Pith. sign in

REVIEW 2 major objections 4 minor 33 references

Explainable AI for Correct Root Cause Analysis of Product Quality in Injection Moulding

T0 review · 2 major / 4 minor · reviewed 2026-08-16 · deepseek-v4-flash

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

Pith's one-line read Permutation-based SHAP correctly identifies the machine setting responsible for a product's weight deviation in injection moulding, whereas ICE misattributes the cause because it ignores interactions between settings.

desk verdict A useful applied comparison of SHAP and ICE for injection moulding root cause analysis, but the central 'correct cause' claim rests on an asserted ground truth and needs quantification. read the letter →

arxiv 2505.01445 v1 pith:AFKQUOM4 submitted 2025-04-29 cs.LG cs.AI

Muhammad Muaz , Sameed Sajid , Tobias Schulze , Chang Liu , Nils Klasen , Benny Drescher This is my paper · ORCID
classification cs.LGcs.AI MSC 62P3062M4591A80 PACS 89.20.Bb07.05.Mh02.50.Le
keywords injectionmouldingrootcauseanalysisexplainableAISHAPindividualconditionalexpectationfeatureinteractionsqualitycontrolattribution
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

The paper claims that when an injection-moulded part deviates from its target weight, the correct way to find which machine setting caused the deviation is permutation-based SHAP, not individual conditional expectation (ICE) plots. It first demonstrates, using a designed experiment with six operator-controlled settings, that those settings genuinely interact in determining product weight. In controlled trials where a single setting was moved from its middle value, SHAP named that changed setting as the top cause in nearly all cases, while ICE named packing time first in every trial, even when packing time had not changed. The conclusion is that choosing an explainability method that accounts for interactions is not a formality but determines whether the diagnosis is actionable.

What carries the argument

The central machinery is the contrast between two local, model-agnostic explanation methods applied to the same fitted black-box models. Permutation-based SHAP assigns each machine setting a Shapley value, the average marginal contribution of that setting to the prediction over all random orderings of the settings, which is the mechanism that lets interaction effects enter each attribution. ICE, by contrast, varies one setting while holding the others fixed and records the standard deviation of the resulting prediction curve, so each setting is judged in isolation. The paper's third piece of machinery is the controlled single-change protocol: because the ideal product weight is defined at the mid value of every setting, changing exactly one setting defines which setting must be the first major cause, giving the comparison a ground truth. The H-statistic then supplies the evidence that interactions are large enough to make the choice between the two explanation methods consequential.

What would settle it

Run the controlled protocol on a synthetic or real dataset where one setting is changed but a known interaction makes a different setting the dominant predictor of weight; if ICE then identifies the true dominant cause more often than SHAP, or SHAP's ranking matches the changed setting no better than chance, the central claim fails. A cheaper check is to collect many more trials around the cooling-time case, where SHAP sometimes fails on random forest, and test whether those failures are systematic rather than noise.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a better feature attribution translates to the correct cause identification: an accurate random forest or multilayer perceptron trained on six machine settings, when explained by permutation-based SHAP, attributes a product weight deviation to the machine setting that was actually changed from its mid value, while the same model explained by ICE attributes the deviation to packing time in every tested cycle. The paper uses the H-statistic and two-way partial dependence plots to show that interactions among the settings are real and substantial, with melt temperature and packing time interacting strongly. It interprets SHAP's correct attributions as a consequence of SHAP distributing credit over all feature orderings, so interaction effects are folded into each setting's score, whereas ICE varies one setting at a time and therefore misses the interaction component. The pattern repeats when the quality characteristic is changed from weight to planarity, supporting the claim that the finding is not specific to one product property.

Load-bearing premise

The comparison's ground truth is the assumption that when exactly one machine setting is changed from its mid value, that setting is necessarily the first major cause of the resulting weight deviation, and that an accurate black-box model fitted to the data will reflect that ordering.

Editorial extensions

If this is right

  • An operator facing a drifting product weight can trust SHAP's top-ranked setting as the first thing to adjust, rather than defaulting to packing time as ICE suggests.
  • Because SHAP on random forest costs only milliseconds per cycle, the explanation step can run in real time alongside the roughly ten-second production cycle.
  • The same attribution logic should apply to other quality characteristics, since the planarity experiment reproduces the pattern for a different property of the same product.
  • In processes monitored on multiple quality characteristics at once, correctly identifying the root cause before adjusting settings becomes more important, since changing the wrong setting to fix one characteristic can damage another.

Reading between the lines

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

  • If interactions are the reason SHAP outperforms ICE, the two methods should give nearly identical rankings in processes where the H-statistic is near zero; measuring that convergence would isolate the mechanism.
  • The same mid-value single-change protocol could benchmark additional explanation methods such as LIME or integrated gradients, without needing external ground truth, as long as the single-change assumption holds.
  • The paper's own cooling-time exception for random forest suggests SHAP is not infallible; a practical safety check would flag settings whose SHAP ranking flips across retrained models or random splits.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. The paper addresses root cause analysis of product quality in injection moulding by comparing two model-agnostic explainability methods, permutation-based SHAP and ICE, on random forest and multilayer perceptron models. Using a face-centered central composite design dataset of 77 machine-setting combinations repeated 20 times (1540 cycles), the authors train both models to high accuracy (MAPE < 0.05%) and use Friedman's H-statistic to argue that machine settings interact. They then rank the impact of six machine settings on product weight for each test sample and compare SHAP and ICE rankings. In controlled experiments where a single machine setting is changed from its mid value, the paper claims that SHAP ranks the changed setting first, whereas ICE always ranks packing time first. The same comparison is repeated for a second quality characteristic, product planarity.

Significance. If the central claim were established, the paper would offer a practically useful, model-agnostic recipe for root cause identification in injection moulding, a domain where small weight deviations matter and operators must decide which machine setting to adjust. The study's strengths include a real experimental dataset with repeated runs, careful model selection with confidence intervals for errors (Tables 5 and 6), use of an interaction diagnostic (H-statistic), comparison across two distinct model families, and an additional validation on a second quality characteristic (planarity). These elements make the paper useful as an empirical comparison of XAI methods in manufacturing. However, the central evaluation that SHAP 'correctly' identifies the cause is not supported by a quantitative measure: the correctness judgment rests on visual inspection of Figures 6, 7, and 9, and the paper's own exception ('apart from the cooling time in RF') is left unquantified.

major comments (2)
  1. [Section 4.2, Equations (7)-(8)] The ground-truth assumption that a single changed machine setting 'should be the first major cause' is not a logical consequence of model accuracy. Shapley values distribute prediction differences and split interaction terms among features, so a changed feature can receive a small or even zero attribution when it is the only feature that moves; for example, with y = x1*x2 and x1 changed from 0 to 1 while x2 remains 0, the Shapley value of x1 is 0. Thus the 'correct cause' verdict is a normative claim, not an oracle. The paper itself reports an exception (cooling time under RF) but does not quantify it. To make the central claim defensible, the authors should report the proportion of test samples in which SHAP's top-ranked feature equals the experimentally changed setting (and the corresponding proportion for ICE), with confidence intervals across the 10 random splits, and explicitly discuss the observed exceptions.
  2. [Section 3.3] The statement that the machine settings 'are independently tuned with zero correlations among themselves, hence, the feature independence holds' is not valid: zero Pearson correlation does not imply statistical independence. For the described face-centered CCD with 77 unique combinations, the joint distribution is not a product of marginals; for instance, for any two-level factor, P(+1,+1)=16/77 while P(+1)^2=(33/77)^2, so the independence assumption is violated. Since the permutation-based SHAP approximation in Equation (7) relies on sampling from a background distribution assuming independence, the authors should either use a dependence-aware SHAP variant or empirically assess the sensitivity of the SHAP rankings to this assumption. Without this, the claim that SHAP correctly captures interactions is not fully supported.
minor comments (4)
  1. [Section 3.4] The sentence 'The rankings by ICE ... are used as the benchmark for SHAP based rankings' is misleading, since the paper later argues that ICE produces incorrect rankings. ICE should be described as a competing method rather than a benchmark.
  2. [Section 4.1, Figure 4] The H-statistic is computed only for the random forest model, but the SHAP-versus-ICE comparison is performed for both RF and MLP; computing the H-statistic for MLP as well would strengthen the claim that both models capture interactions and that the conclusions are model-agnostic.
  3. [Section 2, Table 1] In Table 1, 'min mould height: 200 nm' appears to be a typo for '200 mm'; please correct it.
  4. [Section 1 and Conclusion] The phrase 'the better feature attribution translates to the correct cause identification' is stated as a general conclusion, but the evidence is limited to one dataset and one quality characteristic (weight, and secondarily planarity); the conclusion should be scoped accordingly to avoid overclaiming.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: SHAP and ICE are compared against an external changed-feature benchmark; the benchmark is normative but not derived from the explanation equations.

full rationale

The paper's central claim is that permutation-based SHAP ranks the experimentally changed machine setting as the top cause while ICE always ranks packing time first. The correctness ground truth is defined in Section 4.2 by the controlled CCD design: "When a single machine setting changes from its mid value causing a deviation of the product weight from its expected/desired value, it is obvious that that particular machine setting should be the first major cause of the deviation and should be attributed for the aberration." This is an external, intervention-based benchmark rather than a value fitted from SHAP or ICE. The SHAP values are computed from the game-theoretic permutation formula (Eqs. 7-8), and the ICE impact is computed as the standard deviation of the ICE curve (Eq. 9); neither formula contains the changed-feature label, and no parameter is fitted to make either method agree with that label. The conclusion that SHAP "has followed the correct input/output relationship" is therefore an evaluation against an external definition, not a derivation that reduces to its own input. The only self-referential aspect is that "correct" is defined as "the machine setting that was modified," which is a contestable normative oracle: exact Shapley values need not rank the changed feature first when strong interactions are present, as the paper's own cooling-time exception in Section 4.1 shows. That weakness concerns the validity of the benchmark, not circularity of the derivation. Similarly, the Section 3.3 claim that zero Pearson correlation implies feature independence is mathematically unsupported, but it is an assumption error rather than a circular step. Self-citations such as [10] and [26] are not load-bearing for the cause-analysis conclusion. No fitted parameter is renamed as a prediction, and no uniqueness theorem or ansatz is imported from the authors' prior work. Accordingly, the paper is self-contained against its external test protocol, and the circularity score is 0.

Assumptions & free parameters 2 free parameters · 5 assumptions · 0 invented entities

The central comparison rests on model training choices and on the causal ground truth from the CCD design; no new physical entities or fitted physical constants are introduced. The explanation comparison itself is not a fitted derivation.

free parameters (2)
  • MLP hyperparameter configuration = 2 hidden layers (8,4 neurons), learning rate 0.01, L2 0.0001, batch 8, L-BFGS, random state 42
    Chosen via grid search and empirical stability checks; the explanations are computed on this trained model, so the ranking results depend on this choice.
  • SHAP permutation count P = not reported
    Equation (8) averages over P permutations; the paper does not state P, so the approximation error in SHAP values is not auditable.
assumptions (5)
  • standard math SHAP values fairly distribute model predictions among features and satisfy additivity and efficiency (Lundberg and Lee).
    Used without proof in Section 3.3 to interpret absolute SHAP values as feature impact.
  • standard math Friedman H-statistic computed from partial dependence correctly separates interaction from main effects.
    Used in Section 3.2 and Figure 4 to claim that machine settings interact.
  • domain assumption The six operator-controlled machine settings suffice to determine product weight and planarity; other sensor data can be excluded.
    Section 2 states that other process parameters and sensor time series are not included 'for simplicity', so unobserved factors are assumed not to confound the attribution.
  • domain assumption When a single machine setting is changed from its mid value, that setting is the first major cause of the product quality deviation.
    This is the ground truth used in Section 4.2 to decide which explanation method is correct.
  • domain assumption Zero Pearson correlation among the six machine settings implies the feature independence required by SHAP.
    Section 3.3 states 'zero correlations ... hence, the feature independence holds'. Zero correlation does not generally imply independence, and the CCD sample may not be fully independent.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Explainable AI for Correct Root Cause Analysis of Product Quality in Injection Moulding." pith.science (2026). https://pith.science/paper/AFKQUOM4

@misc{pith2026250501445,
  author       = {Pith},
  title        = {Pith review of: Explainable AI for Correct Root Cause Analysis of Product Quality in Injection Moulding},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AFKQUOM4}},
  note         = {Machine review of arXiv:2505.01445}
}
read the original abstract

If a product deviates from its desired properties in the injection moulding process, its root cause analysis can be aided by models that relate the input machine settings with the output quality characteristics. The machine learning models tested in the quality prediction are mostly black boxes; therefore, no direct explanation of their prognosis is given, which restricts their applicability in the quality control. The previously attempted explainability methods are either restricted to tree-based algorithms only or do not emphasize on the fact that some explainability methods can lead to wrong root cause identification of a product's deviation from its desired properties. This study first shows that the interactions among the multiple input machine settings do exist in real experimental data collected as per a central composite design. Then, the model-agnostic explainable AI methods are compared for the first time to show that different explainability methods indeed lead to different feature impact analysis in injection moulding. Moreover, it is shown that the better feature attribution translates to the correct cause identification and actionable insights for the injection moulding process. Being model agnostic, explanations on both random forest and multilayer perceptron are performed for the cause analysis, as both models have the mean absolute percentage error of less than 0.05% on the experimental dataset.

Figures

Figures reproduced from arXiv: 2505.01445 by the authors.

Figure 1
Figure 1. Explainable AI broad categories used in manufacturing [16]. [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Schematic of production setup for training and deployment. [PITH_FULL_IMAGE:figures/full_fig_p009_2.png] view at source ↗
Figure 3
Figure 3. PDP plots of machine settings for (a) RF, and (b) MLP. [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Bar plot of one-versus-all interactions of machine settings measured by H [PITH_FULL_IMAGE:figures/full_fig_p020_4.png]
Figure 5
Figure 5. Figure 5: Step plots of packing time (left) and packing pressure (right) vs RF-based weight [PITH_FULL_IMAGE:figures/full_fig_p021_5.png]
Figure 6
Figure 6. Figure 6: Comparison of the normalized feature impact values of SHAP and ICE applied [PITH_FULL_IMAGE:figures/full_fig_p024_6.png]
Figure 7
Figure 7. Figure 7: Comparison of the normalized feature impact values of SHAP and ICE applied [PITH_FULL_IMAGE:figures/full_fig_p025_7.png]
Figure 8
Figure 8. Figure 8: Product sample showing the shaded region of measuring the planarity charac [PITH_FULL_IMAGE:figures/full_fig_p027_8.png]
Figure 9
Figure 9. Figure 9: Comparison of the normalized feature impact values of SHAP and ICE applied on [PITH_FULL_IMAGE:figures/full_fig_p028_9.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 26 canonical work pages

  1. [1]

    Ribeiro, Support Vector Machines for Quality Monitoring in a Plastic Injection Molding Process, IEEE Trans

    B. Ribeiro, Support Vector Machines for Quality Monitoring in a Plastic Injection Molding Process, IEEE Trans. Syst., Man, Cybern. C 35 (3) (2005) 401–410. doi:10.1109/TSMCC.2004.843228. URL http://ieeexplore.ieee.org/document/1487588/

    work page arXiv 2005

  2. [2]

    Z. Song, S. Liu, X. Wang, Z. Hu, Optimization and prediction of volume shrinkage and warpage of injection-molded thin-walled parts based on neural network, Int J Adv Manuf Technol 109 (3-4) (2020) 755–769. doi:10.1007/s00170-020-05558-6 . URL https://link.springer.com/10.1007/s00170-020-05558-6

    work page doi:10.1007/s00170-020-05558-6 2020

  3. [3]

    Lockner, C

    Y. Lockner, C. Hopmann, Induced network-based transfer learning in injection molding for process modelling and optimization with artificial neural networks, Int J Adv Manuf Technol 112 (11-12) (2021) 3501– 29

    work page 2021

  4. [4]

    Lockner, C

    Y. Lockner, C. Hopmann, W. Zhao, Transfer learning with ar- tificial neural networks between injection molding processes and different polymer materials, J Manuf Process 73 (2022) 395–408. doi:10.1016/j.jmapro.2021.11.014. URL https://linkinghub.elsevier.com/retrieve/pii/ S1526612521008136

    work page doi:10.1016/j.jmapro.2021.11.014 2022

  5. [5]

    J. C. Chen, G. Guo, W.-N. Wang, Artificial neural network-based online defect detection system with in-mold temperature and pressure sensors for high precision injection molding, Int J Adv Manuf Technol 110 (7-8) (2020) 2023–2033. doi:10.1007/s00170-020-06011-4 . URL https://link.springer.com/10.1007/s00170-020-06011-4

    work page doi:10.1007/s00170-020-06011-4 2020

  6. [6]

    Ke, M.-S

    K.-C. Ke, M.-S. Huang, Quality Prediction for Injection Molding by Using a Multilayer Perceptron Neural Network, Polymers 12 (8) (2020)

    work page 2020

  7. [7]

    H. Jung, J. Jeon, D. Choi, J.-Y. Park, Application of Machine Learning Techniques in Injection Molding Quality Prediction: Implications on Sustainable Manufacturing Industry, Sustainability 13 (8) (2021) 4120. doi:10.3390/su13084120. URL https://www.mdpi.com/2071-1050/13/8/4120

    work page doi:10.3390/su13084120 2021

  8. [8]

    Finkeldey, J

    F. Finkeldey, J. Volke, J.-C. Zarges, H.-P. Heim, P. Wiederkehr, 30 Learning quality characteristics for plastic injection molding processes using a combination of simulated and measured data, J Manuf Proc 60 (2020) 134–143. doi:10.1016/j.jmapro.2020.10.028. URL https://linkinghub.elsevier.com/retrieve/pii/ S1526612520306964

    work page doi:10.1016/j.jmapro.2020.10.028 2020

Show all 33 references
  1. [9]

    R. D. P´ arizs, D. T¨ or¨ ok, T. Ageyeva, J. G. Kov´ acs, Machine Learning in Injection Molding: An Industry 4.0 Method of Quality Prediction, Sensors 22 (7) (2022) 2704. doi:10.3390/s22072704. URL https://www.mdpi.com/1424-8220/22/7/2704

    2022 doi

  2. [10]

    M. Muaz, H. Yu, W. L. Sung, C. Liu, B. Drescher, A mul- titask encoder–decoder model for quality prediction in in- jection moulding, J Manuf Process 103 (2023) 238–247. doi:10.1016/j.jmapro.2023.08.030. URL https://linkinghub.elsevier.com/retrieve/pii/ S1526612523008046

    2023 doi

  3. [11]

    Quatrini, F

    E. Quatrini, F. Costantino, G. Di Gravio, R. Patriarca, Machine learning for anomaly detection and process phase classification to improve safety and maintenance activities, J Manuf Syst 56 (2020) 117–132. doi:10.1016/j.jmsy.2020.05.013. URL https://linkinghub.elsevier.com/ret...

    2020 doi

  4. [12]

    Vilone, L

    G. Vilone, L. Longo, Classification of explainable artificial intelligence methods through their output formats, Mach Learn Knowl Extr 3 (3) 31 (2021) 615–661. doi:10.3390/make3030032. URL https://www.mdpi.com/2504-4990/3/3/32

    2021 doi

  5. [13]

    F. K. Dosilovic, M. Brcic, N. Hlupic, Explainable artificial intelli- gence: A survey, in: 2018 41st International Convention on Informa- tion and Communication Technology, Electronics and Microelectronics (MIPRO), IEEE, Opatija, 2018, pp. 0210–0215. doi:10.23919/MIPRO. 2018.8...

    2018

  6. [14]

    Adadi, M

    A. Adadi, M. Berrada, Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI), IEEE Access 6 (2018) 52138– 52160. doi:10.1109/ACCESS.2018.2870052. URL https://ieeexplore.ieee.org/document/8466590/

    2018

  7. [15]

    Soldatos, D

    A Review of Explainable Artificial Intelligence in Manufacturing, in: J. Soldatos, D. Kyriazis (Eds.), Becoming a Platform in Europe: On the Governance of the Collaborative Economy, Now Publishers, 2021. doi:10.1561/9781680838770.ch5. URL https://nowpublishers.com/article/Book...

    2021 doi

  8. [16]

    H. K. Dam, T. Tran, A. Ghose, Explainable software analytics, in: Pro- ceedings of the 40th International Conference on Software Engineering: New Ideas and Emerging Results, ACM, Gothenburg Sweden, 2018, pp. 53–56. doi:10.1145/3183399.3183424. URL https://dl.acm.org/doi/10.114...

    2018

  9. [17]

    Obregon, J

    J. Obregon, J. Hong, J.-Y. Jung, Rule-based explanations based on ensemble machine learning for detecting sink mark defects in the injection moulding process, J Manuf Syst 60 (2021) 392–405. doi:10.1016/j.jmsy.2021.07.001. URL https://linkinghub.elsevier.com/retrieve/pii/ S027...

    2021 doi

  10. [18]

    J. P. Soler, N. G. Sastre, A. L. Santacruz, V. S. Nu˜ nez, M. C. Y¨ uksel, S. G. Settier, J.-C. Perez-Cortes, Advancing towards zero-defect manu- facturing in the plastic injection industry with global and local explain- ability approaches, in: IECON 2023- 49th Annual Conferen...

    2023

  11. [19]

    Gim, L.-S

    J. Gim, L.-S. Turng, Interpretation of the effect of transient process data on part quality of injection molding based on explainable arti- ficial intelligence, Int J Prod Res 61 (23) (2023) 8192–8212. doi: 10.1080/00207543.2023.2216310. URL https://www.tandfonline.com/doi/ful...

    2023

  12. [20]

    Kozjek, R

    D. Kozjek, R. Vrabiˇ c, D. Kralj, P. Butala, Interpretative identification of the faulty conditions in a cyclic manufacturing process, J Manuf Syst 43 (2017) 214–224. doi:10.1016/j.jmsy.2017.03.001. URL https://linkinghub.elsevier.com/retrieve/pii/ S0278612517300304 33

    2017 doi

  13. [21]

    Kozjek, R

    D. Kozjek, R. Vrabiˇ c, D. Kralj, P. Butala, N. Lavraˇ c, Data mining for fault diagnostics: A case for plastic injection molding, Procedia CIRP 81 (2019) 809–814. doi:10.1016/j.procir.2019.03.204. URL https://linkinghub.elsevier.com/retrieve/pii/ S2212827119305098

    2019 doi

  14. [22]

    Gim, In-mold condition-centered and explainable artificial intelligence-based (IMC-XAI) process optimization for injection molding, J Manuf Syst (2024)

    J. Gim, In-mold condition-centered and explainable artificial intelligence-based (IMC-XAI) process optimization for injection molding, J Manuf Syst (2024)

    2024

  15. [23]

    J. H. Friedman, B. E. Popescu, Predictive learning via rule ensembles, Ann. Appl. Stat. 2 (3) (Sep. 2008). doi:10.1214/07-AOAS148. URL https://projecteuclid.org/journals/ annals-of-applied-statistics/volume-2/issue-3/ Predictive-learning-via-rule-ensembles/10.1214/07-AOAS148. full

    2008 doi

  16. [24]

    Tsai, H.-J

    K.-M. Tsai, H.-J. Luo, An inverse model for injection molding of optical lens using artificial neural network coupled with genetic algorithm, J Intell Manuf 28 (2) (2017) 473–487. doi:10.1007/s10845-014-0999-z . URL http://link.springer.com/10.1007/s10845-014-0999-z

    2017 doi

  17. [25]

    URL http://www.autodesk.com/moldflow

    Moldflow, molding window analysis. URL http://www.autodesk.com/moldflow

  18. [26]

    J. Lee, J. Kim, J. Kim, A study on the architecture of artificial neural network considering injection-molding process steps, Polymers 15 (23) 34 (2023) 4578. doi:10.3390/polym15234578. URL https://www.mdpi.com/2073-4360/15/23/4578

    2023 doi

  19. [27]

    Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable, 2nd Edition, 2022

    C. Molnar, Interpretable Machine Learning: A Guide for Making Black Box Models Explainable, 2nd Edition, 2022. URL https://christophm.github.io/interpretable-ml-book

    2022

  20. [28]

    S. M. Lundberg, S.-I. Lee, A unified approach to interpreting model predictions, 31st Conference on Neural Information Processing Systems (NIPS) 30 (2017) 4768 – 4771

    2017

  21. [29]

    Bengio, P

    Y. Bengio, P. Simard, P. Frasconi, Learning long-term dependencies with gradient descent is difficult, IEEE Trans. Neural Netw. 5 (2) (1994) 157–

    1994

  22. [30]

    Knoll, H.-P

    J. Knoll, H.-P. Heim, Analysis of the Machine-Specific Behavior of In- jection Molding Machines, Polymers 16 (1) (2023) 54. doi:10.3390/ polym16010054. URL https://www.mdpi.com/2073-4360/16/1/54 35

    2023

  23. [166]

    URL https://ieeexplore.ieee.org/document/279181/

    doi:10.1109/72.279181. URL https://ieeexplore.ieee.org/document/279181/

    doi

  24. [1812]

    URL https://www.mdpi.com/2073-4360/12/8/1812

    doi:10.3390/polym12081812. URL https://www.mdpi.com/2073-4360/12/8/1812

    doi

  25. [3513]

    URL http://link.springer.com/10.1007/s00170-020-06511-3

    doi:10.1007/s00170-020-06511-3 . URL http://link.springer.com/10.1007/s00170-020-06511-3

    doi

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.