Pith. sign in

REVIEW 4 major objections 5 minor 39 references

Digital Twin-based Smart Manufacturing: Dynamic Line Reconfiguration for Disturbance Handling

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A digital-twin framework detects a slowed worker and locally re-assigns tasks, keeping a battery line's bottleneck at 43.9 seconds under both tested disturbances and preventing 26% and 63% throughput drops.

desk verdict A genuinely integrated reconfiguration framework with real factory data, but the headline throughput recovery rests on a simulation that omits the very movement costs the paper admits matter. read the letter →

arxiv 2506.07332 v1 pith:MP4LFFHJ submitted 2025-06-09 cs.MA

Bo Fu , Mingjie Bi , Shota Umeda , Takahiro Nakano , Youichi Nonaka , Quan Zhou , Takaharu Matsui , Dawn M. Tilbury
show 1 more author
Kira Barton
This is my paper
classification cs.MA MSC 90C1190B30
keywords digitaltwinmanufacturinglinereconfigurationdisturbancehandlingmixed-integerlinearprogramdiscreteeventsimulationmulti-agentsystemthroughputrecoveryontologymodel
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a manufacturing line can absorb operation-time disturbances, such as a worker slowing to 150% or 300% of normal speed, without a global redesign: a digital twin detects the disturbance, an ontology supplies which agents can do which operations, and a mixed-integer optimizer locally reassigns tasks to neighboring agents or adds one extra agent. A simulation program generated automatically from the optimizer's output evaluates each candidate configuration about 400 times faster than real time. In the battery-line case study, the selected plans keep the bottleneck at the original 43.9 seconds, and throughput in 16 hours stays near 1290 products instead of dropping by 26% or 63%. The authors claim the framework generalizes to any line whose work is a fixed sequence of operations performed by workers, robots, or machines.

What carries the argument

The load-bearing mechanism is the mixed-integer linear program for reconfiguration. It minimizes a weighted sum of expected bottleneck time, number of agent usages, and deviation from the original assignment (Eq. 9), subject to assignment, capability, and contiguous-usage constraints. The novel part is the continuous adjustment variable $\Delta x_{kj}$, which is nonzero only for operations in the disturbed set and for the disturbed or adjacent agents; it allows partial operation sharing, and constraints (10)-(14) scope the change so the rest of the line stays untouched. The second mechanism is the automatic discrete-event simulation generator, which turns the optimizer's assignment into a simulation model and reports bottleneck time and throughput for the configuration selector.

What would settle it

Measure the real station times when the scenario-2 configuration-switch plan is executed: time the neighbor worker's movement between the two stations and the added agent's setup. If the measured average time for any station exceeds the simulated 43.9-second bottleneck, the claimed throughput recovery does not transfer to the physical line.

Watch

Extended reading notes

Core claim

The central claim is that scoped reconfiguration, which modifies only the operations of the disturbed agent and its neighbors, is enough to fully recover throughput in both tested disturbance scenarios. The optimizer's key flexibility is a continuous adjustment variable that lets an operation be split between the disrupted agent and an adjacent agent, so a neighbor can take over part of a task rather than all of it. In the larger disturbance, the selected plan redistributes 57% of one operation to a neighboring worker and moves three other operations to a newly added station, restoring the original 43.9-second bottleneck. The paper reports the reconfiguration optimization runs in 0.03 seconds on average, and that two baseline formulations, one MILP without task sharing and one agent-based rescheduling method, produce higher bottleneck times and lower throughput.

Load-bearing premise

That a task reassigned to a neighboring or newly added agent takes the same average operation time as in the original layout, with zero additional time for moving between stations, setup, or spatial interference.

Editorial extensions

If this is right

  • If the framework is right, a line can respond to a single-agent slowdown in under a second of computation, deciding between a plan switch (no new agents) and a configuration switch (adding one agent).
  • The partial-sharing flexibility is what beats the baselines: the MILP without sharing leaves the larger disturbance at a 55.0-second bottleneck, while sharing plus one added agent restores the original 43.9 seconds.
  • Because simulation generation is automatic and runs at 400x real time, many candidate configurations can be screened before touching the factory floor.
  • The same architecture should transfer to other lines whose work is a fixed operation sequence performed by workers, robots, or machines, provided the ontology entries are populated.
  • The optimizer's scoped search preserves most of the original assignment, so reconfiguration acts as a small local edit rather than a new line balance.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If movement and setup costs turn out to be material, the MILP could be extended by adding a distance-dependent penalty to the objective; the paper's own Sec. IV-D flags this as future work.
  • The continuous partial-assignment trick could be reused in other rebalancing problems, such as dynamic allocation in human-robot assembly cells where tasks are divisible.
  • A direct testable extension would be to repeat the case study with the movement-time penalty included and compare the resulting bottleneck with the 43.9-second claim; if the penalty is small, the claim survives.
  • The framework's monitoring-trigger threshold needs field tuning, because a too-low threshold could cause frequent, unnecessary reconfigurations, a trade-off the paper acknowledges.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes an integrated digital-twin-based framework for dynamically reconfiguring manufacturing lines in response to disturbances such as worker slowdowns. The framework combines a system process digital twin for disturbance monitoring and triggering, a capability-based ontology model for resource representation, a mixed-integer linear programming (MILP) configuration optimizer that generates Pareto-optimal reconfiguration plans, and an automatic SimEvents-based simulation generator that evaluates candidate configurations. The method is demonstrated on a battery production line with 51 operations and 40 available agents across 8 agent types. In two disturbance scenarios (Worker2 slowed by 50% and by 200%), the authors report that the selected reconfiguration plans maintain the original 43.9 s bottleneck time (or nearly so), recovering throughput from 962 to 1292 products per 16 hours in scenario 1 and from 480 to 1275 products per 16 hours in scenario 2, thereby preventing the 26% and 63% throughput drops that would occur without reconfiguration. The optimizer is reported to run in 0.03 s on average for reconfiguration, and the simulation runs at approximately 400 times real time.

Significance. The paper addresses a relevant problem in reconfigurable manufacturing systems and contributes a full pipeline from disturbance detection to optimized reconfiguration to automated simulation evaluation. A notable strength is that the configuration-switch result achieves the theoretical lower bound of 43.9 s imposed by the indivisible operation 38, which provides an internal consistency check on the optimizer. The reported computational efficiency (0.03 s reconfiguration optimization and 400x real-time simulation) is practically appealing for dynamic decision-making. However, the central claims of throughput recovery are established only in a simulation that abstracts physical motions and reuses the same per-operation time distributions for all agents, and the paper itself acknowledges that movement between stations could introduce additional time costs. The lack of uncertainty quantification in the throughput figures and the incomplete description of baseline modifications further limit the strength of the evaluation. The framework is a useful systems integration, but the physical-line claims require additional validation or significant qualification.

major comments (4)
  1. [Section III-D2, Eq. (9) and constraints (5)-(8)] The text states that during reconfiguration, the disrupted agent and its neighboring agents are allowed to perform multiple non-contiguous operations, but the formulation explicitly reuses constraint (8), which forces z_k = sum_j y_kj <= 1. An assignment in which an agent performs, for example, operations 19-22 and then operation 25 (as implied by Fig. 10 and the discussion in Sec. IV-D) yields z_k = 2 and violates (8). As written, the MILP cannot produce the non-contiguous assignments that the reported reconfiguration plans depend on. The authors should either replace (8) with a constraint that permits multiple contiguous segments for the affected agents, or clarify that the implemented model differs from the written one. Without this correction, the optimizer outputs in Table III are not reproducible from the stated formulation.
  2. [Section III-E and Section IV-D] The simulation 'abstracts physical motions,' and the per-operation time distributions T_kj are applied unchanged to any agent that takes over a redistributed operation. The paper itself concedes in Sec. IV-D that 'allowing partial task assignment during reconfiguration increases flexibility but could introduce additional time costs due to the movement between stations' and that additional buffers would be needed before operation 25. In the selected configuration-switch plan, Worker3 alternates between stations 6 and 8, handling 57% of operation 25 in addition to its own station 8 tasks. Walking between stations, setup/retooling, and buffer-induced waiting are therefore not modeled. Because the headline claim is prevention of 26% and 63% throughput drops on the physical battery line, the simulated bottleneck times and throughputs in Table III do not yet establish the real-line benefit. The authors should add a sensitivity analysis or enhanced simulation that accounts for movement and setup costs, or carefully restate the conclusions as applying to the simulation-level evaluation only.
  3. [Section III-D, Eq. (10), and Section III-E] The optimizer allows continuous fractional assignments through Delta_x_kj in Eq. (10), and Fig. 10 reports a 57% sharing of operation 25. However, the discrete-event simulation model described in Sec. III-E treats each operation as a module with sampled operation times and does not explain how a fractional assignment is realized in a flow line. The paper does not state whether parts are split into batches, whether parallel stations are created, or whether an operation can be preempted and resumed. Without this mapping, the simulated throughput values are not traceable to the optimizer's fractional solution, and the feasibility of the plan on a real line remains unclear. A concrete implementation mechanism for fractional tasks is needed.
  4. [Section IV-D and Table III] The comparison with baseline methods is weakened by the statement that 'both baseline methods were modified to approximate the operational context of our study,' without specifying the modifications. The MILP-based and agent-based baselines are drawn from refs. [12], [13], [16] and [8], but the exact adaptations are not described, so the reported improvements in Table III are not reproducible. In addition, Table III gives only point estimates for throughput without confidence intervals or the number of simulation replications; the t-test mentioned in the text applies only to bottleneck time. The authors should report simulation variance and the precise baseline modifications to support the claimed performance advantage.
minor comments (5)
  1. [Fig. 10 caption] The caption says 'the worker at station 7 (red) delayed its operations,' but the disturbance scenario in Sec. IV-C concerns Worker2, whose adjacent stations are described as stations 5, 7, and 8. This appears to be a typo; station 6 is the likely intended station.
  2. [Section III-D, Eq. (9)] The objective includes the term Cx * sum |x_kj + Delta_x_kj - x0_kj|, which is not linear as written. The authors do not explain how this absolute value is linearized for the MILP solver; adding the standard linearization variables and constraints would improve reproducibility.
  3. [Section III-B] The disturbance detection threshold is a free parameter, but the case study does not state its value or the procedure used to select it. The text correctly mentions anomaly classification methods [34] but gives no implementation details, making the triggering behavior difficult to reproduce.
  4. [Section IV-B] The claim that the simulation runs at 'around 400x real-time speed' is reported for a single hardware setup (Apple M1 laptop) without specifying the number of replications, simulation warm-up, or runtime measurement methodology. A brief benchmark description would strengthen this claim.
  5. [Section IV-E] The scalability discussion is qualitative and based on a single problem instance. Statements such as 'the reconfiguration optimization ... has proven scalable for larger problems' are not supported by experiments and could be rephrased as expectations rather than proven properties.

Circularity Check

0 steps flagged · score 1.0 of 10

No load-bearing circularity: the reconfiguration throughput results are direct outputs of the optimizer and simulator, and the self-citations support infrastructure rather than the central claim.

full rationale

The paper's central demonstration, the throughput recovery in Table III, is produced by an optimization-simulation pipeline rather than by fitting or by definition. The optimizer minimizes expected bottleneck time using operation-time inputs T_kj from the ontology (Eqs. (9)-(14)), and the SimEvents simulator evaluates the resulting configurations using the same operation-time distributions. This is a consistent modeling pipeline, not a circular reduction: the simulated throughput values (1292 and 1275) are not assumed in the optimizer, and the comparison against no-reconfiguration and baseline configurations is an independent counterfactual computation within the model. The claim that the selected plans maintain the original 43.9 s bottleneck is the optimizer reaching the lower bound set by operation 38's time, which is a legitimate optimization result rather than a definitional equivalence. The paper does cite prior work by overlapping authors ([9], [33]) for the digital twin and ontology components, and it explicitly says these are leveraged rather than newly introduced; these citations are supporting infrastructure for monitoring and data retrieval, not the load-bearing demonstration of reconfiguration performance. The acknowledged limitation in Sec. IV-D, that partial task assignment could introduce additional movement time costs, is a real threat to physical validity but is explicitly flagged by the authors and is not a circularity: the simulation abstracts physical motions by design, and the paper does not claim those costs were measured. No equation is equivalent to its own input, no fitted parameter is renamed as a prediction, and no uniqueness theorem is imported from the authors' prior work. The appropriate finding is therefore no significant circularity, with only a minor non-load-bearing reliance on self-citations.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central demonstration depends on trusting the operation time distributions and the SimEvents model to represent the real line, and on assuming redistributed tasks have the same operation times. These are domain assumptions rather than fitted parameters. The optimizer weights and detection threshold are hand-set tuning values, so they are listed as free parameters.

free parameters (2)
  • Objective weights Ct, Cz, Cx = Ct varied 0 to 1, Cz = 1 - Ct; Cx not reported
    Trade-off weights between bottleneck time, number of agents used, and plan-change penalty; chosen by hand to generate Pareto configurations and the two reconfiguration modes, not fitted to data.
  • Disturbance detection threshold = Not specified
    The SDT triggers reconfiguration when operation-time deviation exceeds a preset threshold; the paper defers threshold selection to anomaly classification methods, and the case study does not report a concrete value.
assumptions (4)
  • domain assumption Operation times are known random distributions and independent of agent assignment.
    The optimizer uses E(Tkj) and the simulation samples from distributions; this ignores learning, fatigue, and movement effects. Invoked in Sec. III-B and III-E.
  • ad hoc to paper Partial and non-contiguous task assignments are feasible for disturbed and adjacent agents.
    Constraint relaxation in Sec. III-D.2 (continuous Delta xkj) enables task sharing; physical feasibility is assumed and only discussed qualitatively in Sec. IV-D.
  • domain assumption The SimEvents model faithfully represents real line behavior.
    All throughput and bottleneck claims come from simulation; no real-world validation is provided, though Sec. V lists testbed validation as future work.
  • domain assumption The manufacturing line is a flow line with a fixed operation sequence.
    The paper assumes operations are sequential and that reconfiguration does not change the operation sequence (Sec. III-A.8).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Digital Twin-based Smart Manufacturing: Dynamic Line Reconfiguration for Disturbance Handling." pith.science (2026). https://pith.science/paper/MP4LFFHJ

@misc{pith2026250607332,
  author       = {Pith},
  title        = {Pith review of: Digital Twin-based Smart Manufacturing: Dynamic Line Reconfiguration for Disturbance Handling},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/MP4LFFHJ}},
  note         = {Machine review of arXiv:2506.07332}
}
read the original abstract

The increasing complexity of modern manufacturing, coupled with demand fluctuation, supply chain uncertainties, and product customization, underscores the need for manufacturing systems that can flexibly update their configurations and swiftly adapt to disturbances. However, current research falls short in providing a holistic reconfigurable manufacturing framework that seamlessly monitors system disturbances, optimizes alternative line configurations based on machine capabilities, and automates simulation evaluation for swift adaptations. This paper presents a dynamic manufacturing line reconfiguration framework to handle disturbances that result in operation time changes. The framework incorporates a system process digital twin for monitoring disturbances and triggering reconfigurations, a capability-based ontology model capturing available agent and resource options, a configuration optimizer generating optimal line configurations, and a simulation generation program initializing simulation setups and evaluating line configurations at approximately 400x real-time speed. A case study of a battery production line has been conducted to evaluate the proposed framework. In two implemented disturbance scenarios, the framework successfully recovers system throughput with limited resources, preventing the 26% and 63% throughput drops that would have occurred without a reconfiguration plan. The reconfiguration optimizer efficiently finds optimal solutions, taking an average of 0.03 seconds to find a reconfiguration plan for a manufacturing line with 51 operations and 40 available agents across 8 agent types.

Figures

Figures reproduced from arXiv: 2506.07332 by the authors.

Figure 1
Figure 1. If Worker2 spends more time completing its operations, [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 1
Figure 1. An example manufacturing line operating under a disturbance due to a slower worker. There are 8 stations in the figure, with the numbers on the [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. System diagram of the dynamic reconfiguration framework. The blue [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figures from the paper (8 more)
Figure 3
Figure 3. Figure 3: A part of the ontology graph. An operation needs several capabilities [PITH_FULL_IMAGE:figures/full_fig_p006_3.png]
Figure 4
Figure 4. Figure 4: A graphical model for a matching problem with 51 operations and 40 [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: A battery manufacturing line. throughput, then the selector chooses the line plan that uses the fewest agents. Compared to the configuration optimizer, the selector uses more detailed performance metrics that become available through the simulation environment to deter…
Figure 8
Figure 8. Figure 8: Initial configuration: The mean station and operation times according [PITH_FULL_IMAGE:figures/full_fig_p010_8.png]
Figure 6
Figure 6. Figure 6: An illustration of the system process digital twin and its connection [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: Trade-off between bottleneck time and the number of used agents. [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 9
Figure 9. Figure 9: The time distribution for Worker2 to complete operation 25 when [PITH_FULL_IMAGE:figures/full_fig_p011_9.png]
Figure 10
Figure 10. Figure 10: Configuration switch, disturbance scenario 2: the mean station [PITH_FULL_IMAGE:figures/full_fig_p012_10.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

39 extracted references · 39 canonical work pages

  1. [12]

    Workforce planning and production scheduling in a reconfigurable manufacturing system facing the covid-19 pandemic,

    B. Vahedi-Nouri, R. Tavakkoli-Moghaddam, Z. Hanz ´alek, and A. Dol- gui, “Workforce planning and production scheduling in a reconfigurable manufacturing system facing the covid-19 pandemic,” Journal of Man- ufacturing Systems, vol. 63, pp. 563–574, 2022

    work page 2022

  2. [13]

    Mixed integer programming models for concurrent configuration design and scheduling in a reconfigurable manufacturing system,

    J. Dou, C. Su, and X. Zhao, “Mixed integer programming models for concurrent configuration design and scheduling in a reconfigurable manufacturing system,” Concurrent Engineering, vol. 28, no. 1, pp. 32– 46, 2020

    work page 2020

  3. [16]

    An optimi- sation model for the dynamic management of cellular reconfigurable manufacturing systems under auxiliary module availability constraints,

    M. Bortolini, E. Ferrari, F. G. Galizia, and A. Regattieri, “An optimi- sation model for the dynamic management of cellular reconfigurable manufacturing systems under auxiliary module availability constraints,” Journal of manufacturing systems , vol. 58, pp. 442–451, 2021

    work page 2021

  4. [8]

    Dynamic distributed decision-making for resilient resource real- location in disrupted manufacturing systems,

    ——, “Dynamic distributed decision-making for resilient resource real- location in disrupted manufacturing systems,” International Journal of Production Research, vol. 62, no. 5, pp. 1737–1757, 2024

    work page 2024

  5. [1]

    Agent-based distributed manufacturing control: A state- of-the-art survey,

    P. Leit ˜ao, “Agent-based distributed manufacturing control: A state- of-the-art survey,” Engineering applications of artificial intelligence , vol. 22, no. 7, pp. 979–991, 2009

    work page 2009

  6. [2]

    A distributed approach for agile supply chain decision-making based on network attributes,

    M. Bi, D. M. Tilbury, S. Shen, and K. Barton, “A distributed approach for agile supply chain decision-making based on network attributes,” IEEE Transactions on Automation Science and Engineering , vol. 21, no. 3, pp. 2223–2236, 2023

    work page 2023

  7. [3]

    Design of reconfigurable manufacturing systems,

    Y . Koren and M. Shpitalni, “Design of reconfigurable manufacturing systems,” Journal of manufacturing systems , vol. 29, no. 4, pp. 130– 141, 2010

    work page 2010

  8. [4]

    Reconfigurable manufacturing systems: Principles, design, and future trends,

    Y . Koren, X. Gu, and W. Guo, “Reconfigurable manufacturing systems: Principles, design, and future trends,” Frontiers of Mechanical Engineer- ing, vol. 13, pp. 121–136, 2018

    work page 2018

Show all 39 references
  1. [5]

    Re- configurable manufacturing systems from an optimisation perspective: a focused review of literature,

    A. R. Yelles-Chaouche, E. Gurevsky, N. Brahimi, and A. Dolgui, “Re- configurable manufacturing systems from an optimisation perspective: a focused review of literature,” International Journal of Production Research, vol. 59, no. 21, pp. 6400–6418, 2021

    2021

  2. [6]

    Digital twins-based flexible operating of open architecture production line for individualized manufacturing,

    J. Leng, Z. Chen, W. Sha, Z. Lin, J. Lin, and Q. Liu, “Digital twins-based flexible operating of open architecture production line for individualized manufacturing,” Advanced Engineering Informatics , vol. 53, p. 101676, 2022

    2022

  3. [7]

    Dynamic resource allocation using multi-agent control for manufacturing systems,

    M. Bi, I. Kovalenko, D. M. Tilbury, and K. Barton, “Dynamic resource allocation using multi-agent control for manufacturing systems,” IFAC- PapersOnLine, vol. 54, no. 20, pp. 488–494, 2021

    2021

  4. [9]

    An integrated framework for dynamic manufacturing planning to obtain new line configurations,

    L. Poudel, I. Kovalenko, R. Geng, M. Takaharu, Y . Nonaka, N. Takahiro, U. Shota, D. M. Tilbury, and K. Barton, “An integrated framework for dynamic manufacturing planning to obtain new line configurations,” in 2022 IEEE 18th International Conference on Automation Science and ...

    2022

  5. [10]

    Toward an automated learning control architecture for cyber-physical manufacturing systems,

    I. Kovalenko, J. Moyne, M. Bi, E. C. Balta, W. Ma, Y . Qamsane, X. Zhu, Z. M. Mao, D. M. Tilbury, and K. Barton, “Toward an automated learning control architecture for cyber-physical manufacturing systems,” IEEE Access, vol. 10, pp. 38 755–38 773, 2022

    2022

  6. [11]

    Intelligent scheduling and reconfiguration via deep reinforcement learning in smart manufacturing,

    S. Yang and Z. Xu, “Intelligent scheduling and reconfiguration via deep reinforcement learning in smart manufacturing,” International Journal of Production Research, vol. 60, no. 16, pp. 4936–4953, 2022

    2022

  7. [14]

    Mini- mizing task reassignments under balancing multi-product reconfigurable manufacturing lines,

    A. R. Yelles-Chaouche, E. Gurevsky, N. Brahimi, and A. Dolgui, “Mini- mizing task reassignments under balancing multi-product reconfigurable manufacturing lines,” Computers & Industrial Engineering , vol. 173, p. 108660, 2022

    2022

  8. [15]

    Manufacturing line design configuration with optimized resource groups,

    T. Nakano, K. Daiki, H. Chen, I. Kovalenko, E. Balta, Y . Qamsane, and K. Barton, “Manufacturing line design configuration with optimized resource groups,” in 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE) . IEEE, 2021, pp. 625– 632

    2021

  9. [17]

    A unified digital twin framework for real- time monitoring and evaluation of smart manufacturing systems,

    Y . Qamsane, C.-Y . Chen, E. C. Balta, B.-C. Kao, S. Mohan, J. Moyne, D. Tilbury, and K. Barton, “A unified digital twin framework for real- time monitoring and evaluation of smart manufacturing systems,” in 2019 IEEE 15th international conference on automation science and eng...

    2019

  10. [18]

    Automatic simulation model generation for simulation-based, real-time shop floor control,

    Y . J. Son and R. A. Wysk, “Automatic simulation model generation for simulation-based, real-time shop floor control,” Computers in Industry , vol. 45, no. 3, pp. 291–308, 2001

    2001

  11. [19]

    Automatic generation of a simulation-based digital twin of an industrial process plant,

    G. S. Mart ´ınez, S. Sierla, T. Karhela, and V . Vyatkin, “Automatic generation of a simulation-based digital twin of an industrial process plant,” in IECON 2018-44th Annual Conference of the IEEE Industrial Electronics Society. IEEE, 2018, pp. 3084–3089

    2018

  12. [20]

    Modelling and solving approaches for scheduling problems in reconfigurable manufac- turing systems,

    X. Delorme, G. Fleury, P. Lacomme, and D. Lamy, “Modelling and solving approaches for scheduling problems in reconfigurable manufac- turing systems,” International Journal of Production Research , vol. 62, no. 7, pp. 2683–2704, 2024

    2024

  13. [21]

    Digital twin framework for reconfigurable manufacturing systems (rmss): design and simulation,

    J. Kombaya Touckia, N. Hamani, and L. Kermad, “Digital twin framework for reconfigurable manufacturing systems (rmss): design and simulation,” The International Journal of Advanced Manufacturing Technology, vol. 120, no. 7, pp. 5431–5450, 2022

    2022

  14. [22]

    Towards smart layout design for a reconfigurable manufacturing sys- tem,

    H. Arnarson, H. Yu, M. M. Olavsbr ˚aten, B. A. Bremdal, and B. Solvang, “Towards smart layout design for a reconfigurable manufacturing sys- tem,” Journal of Manufacturing Systems , vol. 68, pp. 354–367, 2023

    2023

  15. [23]

    Digital twin frame- work for reconfiguration management: concept & evaluation,

    B. Caesar, K. Barton, D. M. Tilbury, and A. Fay, “Digital twin frame- work for reconfiguration management: concept & evaluation,” IEEE Access, vol. 11, pp. 127 364–127 387, 2023

    2023

  16. [24]

    Digital twin-driven rapid reconfiguration of the automated manufacturing system via an open architecture model,

    J. Leng, Q. Liu, S. Ye, J. Jing, Y . Wang, C. Zhang, D. Zhang, and X. Chen, “Digital twin-driven rapid reconfiguration of the automated manufacturing system via an open architecture model,” Robotics and computer-integrated manufacturing, vol. 63, p. 101895, 2020

    2020

  17. [25]

    A reconfigurable modeling approach for digital twin-based manufacturing system,

    C. Zhang, W. Xu, J. Liu, Z. Liu, Z. Zhou, and D. T. Pham, “A reconfigurable modeling approach for digital twin-based manufacturing system,” Procedia Cirp, vol. 83, pp. 118–125, 2019

    2019

  18. [26]

    A framework for manufac- turing system reconfiguration and optimisation utilising digital twins and modular artificial intelligence,

    F. Mo, H. U. Rehman, F. M. Monetti, J. C. Chaplin, D. Sanderson, A. Popov, A. Maffei, and S. Ratchev, “A framework for manufac- turing system reconfiguration and optimisation utilising digital twins and modular artificial intelligence,” Robotics and Computer-Integrated Manufac...

    2023

  19. [27]

    Reconfigurable manufac- turing system: a systematic review, meta-analysis and future research directions,

    R. Pansare, G. Yadav, and M. R. Nagare, “Reconfigurable manufac- turing system: a systematic review, meta-analysis and future research directions,” Journal of Engineering, Design and Technology , vol. 21, no. 1, pp. 228–265, 2023

    2023

  20. [28]

    A requirements driven digital twin framework: Specification and opportunities,

    J. Moyne, Y . Qamsane, E. C. Balta, I. Kovalenko, J. Faris, K. Barton, and D. M. Tilbury, “A requirements driven digital twin framework: Specification and opportunities,” Ieee Access , vol. 8, pp. 107 781– 107 801, 2020

    2020

  21. [29]

    An ontology-based approach for product lifecycle management,

    A. Matsokis and D. Kiritsis, “An ontology-based approach for product lifecycle management,” Computers in industry , vol. 61, no. 8, pp. 787– 797, 2010

    2010

  22. [30]

    Industrial automation based on cyber-physical systems technologies: Prototype implementa- tions and challenges,

    P. Leit ˜ao, A. W. Colombo, and S. Karnouskos, “Industrial automation based on cyber-physical systems technologies: Prototype implementa- tions and challenges,” Computers in industry , vol. 81, pp. 11–25, 2016

    2016

  23. [31]

    The development of an ontology for describing the capabilities of manufacturing resources,

    E. J ¨arvenp¨a¨a, N. Siltala, O. Hylli, and M. Lanz, “The development of an ontology for describing the capabilities of manufacturing resources,” Journal of Intelligent Manufacturing, vol. 30, no. 2, pp. 959–978, 2019. 14

    2019

  24. [32]

    The model-based product agent: A control oriented architecture for intelligent products in multi- agent manufacturing systems,

    I. Kovalenko, D. Tilbury, and K. Barton, “The model-based product agent: A control oriented architecture for intelligent products in multi- agent manufacturing systems,” Control Engineering Practice , vol. 86, pp. 105–117, 2019

    2019

  25. [33]

    A methodology to develop and implement digital twin solutions for manufacturing systems,

    Y . Qamsane, J. Moyne, M. Toothman, I. Kovalenko, E. C. Balta, J. Faris, D. M. Tilbury, and K. Barton, “A methodology to develop and implement digital twin solutions for manufacturing systems,” IEEE Access, vol. 9, pp. 44 247–44 265, 2021

    2021

  26. [34]

    A digital twin framework for prognostics and health management,

    M. Toothman, B. Braun, S. J. Bury, J. Moyne, D. M. Tilbury, Y . Ye, and K. Barton, “A digital twin framework for prognostics and health management,” Computers in Industry , vol. 150, p. 103948, 2023

    2023

  27. [35]

    An ontology- based resource reconfiguration method for manufacturing cyber-physical systems,

    J. Wan, B. Yin, D. Li, A. Celesti, F. Tao, and Q. Hua, “An ontology- based resource reconfiguration method for manufacturing cyber-physical systems,” IEEE/ASME Transactions on Mechatronics, vol. 23, no. 6, pp. 2537–2546, 2018

    2018

  28. [36]

    The prot ´eg´e project: a look back and a look forward,

    M. A. Musen, “The prot ´eg´e project: a look back and a look forward,” AI matters, vol. 1, no. 4, pp. 4–12, 2015

    2015

  29. [37]

    A review of technology standards and patent portfolios for enabling cyber-physical systems in advanced manufacturing,

    A. J. Trappey, C. V . Trappey, U. H. Govindarajan, J. J. Sun, and A. C. Chuang, “A review of technology standards and patent portfolios for enabling cyber-physical systems in advanced manufacturing,” Ieee Access, vol. 4, pp. 7356–7382, 2016

    2016

  30. [38]

    International Organization for Standardization Geneva, Information technology — Top-level ontologies (TLO), International Organization for Standardization Geneva, Switzerland Std

    S. International Organization for Standardization Geneva, Information technology — Top-level ontologies (TLO), International Organization for Standardization Geneva, Switzerland Std. ISO/IEC 21 838, 2021

    2021

  31. [39]

    ISO/DIS 23 247, 2020

    ——, Automation systems and integration—digital twin framework for manufacturing, International Organization for Standardization Geneva, Switzerland Std. ISO/DIS 23 247, 2020. Bo Fu received the B.S. degree in vehicle engineer- ing from Tongji University, China, in 2017, the M....

    2020

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.