involves large volumes of data. Many manufacturers apply data analytics (e.g., for quality management or improvement purposes) to translate the data into a business value. However, isolated, rigid, and area-specific IT solutions often carry this out. Today’s complex manufacturing requires quality management approaches that constitute a holistic view of and understanding of process–product interactions along the process chain instead of focusing solely on single processes. A novel platform approach to support quality management in manufacturing systems is proposed in this paper to overcome this deficit. It integrates state-of-the-art concepts of IT with modeling approaches for planning and operation of quality management. A conceptual framework and the technical architecture for implementing a digitalization platform are presented in this regard. Moreover, the approach is validated and implemented within a web application based on a use case of data-driven quality management in electronics production.

Joint optimization of fleet-level sequential selective maintenance and repairpersons assignment for multi-state manufacturing systems

Due to limited maintenance resources, the joint optimization of fleet-level selective maintenance and repairpersons assignment aims to determine the optimal subset of maintenance actions for each repairperson to ensure that manufacturing systems in the fleet complete subsequent missions. However, related studies ignore the uncertain maintenance duration and flow dependency (a stochastic dependency caused by dynamic material flow) resulting in overestimation of system reliability and total profit. Therefore, this paper proposes a novel joint optimization model of fleet-level sequential selective maintenance and repairpersons assignment under flow dependency and uncertain maintenance duration. The model requires identifying a subset of maintenance actions, assigning the selected maintenance actions to repairpersons, and planning sequence of maintenance actions executed by each repairperson. Moreover, the degradation model of multi-state machines under flow dependency is established by analyzing dynamic characteristics of production processes. Further, the reliability of multi-state manufacturing systems considering the interaction among repairpersons, flow dependency, and uncertain maintenance duration is incorporated into the optimization model. The objective is to maximize the total profit under predetermined system reliability thresholds. Then, the optimization problem is solved by a tailored genetic algorithm with segment-based evolution strategy. Finally, numerical examples are given to verify the effectiveness of the proposed method.

Optimal replacement policy for multi-state manufacturing system with economic and resource dependence under epistemic uncertainty

This paper develops an optimal replacement policy V* for a multi-state manufacturing system. The manufacturing system would be repaired imperfectly once its performance cannot meet the production demand, and would be replaced when the production demand is not met for the V*-th time. Due to imprecise state assignments and unpredictable external working conditions, the performance and transition intensity of the multi-state machine cannot be accurately identified and then inevitably lead to epistemic uncertainty. In addition, the economic dependence and resource dependence that prevailed in the manufacturing system should be considered. In this paper, economic dependence is described as the time and cost saved by simultaneously repairing multiple identical machines, and resource dependence is caused by finite capacity buffers. To take these into account, the fuzzy Markov model and fuzzy stochastic flow manufacturing network (FSFMN) are tailored to evaluate the fuzzy reliability of machines and manufacturing systems, respectively. To obtain the optimal replacement policy V*, we derive the expression of the long run fuzzy profit rate under epistemic uncertainty. The replacement policy is demonstrated on the ferrite phase shifting unit manufacturing system, and the results of the subsequent comparative study and sensitivity analysis show that this policy is more effective.

Inspection strategies for quality products with rewards in a multi-stage production

In a multi-stage manufacturing system, defective components are generated due to deteriorating machine parts and failure to install the feed load. In these circumstances, the system requires inspection counters to distinguish imperfect items and takes a few discreet decisions to produce impeccable items. Whereas the prioritisation of employee appreciation and working on reward is one of the important policies to improve productivity. Here we look at the multi-stage manufacturing system as an M/PH/1 queue model and rewards are given for using certain inspection strategies to produce the quality items. A matrix analytical method is proposed to explain a continuous-time Markov process in which the reward points are given to the strategy of inspection in each state of the system. By constructing the value functions of this dynamic programming model, we derive the optimal policy and the optimal average reward of the entire system in the long run. In addition, we obtain the percentage of time spent on each system state for the probability of conformity and non-conformity of the product over the long term. The results of our computational experiments and case study suggest that the average reward increases due to the actions are taken at each decision epoch for rework and disposal of the non-conformity items.

Multiple degradation-driven preventive maintenance policy for serial-parallel multi-station manufacturing systems

Machine reliability degradation is common in production systems; besides, machine production rate (PR) degradation also widely occurs as the systems deteriorate. The PR has an essential impact on the capacity of the systems. However, the few efforts involving the capacity requirements in preventive maintenance (PM) policies have not yet taken PR degradation into account, assuming all the PRs are constant. It is necessary for PM policies to consider the dynamic capacity caused by the deteriorating PR. With the two types of degradation involved, a multiple degradation-driven preventive maintenance (MDPM) policy for the serial-parallel multi-station manufacturing systems (SMMS) is proposed. Since PM has an impact on PR each time it is executed, an evolution model of PR, with the imperfect maintenance effect considered, is developed. The capacity efficiency based on deteriorating PR is put forward. By maximizing capacity efficiency and minimizing cost rate, a station-level PM optimization model is proposed. Then, the process of the MDPM scheduling is formulated, where three rules, including minimum remaining time, maintenance time window and capacity balance, are suggested for the system-level decision. Finally, the case study is illustrated to show the application of the PR evolution model and the advantages of the proposed MDPM policy.