REVIEW 3 major objections 6 minor 3 references
Deep Learning Aided Multi-Objective Optimization and Multi-Criteria Decision Making in Thermal Cracking Process for Olefines Production
T0 review · 3 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash
Pith's one-line read Trained multilayer perceptron surrogates let MOPSO map Pareto fronts for an LPG thermal cracking furnace in under a minute, roughly three orders of magnitude faster than optimizing with the full mathematical model, and a five-method MCDM…
desk verdict A credible application of an existing surrogate-based MOO+MCDM pipeline to LPG cracking, with a real speedup claim but weak validation of the recommended points. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The load-bearing mechanism is the trained MLP surrogate: each of the six process objectives is predicted almost instantly from five decision variables (feed flow rate, coil outlet temperature, coil outlet pressure, steam-to-feed ratio, and coil inlet temperature), so the swarm can evaluate candidate operating conditions hundreds of times per minute instead of waiting on the slow mathematical model. On top of the surrogate, the Pareto-optimal archive produced by MOPSO is processed by CRITIC weighting, which assigns objective weights from the standard deviation and pairwise correlations of the normalized objective matrix, and then by five MCDM methods that rank the Pareto solutions; the paper's final selection rule is majority voting supplemented by similarity among the recommended solutions.
What would settle it
Take every recommended operating point from the six case studies, re-evaluate the six objectives with the full mathematical model, and compare them with the MLP predictions; if the errors are not small relative to the spread of the Pareto front, or if full-model MOPSO finds clearly better points, the surrogate-optimality claim collapses. The paper's own spot-check covers only four points in Case A, so this full check is the direct experiment that would settle the question.
Extended reading notes
Core claim
The central claim is that replacing the full mathematical model with MLP surrogates inside MOPSO preserves the essential optimum structure of the LPG cracking process while making each optimization tractable in under a minute on the same computer that would need about two days for the conventional approach. The paper trains one single-hidden-layer MLP for each of six outputs—annual ethylene and propylene production, heat duty, run length, ethylene selectivity, and severity index—from a complete factorial dataset over five operating variables, then runs MOPSO for 100 generations with 60 or 70 particles in six case studies. It reports test $R^2 \geq 0.995$ for all surrogate models, and in Case A four surrogate-optimal points are shown to give objective values close to those obtained by re-evaluating the mathematical model. The resulting Pareto fronts align with cracking chemistry, and the CRITIC plus five-MCDM procedure with majority voting yields a single recommended solution in each case.
Load-bearing premise
The scheme rests on the assumption that the mathematical model used to generate the training data faithfully represents the real LPG cracking furnace, so that a point that is best for the neural-network stand-in is also best for the actual process.
Editorial extensions
If this is right
- In all six case studies the surrogate-based optimization completes in under one minute, so the paper's claim implies that routine re-optimization of cracking conditions becomes cheap enough for industrial practice.
- The Pareto fronts quantify the trade-offs among ethylene, propylene, heat duty, run length, and selectivity, giving operators a map of how much of one objective is sacrificed for another.
- Because CRITIC weights come from the front itself and several MCDM methods often converge on the same solution, the workflow converts the Pareto set into one defensible setpoint rather than leaving the choice to the engineer.
- The consistently high test $R^2 \geq 0.995$ implies the MLP models are accurate enough to serve as objective functions for optimization screening, with full-model verification reserved for the final candidate.
Reading between the lines
- A consequence the paper leaves implicit is that the one-minute runtime makes near-real-time re-optimization possible: when feed quality or product prices shift, the operating point can be updated within the hour rather than after a multi-day study.
- Because the paper validates surrogate predictions against the full model at only four points in Case A, a direct test would be to feed every MCDM-recommended solution from all six cases back into the full model and compare objective values.
- In this reader's view, the agreement among several MCDM methods is evidence of stability but not a formal guarantee; changing the CRITIC weighting or using other weight schemes could still move the majority-vote winner.
- The success of a single-hidden-layer MLP suggests the true input–output map of this furnace is smooth, so simpler response-surface surrogates might achieve much of the same speedup; the paper does not compare against them.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The manuscript proposes a surrogate-based multi-objective optimization workflow for the LPG thermal cracking process. A complete factorial design over five operating variables is run on an (unidentified) mathematical model to generate training data; single-hidden-layer MLPs are trained for six objectives and report test-set R² values of at least 0.995; MLP-aided MOPSO is then applied in six case studies to produce Pareto fronts; CRITIC weights and five MCDM methods (MABAC, PROBID, SAW, sPROBID, TOPSIS) are used to recommend a single operating point. The paper's central practical claims are that the MLP-aided optimization completes in under one minute versus about two days for direct simulation-based MOPSO, and that the recommended points are valid Pareto-optimal solutions for the thermal cracking process.
Significance. If the surrogate fidelity at the optimizer-selected points and the validity of the underlying process model are established, the workflow is a useful template for accelerating MOO of cracking furnaces: it combines factorial data generation, MLP surrogates, MOPSO, CRITIC weighting, and a transparent MCDM selection stage, and the six case studies illustrate plausible trade-offs among ethylene, propylene, heat duty, run length, and selectivity indices. The four-point spot check in Table 2 is a useful start, and the majority-voting procedure across five MCDM methods is a reasonable safeguard against method-specific ranking artifacts. However, the current evidence does not yet establish that the recommended operating conditions are Pareto-optimal for the actual mathematical model, so the significance is conditional on additional validation.
major comments (3)
- [Section 3, first paragraph] The 'mathematical model' used to generate all training data is never identified or cited in the main text, and its accuracy against plant or experimental data is not addressed. Since every surrogate prediction inherits any error in this model, the absence of model identification and validation is load-bearing rather than stylistic. The reference list contains Nabavi et al. (2009), which appears uncited in the text and may be the intended source; the authors should cite it explicitly and provide the model equations or a clear pointer to them.
- [Section 3.1, Table 2] The only full-model validation is a set of four points (I-IV) from Case A, and these are not the MCDM-recommended solution reported in Table 3; Cases B-F have no full-model checks at all. Because the optimizer drives decision variables to the boundaries of their ranges in the recommended solutions (e.g., Fin=14, Tin=703, COT=853, COP=1.5, SR=0.4 in Tables 3-8), a global test-set R² over a factorial design does not establish surrogate accuracy in the regions where the optimizer actually selects solutions. The authors should evaluate the full mathematical model at the recommended solutions of Tables 3-8 and report the resulting errors, and ideally evaluate a sample of nondominated solutions from each case to show that the surrogate Pareto front is not an artifact of MLP approximation.
- [Section 3.1, first paragraph; Section 4] The headline speedup ('within one minute' versus 'an average of two days') is not substantiated. The manuscript provides no hardware details, no measured wall-clock times, no number of function evaluations for either approach, no repetitions, and no description of the conventional MOPSO implementation used as the baseline. Since this acceleration claim appears in the abstract and conclusions, it should either be supported by a reproducible benchmark or removed and replaced by a more modest qualitative statement.
minor comments (6)
- [Sections 2.5 and 2.7; Table 7] There are typographical inconsistencies in the MCDM method names: 'PRBOID' and 'sPRBOID' in Sections 2.5 and 2.7 should be 'PROBID' and 'sPROBID', and Table 7 contains 'sPROBOD' and 'TOPISIS'.
- [Section 2.3, Eq. (7)] The summation in the formula for c_j is not clearly typeset; please state explicitly that the sum runs over k=1 to n and clarify how the j=k term is treated, since this directly affects the CRITIC weights.
- [Section 3.1] The factorial design is described only qualitatively; the number of levels per decision variable and the total dataset size should be reported so that the reported R² values can be interpreted and the coverage of the boundary regions can be assessed.
- [Sections 3.4 and 3.5] There are several language slips that should be corrected, including 'taken removed from the production line', 'heat fraction process', and 'speed of heat fractions'; in context these appear to refer to cracking rather than 'fraction' processes.
- [Section 3.5] The definition of the severity index C-3/C3= should be double-checked: as written, it includes propylene in both the numerator and the denominator, which makes the index's behavior and interpretation unclear.
- [Reproducibility] For a data-driven workflow, providing the trained MLP weights or the code used for surrogate training, MOPSO, and MCDM would substantially improve reproducibility; currently only architecture-level information is indicated.
Circularity Check
No significant circularity: surrogate-based optimization is a standard workflow, and the reported acceleration and Pareto fronts are not derived from their own inputs by construction.
full rationale
The paper's central claim is that fitting MLP surrogates to data generated by the underlying LPG cracking model lets MOPSO run within one minute instead of two days. This is a conventional surrogate-based optimization workflow rather than a circular derivation. The MLPs are trained on data generated by the mathematical model and evaluated on held-out test data, with reported test R2 values of at least 0.995; the four spot checks in Section 3.1 compare MLP predictions with the mathematical model at previously unseen points, which is a standard validation protocol and not a statistically forced prediction. The Pareto fronts are Pareto fronts of the fitted surrogates, which is inherent to surrogate-based optimization and is not disguised as an independent first-principles result. The MCDM methods are external algorithms applied after optimization, and the cited MCDM and model references, including self-citations, are references to established methods and a prior first-principles model rather than load-bearing uniqueness assertions. No equation in the paper reduces to its own inputs, and no fitted parameter is renamed as a prediction. The main limitations are about surrogate fidelity at optimizer-selected extremes and the absence of experimental validation, which are correctness or robustness concerns, not circularity.
Assumptions & free parameters
free parameters (3)
- MLP hidden-layer neuron counts, learning rates, and momentum constants =
reported in Supporting Information Table S1, not in main text
- MOPSO algorithm parameters (population size, generations, inertia, learning factors, mutation, archive size) =
partially reported: 60 or 70 particles, 100 generations; other values not given
- MLP weights and biases for the six surrogate models =
not reported
assumptions (4)
- domain assumption The mathematical model of the LPG thermal cracking process used to generate training data is an accurate representation of the real process and covers the operating ranges in Table 1.
- domain assumption A single-hidden-layer MLP can accurately represent the input-output mapping of the thermal cracking model throughout the decision-variable ranges, not only on the training/test points.
- domain assumption MOPSO with 100 generations and 60 or 70 particles converges to sufficiently complete Pareto-optimal fronts for the six cases.
- domain assumption The five MCDM methods and CRITIC weights provide a meaningful way to select one solution for implementation, i.e., the weighting reflects the plant's actual priorities.
Cite this review
Pith. "Pith review of Deep Learning Aided Multi-Objective Optimization and Multi-Criteria Decision Making in Thermal Cracking Process for Olefines Production." pith.science (2026). https://pith.science/paper/P2BQOTXQ
@misc{pith2026241211035,
author = {Pith},
title = {Pith review of: Deep Learning Aided Multi-Objective Optimization and Multi-Criteria Decision Making in Thermal Cracking Process for Olefines Production},
year = {2026},
howpublished = {\url{https://pith.science/paper/P2BQOTXQ}},
note = {Machine review of arXiv:2412.11035}
}
read the original abstract
Background: Multilayer perceptron (MLP) aided multi-objective particle swarm optimization algorithm (MOPSO) is employed in the present article to optimize the liquefied petroleum gas (LPG) thermal cracking process. This new approach significantly accelerated the multi-objective optimization (MOO), which can now be completed within one minute compared to the average of two days required by the conventional approach. Methods: MOO generates a set of equally good Pareto-optimal solutions, which are then ranked using a combination of a weighting method and five multi-criteria decision making (MCDM) methods. The final selection of a single solution for implementation is based on majority voting and the similarity of the recommended solutions from the MCDM methods. Significant Findings: The deep learning (DL) aided MOO and MCDM approach provides valuable insights into the trade-offs between conflicting objectives and a more comprehensive understanding of the relationships between them. Furthermore, this approach also allows for a deeper understanding of the impact of decision variables on the objectives, enabling practitioners to make more informed, data-driven decisions in the thermal cracking process.
Reference graph
Works this paper leans on
-
[13]
https://doi.org/10.1016/j.jtice.2017.10.025 Fishburn, P. C. (1967). A Problem -based selection of multi -attribute decision making methods. Blackwell Publishing. Geng, Z., Wang, Z., Zhu, Q., & Han, Y. (2016). Multi -objective operation optimization of ethylene cracking furnace based on AMOPSO algorithm. Chemical Engineering Science, 153, 21–33. https://do...
-
[376]
https://doi.org/10.1016/j.jtice.2012.11.020 Aghbashlo, M., Hosseinpour, S., & Mujumdar, A. S. (2015). Application of artificial neural networks (ANNs) in drying technology: A comprehensive review. Drying Technology, 33(12), 1397–1462. https://doi.org/10.1080/07373937.2015.1036288 Amghizar, I., Vandewalle, L. A., Van Geem, K. M., & Marin, G. B. (2017). New...
-
[4607]
https://doi.org/10.1021/acs.energyfuels.5b00917 Pamučar, D., & Ćirović, G. (2015). The selection of transport and handling resources in logistics centers using Multi -Attributive Border Approximation area Comparison (MABAC). Expert Systems with Applications , 42(6), 3016 –3028. https://doi.org/10.1016/j.eswa.2014.11.057 36 Park, M., Wang, Z., Li, L., & Wa...
Reviewed August 11, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.