Pith. sign in

REVIEW 6 major objections 6 minor 77 references

Reinforcement Learning for Multi-Objective Multi-Echelon Supply Chain Optimisation

T0 review · 6 major / 6 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A decomposition-based multi-objective reinforcement learning method with a shared experience buffer is the most balanced approach the authors test for approximating the Pareto front in non-stationary, multi-echelon supply chains balancing…

desk verdict A genuinely useful generalized MOMDP formulation for multi-echelon supply chains, but the algorithm ranking is built on a self-referential benchmark; the 'MORL/D is most balanced' claim is not yet proven. read the letter →

arxiv 2507.19788 v1 pith:DIRPX2YO submitted 2025-07-26 cs.AI

Rifny Rachman , Josh Tingey , Richard Allmendinger , Pradyumn Shukla , Wei Pan This is my paper · ORCID
classification cs.AI MSC 90C2990C4090B06
keywords multi-objectivereinforcementlearningsupplychainoptimisationmulti-echelonnetworkMarkovdecisionprocessParetofrontapproximationsharedexperiencebuffersustainability
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 decomposition-based multi-objective reinforcement learning algorithm, MORL/D, is the most balanced method for approximating the Pareto front in non-stationary, multi-echelon supply chain problems that trade off profit, greenhouse-gas emissions, and service-level inequality. It builds a generalised multi-objective Markov decision process formulation of the supply chain, provides a configurable simulator, and benchmarks MORL/D against weighted-sum PPO, a single-objective RL algorithm adapted to multiple objectives, and NSGA-II, a multi-objective evolutionary algorithm. In complex networks the paper reports up to 75% higher hypervolume than NSGA-II and roughly eleven times denser solution sets than weighted-sum PPO, while also producing more stable production and inventory levels and lower demand loss. A sympathetic reader would care because, if correct, the result gives practitioners a concrete MORL configuration that jointly covers optimality, diversity, and density in a class of problems where social objectives are rarely included.

What carries the argument

The load-bearing object is the multi-objective Markov decision process formulation of the supply chain: the state includes inventory, outstanding orders, cumulative emissions, and average service-level inequality; actions are per-period manufacturing and delivery quantities; and rewards form a three-dimensional vector of profit, negative emissions, and negative service-level inequality, with an inventory-penalty term. MORL/D is the central mechanism: it decomposes the three objectives into scalarised subproblems, maintains a population of policies with an external Pareto archive, uses the multi-objective soft actor-critic algorithm as the underlying continuous-action policy, optionally adapts weights via Pareto simulated annealing, and transfers experience between neighbouring subproblems through a shared buffer, producing a dense, diverse approximation of the Pareto front in a single run.

What would settle it

Run the simple network to exact Pareto optimality by exhaustive enumeration over its eight action dimensions, then recompute hypervolume, sparsity, and averaged Hausdorff distance using a fixed, independently generated reference front; if MORL/D's hypervolume no longer ranks as the best balanced trade-off or its density advantage shrinks below the reported eleven-fold, the central balance claim fails.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that MORL/D generalises an earlier case-specific study to arbitrary network topologies and, with a shared experience buffer that lets subproblem policies exchange experience, offers the best tested balance of optimality, diversity, and density. The paper reports that MORL/D's hypervolume and expected utility metric rise steadily during training while sparsity falls, that the shared buffer improves hypervolume, EUM, and averaged Hausdorff distance in all three networks, and that Pareto simulated annealing weight adaptation helps in the more complex settings. It also finds that the apparent hypervolume winner, weighted-sum PPO, is actually the sparsest and becomes unstable in EUM as complexity grows, and that NSGA-II concentrates solutions in narrow, lower-quality regions. The conclusion is that decomposition-based MORL with experience sharing is the recommended approach for this generalised class of sustainable, non-stationary supply chain problems.

Load-bearing premise

The comparison assumes that the 'true Pareto front' formed by merging all algorithms' own solution sets is an unbiased reference, and if that reference is skewed by how many points each algorithm returns or by the hand-chosen hypervolume reference points, the headline claims about hypervolume and density do not establish that MORL/D is more balanced.

Editorial extensions

If this is right

  • The same MOMDP formulation can be reused across simple, moderate, and complex supply chain networks without rebuilding the model, because its sets and transition functions generalise over echelons, facilities, and markets.
  • The shared experience buffer is a recommended enhancement: the paper reports it improves hypervolume, EUM, and averaged Hausdorff distance while lowering sparsity across all tested network complexities.
  • Hypervolume alone is not a reliable robustness signal in this setting, because weighted-sum PPO scores highest on hypervolume while being the sparsest and most unstable on expected utility.
  • NSGA-II's concentration in narrow feasible regions, especially in the 5,900-decision-variable complex network, suggests that MOEA-based baselines need larger populations or different search operators before they can compete in this class of problems.
  • MORL/D's reduced sensitivity to demand fluctuations yields more stable production and inventory levels and lower demand loss, although it may sacrifice some precision in lean-inventory management.

Reading between the lines

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

  • Editorial inference: the paper's 'true Pareto front' is assembled from the algorithms' own solution sets, so an independent reference constructed by exhaustive enumeration on the simple network or by a much larger external archive would be a stricter test of the reported hypervolume and density gaps.
  • Editorial inference: the eleven-fold density advantage partly follows from MORL/D returning 30 to 70 points while weighted-sum PPO returns a fixed 21; normalising density by archive size or reporting density per unit hypervolume would give a fairer operational comparison.
  • Editorial inference: because profit is far larger in scale than the other two objectives, hypervolume and density comparisons are sensitive to the chosen reference points and objective normalisation; the balance claim would be most credible if replicated with several reference points and normalisation schemes.
  • Editorial inference: a testable extension would run the same MOMDP environment under other MORL algorithms or longer, non-stationary demand regimes to see whether MORL/D's balanced-front advantage persists beyond the three instantiations in the paper.
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

6 major / 6 minor

Summary. The paper develops a generalized multi-objective, multi-echelon supply chain (SC) model cast as a multi-objective Markov decision process (MOMDP), together with a customisable SC simulator called Messiah. The authors compare three solution approaches: MORL/D (decomposition-based multi-objective reinforcement learning) with and without shared experience buffer and Pareto simulated annealing, a modified single-objective PPO with weighted-sum scalarisation, and NSGA-II. Experiments are run on three SC network configurations of increasing complexity, with objectives of profit maximisation, GHG emission minimisation, and service-level inequality minimisation. The central claim is that MORL/D provides the most balanced Pareto-front approximation in terms of optimality, diversity, and density, and that a shared experience buffer improves its performance. The abstract further claims up to 75% higher hypervolume than the MOEA-based method and roughly eleven times denser solutions than the weighted-sum PPO in complex settings.

Significance. If the comparative evaluation were fully sound, the paper would make a useful contribution by demonstrating a decomposition-based MORL algorithm with a shared experience buffer on a reasonably realistic, multi-echelon, multi-objective SC problem with three sustainability pillars. The model formulation is explicit and the Messiah simulator is potentially a reusable environment. However, the evidence for the central 'most balanced' claim is weakened by several methodological issues: the AHD reference front is constructed from the algorithms' own outputs, hypervolume reference points are hand-chosen without sensitivity analysis, solution-set cardinalities differ substantially across methods, and the quantitative claims in the abstract are not backed by any table in the text. The paper ships no code, data, or full parameter tables, so the experiments cannot be reproduced. These issues are fixable within the scope of a revision, but the current empirical support is not robust enough for the headline conclusions.

major comments (6)
  1. [Section 4.3, Eq. (32)] The 'true PF' used for AHD is defined by 'merging all resultant solution sets across all algorithm runs and then selecting the best PF approximation set using Equation (15)'. This reference set is not independent of the compared algorithms. MORL/D is explicitly designed to produce diverse solutions, so any region that it newly populates is added to the reference and then penalises other algorithms in the inverted generational distance term. Consequently, the AHD results in Figure 7 and the 'most thoughtful balance' conclusion in Section 5.2.1 are partly circular. Please compute AHD against an independent reference front (for example, from a much larger multi-start run, a dedicated global solver, or a leave-one-algorithm-out cross-validation), and report GD and IGD separately.
  2. [Section 5.1.1] The hypervolume reference points are chosen by hand as 'slightly lower than the minimum possible points', with values {0, -2e5, -100}, {0, -4e5, -200}, and {0, -1e6, -500}. No sensitivity analysis is reported, even though the text itself notes that the profit scale is much larger than the other objectives and can dominate hypervolume. The ranking between PPO and MORL/D could change under alternative reference points or after objective normalisation. Please provide a sensitivity analysis over reference points, report hypervolume with normalised objectives, and include numeric values with standard errors. The abstract's '75% higher hypervolume' claim needs a concrete table reference.
  3. [Section 3.3.4, Eq. (24)] The reward penalty is defined as rho_t = sum_j min(I_tj, 0)*M. When inventory is negative, min(I_tj,0) is negative, so rho_t is non-positive. The reward vector is then defined as {Profit_t, -E_t, -F_t} * rho_t, which multiplies all rewards by a non-positive scalar. This flips the intended sign of every objective: profit becomes negative, while -E_t and -F_t become positive, so maximising the total return would maximise emissions and inequality rather than minimise them. This is inconsistent with the optimisation model's constraint (10) and with the stated maximisation convention. Please correct the reward definition (likely an additive penalty or a positive scaling with a sign-preserving structure) and verify that the RL training signal is not corrupted.
  4. [Section 5.2.2 and Table 8] The compared methods produce solution sets of very different sizes: PPO yields 21 points per set, MORL/D yields 30 to 70 points, and NSGA-II yields 50 to 130 points. Hypervolume and sparsity are sensitive to the number of points in the front, so the density comparisons in Figure 6 and the abstract's 'approximately eleven times denser' statement are not controlled for cardinality. AHD is introduced to mitigate this, but its reference set is the problematic circular construct noted above. Please equalise the number of points per front (e.g., by subsampling to a common cardinality) or use indicators that are unbiased with respect to cardinality, and report how the results change with front size.
  5. [Section 5.2 and results] The experimental results are reported only as figures; the precise quantitative claims in the abstract ('up to 75% higher hypervolume' and 'approximately eleven times denser') do not appear in the main text or in any table. In addition, Section 5.2.2 states that the authors 'select the optimal PF approximation sets for each approach based on their hypervolume analysis' before visual comparison, which can bias the comparison even if the intention is only illustrative. Please add a table with means and standard deviations across all runs for hypervolume, sparsity, EUM, and AHD for each scenario and algorithm, state the number of independent runs, and replace the best-run selection with an analysis of the full distribution of results.
  6. [Section 5.1.1 and Section 4.2.3] The paper refers to a supplementary material with 'the values of the detailed parameters' but the arXiv submission does not include it, and the Messiah simulator is described extensively but not released. No code, data, or seeds are provided. Given the number of hyperparameters and the in-house simulator, the experiments are not reproducible. Please provide the simulator code or a link, the complete parameter tables, and all seeds, or clearly state the conditions under which the implementation is available.
minor comments (6)
  1. [Section 3.1, Eq. (11)] The summation limits in Eq. (11) are written as 'sum_{t=(t-L)}^T', where the index t also appears in the lower limit; this is not well-formed. Please clarify the intended time window for the service-level calculation. A similar notational issue appears in Eq. (28).
  2. [Section 4.2.1] The text says 21 pre-defined weight vectors 'whose sum values range from 0 to 1'; this should read 'whose components sum to 1', since the weights define a convex combination. The current phrasing is ambiguous.
  3. [Section 4.3] The hypervolume indicator is described as 'monotonic'; the standard technical property is that it is strictly Pareto-compliant. The weaker word can be misleading in a comparative study.
  4. [Section 5.2.2] The statement that the authors 'select the optimal PF approximation sets for each approach based on their hypervolume analysis' is not a statistically robust way to compare algorithms. Even if it is meant only for illustration, the selection criterion should be reported and the full set of runs should be shown in the supplementary figures.
  5. [Table 8 and Section 5.1.2] The computational time comparison mixes runs with different numbers of seeds and different ways of constructing a final front (e.g., PPO runs 21 weighted policies per seed, while MORL/D and NSGA-II run once per seed). A fairer comparison would report the total compute budget needed to produce a final front of a fixed size.
  6. [References] Several references are incomplete: the Gymnasium citation in Section 4.2.3 lacks a year and venue, and the Reymond et al. (2022) citation lacks page numbers or a DOI. Please check that all references are complete and consistent.

Circularity Check

1 steps flagged · score 3.0 of 10

AHD's internal 'true PF' is the non-dominated union of the algorithms' own outputs, making one evaluation leg self-referential; the central 'balance' claim still has independent support.

  1. self definitional [Section 4.3, Performance Measurements (AHD, Eq. (32)); used in Section 5.2.1 (Figure 7) and Section 6.]
    "Each distance is computed between each solution of the resultant PF approximation sets and the true PF set. In this research, the true PF is estimated by merging all resultant solution sets across all algorithm runs and then selecting the best PF approximation set using Equation (15)."

    The AHD indicator in Eq. (32) evaluates each algorithm's front X against the reference set Y. Section 4.3 defines Y as the non-dominated subset of the union of the three algorithms' own solution sets. Consequently, any non-dominated solution returned by an algorithm is, by construction, an element of Y: that algorithm's generational distance to Y is zero for those points, and each region it newly populates reduces its inverted generational distance relative to algorithms that do not cover that region. The reference yardstick is thus built from the very outputs it is used to judge, so AHD-based statements about front proximity are not independent comparisons.

full rationale

Most of the paper's derivation chain is not circular. The MOMDP model, rewards, and constraints are stated explicitly in Sections 3.2-3.3 and are a direct extension of the authors' prior case study, but they are not derived from that prior work by assumption-forcing citations; the equations are self-contained in this paper. The MORL/D, PPO, and NSGA-II implementations are drawn from external libraries and external algorithm papers, and the comparison is empirical rather than a fitted-parameter prediction. The one genuinely self-referential step is the AHD benchmark: Section 4.3 defines the 'true PF' as the non-dominated union of all algorithms' own output sets, and Eq. (32) then measures each algorithm against that internal reference. That makes the AHD leg of the evaluation partly circular, because an algorithm's own non-dominated points are in the reference by construction. However, the paper's headline that MORL/D gives the most balanced trade-off does not reduce to this step: it is supported by hypervolume trends, sparsity, EUM, visual front inspection, and operational behaviour (production stability, demand loss), and the paper itself reports that PPO attains higher hypervolume and lower AHD in several scenarios. The hand-picked hypervolume reference points are a sensitivity risk but not a circularity, so I do not score them as a separate circular step. I also note that Section 5 opens with the self-declared preliminary status, 'This section discusses the preliminary implementation that has been done so far,' which tempers the strength of the conclusions but is not a circularity. Overall this is a partial, contained self-referentiality in one performance indicator rather than a central derivation that collapses into its inputs.

Assumptions & free parameters 3 free parameters · 4 assumptions · 1 invented entities

The central claims depend on several hand-set evaluation choices and on proprietary simulation infrastructure. The model itself introduces no new physical or mathematical entities, but the results cannot be reproduced independently because the simulator and detailed parameters are not public.

free parameters (3)
  • Hypervolume reference points = {0, -2e5, -100}, {0, -4e5, -200}, {0, -1e6, -500}
    Chosen in Section 5.1.1 as 'slightly lower points than the minimum possible points'; they directly affect hypervolume values and the headline percentage comparisons.
  • Big-M penalty coefficient (rho) = unspecified
    Equation (24) multiplies negative inventory by M, a 'very large number'; the value is not given, and it shapes agent behavior and reward magnitudes.
  • Demand distribution parameters = Normal(mu=150,sigma=60; mu=100,sigma=40); Poisson(lambda=200,100,150) with sinusoidal multipliers
    These hand-set scenario parameters define the non-stationary demand in all experiments; the comparative results are tied to these particular distributions.
assumptions (4)
  • domain assumption State transitions are deterministic (Section 3.3.1: 'no inherent randomness is introduced')
    Removes lead-time and yield uncertainty; if real supply chains have stochastic transitions, the learned policies may not transfer.
  • domain assumption Supply from suppliers is unlimited (Section 5.1.1)
    Constraints out supplier-side scarcity, which can change optimal production and delivery policies.
  • ad hoc to paper The 'true PF' can be estimated by the union of the algorithms' own solution sets (Section 4.3)
    Used for AHD; makes quality assessment self-referential and favors algorithms that sample densely.
  • domain assumption Big-M penalty in the reward is sufficient to enforce inventory non-negativity (Eq. 24)
    The finite penalty replaces a hard constraint and can be violated in practice if M is too small.
invented entities (1)
  • Messiah simulator
    purpose: Customisable multi-echelon supply chain environment for MDP-based RL training and evaluation
    Proprietary to Peak AI, no public release, so all experimental results depend on an inaccessible tool.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Reinforcement Learning for Multi-Objective Multi-Echelon Supply Chain Optimisation." pith.science (2026). https://pith.science/paper/DIRPX2YO

@misc{pith2026250719788,
  author       = {Pith},
  title        = {Pith review of: Reinforcement Learning for Multi-Objective Multi-Echelon Supply Chain Optimisation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DIRPX2YO}},
  note         = {Machine review of arXiv:2507.19788}
}
read the original abstract

This study develops a generalised multi-objective, multi-echelon supply chain optimisation model with non-stationary markets based on a Markov decision process, incorporating economic, environmental, and social considerations. The model is evaluated using a multi-objective reinforcement learning (RL) method, benchmarked against an originally single-objective RL algorithm modified with weighted sum using predefined weights, and a multi-objective evolutionary algorithm (MOEA)-based approach. We conduct experiments on varying network complexities, mimicking typical real-world challenges using a customisable simulator. The model determines production and delivery quantities across supply chain routes to achieve near-optimal trade-offs between competing objectives, approximating Pareto front sets. The results demonstrate that the primary approach provides the most balanced trade-off between optimality, diversity, and density, further enhanced with a shared experience buffer that allows knowledge transfer among policies. In complex settings, it achieves up to 75\% higher hypervolume than the MOEA-based method and generates solutions that are approximately eleven times denser, signifying better robustness, than those produced by the modified single-objective RL method. Moreover, it ensures stable production and inventory levels while minimising demand loss.

Figures

Figures reproduced from arXiv: 2507.19788 by the authors.

Figure 1
Figure 1. Generalised SC network consisting of S number of suppliers, M number of manufacturers, ω echelons of warehouses with customisable number at each echelon, D number of distribution centre, and R number of retailers and markets. The warehouses and distribution centres are optional, defining the number of the whole network’s echelon number. This section presents the SC challenges addressed in this study. Our formulation… view at source ↗
Figure 2
Figure 2. In NSGA-II, an individual solution is depicted by a vector with the size of total [PITH_FULL_IMAGE:figures/full_fig_p018_2.png] view at source ↗
Figure 3
Figure 3. Agent-SC environment interaction along the time steps enabling more informed sequential [PITH_FULL_IMAGE:figures/full_fig_p019_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: In a MOMDP, an action is represented by a vector whose size corresponds to the total [PITH_FULL_IMAGE:figures/full_fig_p019_4.png]
Figure 5
Figure 5. Figure 5: The three examples of multi-echelon SC networks simulated in this work. The network [PITH_FULL_IMAGE:figures/full_fig_p024_5.png]
Figure 6
Figure 6. Figure 6: The overall performance of the three methods. PPO consistently shows the highest [PITH_FULL_IMAGE:figures/full_fig_p029_6.png]
Figure 7
Figure 7. Figure 7: AHD values of all algorithms expressed on logarithmic scales. MORL/D AHD exhibits the [PITH_FULL_IMAGE:figures/full_fig_p031_7.png]
Figure 8
Figure 8. Figure 8: The resulting PF approximation sets of the three methods. MORL/D shows the most [PITH_FULL_IMAGE:figures/full_fig_p032_8.png]
Figure 9
Figure 9. Figure 9: The comparison of manufacturing quantities resulted from the three algorithms in each [PITH_FULL_IMAGE:figures/full_fig_p034_9.png]
Figure 10
Figure 10. Figure 10: Inventory build-up coincides with increasing complexity across all algorithms. In simpler [PITH_FULL_IMAGE:figures/full_fig_p035_10.png]
Figure 11
Figure 11. Figure 11: NSGA-II outperforms the other two in terms of demand fulfilment. MORL/D exhibits [PITH_FULL_IMAGE:figures/full_fig_p036_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

77 extracted references · 60 canonical work pages

  1. [1]

    , author Salehi Esfandarani, M

    author Abdolazimi, O. , author Salehi Esfandarani, M. , author Salehi, M. , author Shishebori, D. , year 2020 . title Robust design of a multi-objective closed-loop supply chain by integrating on-time delivery, cost, and environmental aspects, case study of a Tire Factory . journal Journal of Cleaner Production volume 264 , pages 121566 . https://linkingh...

    work page arXiv 2020

  2. [2]

    , author Pham, C

    author Abu Zwaida, T. , author Pham, C. , author Beauregard, Y. , year 2021 . title Optimization of Inventory Management to Prevent Drug Shortages in the Hospital Supply Chain . journal Applied Sciences volume 11 , pages 2726 . https://www.mdpi.com/2076-3417/11/6/2726, :10.3390/app11062726

    work page doi:10.3390/app11062726 2021

  3. [3]

    , author Searcy, C

    author Ahi, P. , author Searcy, C. , year 2013 . title A comparative literature analysis of definitions for green and sustainable supply chain management . journal Journal of Cleaner Production volume 52 , pages 329--341 . https://linkinghub.elsevier.com/retrieve/pii/S095965261300067X, :10.1016/j.jclepro.2013.02.018

    work page doi:10.1016/j.jclepro.2013.02.018 2013

  4. [4]

    , author Bazzan, A.L.C

    author Alegre, L.N. , author Bazzan, A.L.C. , author Roijers, D.M. , author Nowé, A. , author da Silva, B.C. , year 2023 . title Sample- Efficient Multi - Objective Learning via Generalized Policy Improvement Prioritization https://arxiv.org/abs/2301.07784, :10.48550/ARXIV.2301.07784. note publisher: arXiv Version Number: 2

    arXiv doi:10.48550/arxiv.2301.07784 2023

  5. [5]

    , author Guo, Y

    author Allaoui, H. , author Guo, Y. , author Choudhary, A. , author Bloemhof, J. , year 2018 . title Sustainable agro-food supply chain design using two-stage hybrid multi-objective decision-making approach . journal Computers & Operations Research volume 89 , pages 369--384 . https://linkinghub.elsevier.com/retrieve/pii/S030505481630260X, :10.1016/j.cor....

    work page doi:10.1016/j.cor.2016.10.012 2018

  6. [6]

    , author Mateus, G.R

    author Alves, J.C. , author Mateus, G.R. , year 2020 . title Deep Reinforcement Learning and Optimization Approach for Multi -echelon Supply Chain with Uncertain Demands , in: editor Lalla-Ruiz, E. , editor Mes, M. , editor Voß, S. (Eds.), booktitle Computational Logistics . publisher Springer International Publishing , address Cham . volume volume 12433 ...

    work page doi:10.1007/978-3-030-59747-4_38 2020

  7. [7]

    , author Torabi, S.A

    author Asgharizadeh, E. , author Torabi, S.A. , author Mohaghar, A. , author Zare-Shourijeh, M.A. , year 2019 . title Sustainable Supply Chain Network Design : A Review on Quantitative Models Using Content Analysis . journal Environmental Energy and Economic Research volume 3 . https://doi.org/10.22097/eeer.2019.184458.1081, :10.22097/eeer.2019.184458.1081

    work page arXiv 2019

  8. [8]

    , author Goldsby, T.J

    author Autry, C.W. , author Goldsby, T.J. , author Bell, J. , author Moon, M.A. , author Munson, C. , author Watson, M. , author Lewis, S. , author Cacioppi, P. , author Jayaraman, J. , year 2013 . title The Definitive Guide to Modern Supply Chain Management ( Collection ) . publisher Pearson Education , address New Jersey

    work page 2013

Show all 77 references
  1. [9]

    , author Fatemi Ghomi, S

    author Badri, H. , author Fatemi Ghomi, S. , author Hejazi, T.H. , year 2017 . title A two-stage stochastic programming approach for value-based closed-loop supply chain network design . journal Transportation Research Part E: Logistics and Transportation Review volume 105 , p...

    2017 doi

  2. [10]

    , author Kanellopoulos, A

    author Banasik, A. , author Kanellopoulos, A. , author Claassen, G. , author Bloemhof-Ruwaard, J.M. , author van der Vorst, J.G. , year 2017 . title Closing loops in agricultural supply chains using multi-objective optimization: A case study of an industrial mushroom supply ch...

    2017 doi

  3. [11]

    , author Mula, J

    author Becerra, P. , author Mula, J. , author Sanchis, R. , year 2021 . title Green supply chain quantitative models for sustainable inventory management: A review . journal Journal of Cleaner Production volume 328 , pages 129544 . https://linkinghub.elsevier.com/retrieve/pii/...

    2021

  4. [12]

    , author Deb, K

    author Blank, J. , author Deb, K. , year 2020 . title Pymoo: Multi - Objective Optimization in Python . journal IEEE Access volume 8 , pages 89497--89509 . https://ieeexplore.ieee.org/document/9078759/, :10.1109/ACCESS.2020.2990567

    2020

  5. [13]

    , author Gijsbrechts, J

    author Boute, R.N. , author Gijsbrechts, J. , author van Jaarsveld, W. , author Vanvuchelen, N. , year 2021 . title Deep Reinforcement Learning for Inventory Control : A Roadmap . journal SSRN Electronic Journal https://www.ssrn.com/abstract=3861821, :10.2139/ssrn.3861821

    2021 doi

  6. [14]

    , author Warsing, D.P

    author Bozarth, C.C. , author Warsing, D.P. , author Flynn, B.B. , author Flynn, E.J. , year 2009 . title The impact of supply chain complexity on manufacturing plant performance . journal Journal of Operations Management volume 27 , pages 78--93 . http://doi.wiley.com/10.1016...

    2009 doi

  7. [15]

    , year 2008

    author Branke, J. , year 2008 . title Multiobjective Optimization : interactive and evolutionary approaches . Number number 5252 in series Lecture notes in computer science , publisher Springer-Verlag , address Berlin

    2008

  8. [16]

    , author Wang, Y

    author Chen, D. , author Wang, Y. , author Gao, W. , year 2020 . title Combining a gradient-based method and an evolution strategy for multi-objective reinforcement learning . journal Applied Intelligence volume 50 , pages 3301--3317 . https://link.springer.com/10.1007/s10489-...

    2020 doi

  9. [17]

    , author Andresen, S

    author Chen, Z. , author Andresen, S. , year 2014 . title A Multiobjective Optimization Model of Production - Sourcing for Sustainable Supply Chain with Consideration of Social , Environmental , and Economic Factors . journal Mathematical Problems in Engineering volume 2014 , ...

    2014 doi

  10. [18]

    , author Jaszkiewicz, A

    author Czyzżak, P. , author Jaszkiewicz, A. , year 1998 . title Pareto simulated annealing—a metaheuristic technique for multiple‐objective combinatorial optimization . journal Journal of Multi‐Criteria Decision Analysis volume 7 , pages 34--47 . https://onlinelibrary.wiley.co...

    1998 doi

  11. [19]

    , author Hemmati, M

    author Darestani, S.A. , author Hemmati, M. , year 2019 . title Robust optimization of a bi-objective closed-loop supply chain network for perishable goods considering queue system . journal Computers & Industrial Engineering volume 136 , pages 277--292 . https://linkinghub.el...

    2019 doi

  12. [20]

    , author Dennis, J.E

    author Das, I. , author Dennis, J.E. , year 1998 . title Normal- Boundary Intersection : A New Method for Generating the Pareto Surface in Nonlinear Multicriteria Optimization Problems . journal SIAM Journal on Optimization volume 8 , pages 631--657 . http://epubs.siam.org/doi...

    1998 doi

  13. [21]

    , author Pratap, A

    author Deb, K. , author Pratap, A. , author Agarwal, S. , author Meyarivan, T. , year 2002 . title A fast and elitist multiobjective genetic algorithm: NSGA - II . journal IEEE Transactions on Evolutionary Computation volume 6 , pages 182--197 . http://ieeexplore.ieee.org/docu...

    2002

  14. [22]

    , author Catanzaro, D

    author Dehaybe, H. , author Catanzaro, D. , author Chevalier, P. , year 2024 . title Deep Reinforcement Learning for inventory optimization with non-stationary uncertain demand . journal European Journal of Operational Research volume 314 , pages 433--445 . https://linkinghub....

    2024 doi

  15. [23]

    , author Dejax, P

    author Eskandarpour, M. , author Dejax, P. , author Miemczyk, J. , author Péton, O. , year 2015 . title Sustainable supply chain network design: An optimization-oriented review . journal Omega volume 54 , pages 11--32 . https://linkinghub.elsevier.com/retrieve/pii/S03050483150...

    2015 doi

  16. [24]

    , author Farahani, R.Z

    author Fahimnia, B. , author Farahani, R.Z. , author Marian, R. , author Luong, L. , year 2013 . title A review and critique on integrated production–distribution planning models and techniques . journal Journal of Manufacturing Systems volume 32 , pages 1--19 . https://linkin...

    2013 doi

  17. [25]

    , author Makui, A

    author Farrokhi-Asl, H. , author Makui, A. , author Ghousi, R. , author Rabbani, M. , year 2020 . title Developing a hazardous waste management system with consideration of health, safety, and environment . journal Computers & Electrical Engineering volume 82 , pages 106553 . ...

    2020

  18. [26]

    , author Alegre, L.N

    author Felten, F. , author Alegre, L.N. , author Nowé, A. , author Bazzan, A.L.C. , author Talbi, E.G. , author Danoy, G. , author Silva, B.C.d. , year 2023 a. title A toolkit for reliable benchmarking and research in multi-objective reinforcement learning , in: booktitle Proc...

    2023

  19. [27]

    , author Talbi, E.G

    author Felten, F. , author Talbi, E.G. , author Danoy, G. , year 2023 b. title Multi- Objective Reinforcement Learning Based on Decomposition : A Taxonomy and Framework https://arxiv.org/abs/2311.12495, :10.48550/ARXIV.2311.12495. note publisher: arXiv Version Number: 2

    2023 doi arXiv

  20. [28]

    , author Sundararaj, D

    author Ganesan, V.K. , author Sundararaj, D. , author Srinivas, A.P. , year 2019 . title Adaptive Inventory Replenishment for Dynamic Supply Chains with Uncertain Market Demand , publisher Springer , address Singapore . pp. pages 325--335

    2019

  21. [29]

    , author Millman, K.J

    author Harris, C.R. , author Millman, K.J. , author Van Der Walt, S.J. , author Gommers, R. , author Virtanen, P. , author Cournapeau, D. , author Wieser, E. , author Taylor, J. , author Berg, S. , author Smith, N.J. , author Kern, R. , author Picus, M. , author Hoyer, S. , au...

    2020

  22. [30]

    , author Rădulescu, R

    author Hayes, C.F. , author Rădulescu, R. , author Bargiacchi, E. , author Källström, J. , author Macfarlane, M. , author Reymond, M. , author Verstraeten, T. , author Zintgraf, L.M. , author Dazeley, R. , author Heintz, F. , author Howley, E. , author Irissappane, A.A. , auth...

    2022

  23. [31]

    , author Zhen, L

    author Huang, L. , author Zhen, L. , author Yin, L. , year 2020 . title Waste material recycling and exchanging decisions for industrial symbiosis network optimization . journal Journal of Cleaner Production volume 276 , pages 124073 . https://linkinghub.elsevier.com/retrieve/...

    2020

  24. [32]

    , author Agdas, D

    author Jayarathna, C.P. , author Agdas, D. , author Dawes, L. , author Yigitcanlar, T. , year 2021 . title Multi- Objective Optimization for Sustainable Supply Chain and Logistics : A Review . journal Sustainability volume 13 , pages 13617 . https://www.mdpi.com/2071-1050/13/2...

    2021 doi

  25. [33]

    , author von Kleist, H

    author Kemmer, L. , author von Kleist, H. , author de Rochebou¨et, D. , author Tziortziotis, N. , year 2018 . title Reinforcement Learning for Supply Chain Optimization . journal European Workshop on Reinforcement Learning volume 14 , pages 1--9

    2018

  26. [34]

    , author Tavana, M

    author Khalili Nasr, A. , author Tavana, M. , author Alavi, B. , author Mina, H. , year 2021 . title A novel fuzzy multi-objective circular supplier selection and order allocation model for sustainable closed-loop supply chains . journal Journal of Cleaner Production volume 28...

    2021

  27. [35]

    , author Kayvanfar, V

    author Khodaee, V. , author Kayvanfar, V. , author Haji, A. , year 2022 . title A humanitarian cold supply chain distribution model with equity consideration: The case of COVID -19 vaccine distribution in the European Union . journal Decision Analytics Journal volume 4 , pages...

    2022

  28. [36]

    , author Kumar, A

    author Kumar, K. , author Kumar, A. , year 2024 . title Application of Optimization Models in Sustainable Supply Chain Management : A Systematic Review Based on PRISMA Guidelines . journal Process Integration and Optimization for Sustainability https://link.springer.com/10.100...

    2024 doi

  29. [37]

    , author Pitrou, A

    author Lam, S.K. , author Pitrou, A. , author Seibert, S. , year 2015 . title Numba: a LLVM -based Python JIT compiler , in: booktitle Proceedings of the Second Workshop on the LLVM Compiler Infrastructure in HPC , publisher ACM , address Austin Texas . pp. pages 1--6 . https:...

    2015

  30. [38]

    , author Sikora, R

    author Lee, Y.S. , author Sikora, R. , year 2019 . title Application of adaptive strategy for supply chain agent . journal Information Systems and e-Business Management volume 17 , pages 117--157 . http://link.springer.com/10.1007/s10257-018-0378-y, :10.1007/s10257-018-0378-y

    2019 doi

  31. [39]

    , year 1998

    author Miettinen, K. , year 1998 . title Nonlinear Multiobjective Optimization . Number number v.12 in series International Series in Operations Research and Management Science Ser . edition 1st ed ed., publisher Springer , address Boston

    1998

  32. [40]

    , author Kelle, P

    author Modak, N.M. , author Kelle, P. , year 2021 . title Using social work donation as a tool of corporate social responsibility in a closed-loop supply chain considering carbon emissions tax and demand uncertainty . journal Journal of the Operational Research Society volume ...

    2021

  33. [41]

    , author Gomes, M.I

    author Mota, B. , author Gomes, M.I. , author Carvalho, A. , author Barbosa-Povoa, A.P. , year 2015 . title Towards supply chain sustainability: economic, environmental and social design and planning . journal Journal of Cleaner Production volume 105 , pages 14--27 . https://l...

    2015 doi

  34. [42]

    , author Nazari, M

    author Oroojlooyjadid, A. , author Nazari, M. , author Snyder, L.V. , author Tak´aˇc, M. , year 2017 . title A Deep Q - Network for the Beer Game : A Reinforcement Learning Algorithm to Solve Inventory Optimization Problems , pages 1--39

    2017

  35. [43]

    , author Nayak, M.M

    author Pattnaik, S. , author Nayak, M.M. , author Abbate, S. , author Centobelli, P. , year 2021 . title Recent Trends in Sustainable Inventory Models : A Literature Review . journal Sustainability volume 13 , pages 11756 . https://www.mdpi.com/2071-1050/13/21/11756, :10.3390/...

    2021 doi

  36. [44]

    , year 2020

    author Peng, H. , year 2020 . title A Comprehensive Overview and Survey of Recent Advances in Meta - Learning . https://arxiv.org/abs/2004.11149, :10.48550/ARXIV.2004.11149. note version Number: 7

    2020 doi arXiv

  37. [45]

    , author Hubbs, C.D

    author Perez, H.D. , author Hubbs, C.D. , author Li, C. , author Grossmann, I.E. , year 2021 . title Algorithmic Approaches to Inventory Management Optimization . journal Processes volume 9 , pages 102 . https://www.mdpi.com/2227-9717/9/1/102, :10.3390/pr9010102

    2021 doi

  38. [46]

    , author Das, T.K

    author Pontrandolfo, P. , author Das, T.K. , author Okogbaa, O.G. , author Gosavi, A. , year 2002 . title Global Supply Chain Management : A reinforcement Learning Approach . journal International Journal of Production Research , pages 1--27 :10.1080/00207540110118640

    2002 doi

  39. [47]

    , author Tingey, J

    author Rachman, R. , author Tingey, J. , author Allmendinger, R. , author Shukla, P. , author Pan, W. , year 2025 . title Multi-objective Sequential Decision Making for Holistic Supply Chain Optimization , in: editor Singh, H. , editor Ray, T. , editor Knowles, J. , editor Li,...

    2025

  40. [48]

    , author Hill, A

    author Raffin, A. , author Hill, A. , author Gleave, A. , author Kanervisto, A. , author Ernestus, M. , author Dormann, N. , year 2021 . title Stable-baselines3: Reliable reinforcement learning implementations . journal Journal of Machine Learning Research volume 22 , pages 1-...

    2021

  41. [49]

    , author Bargiacchi, E

    author Reymond, M. , author Bargiacchi, E. , author Nowé, A. , year 2022 . title Pareto Conditioned Networks , in: booktitle Pareto Conditioned Networks

    2022

  42. [50]

    , author Chou, S.Y

    author Rizqi, Z.U. , author Chou, S.Y. , year 2024 . title Neuroevolution reinforcement learning for multi-echelon inventory optimization with delivery options and uncertain discount . journal Engineering Applications of Artificial Intelligence volume 134 , pages 108670 . http...

    2024

  43. [51]

    , author Whiteson, S

    author Roijers, D.M. , author Whiteson, S. , author Vamplew, P. , author Dazeley, R. , year 2013 . title A Survey of Multi - Objective Sequential Decision - Making . journal Journal of Artificial Intelligence Research volume 48 , pages 63--113

    2013

  44. [52]

    , author Jackson, I

    author Rolf, B. , author Jackson, I. , author Müller, M. , author Lang, S. , author Reggelin, T. , author Ivanov, D. , year 2022 . title A review on reinforcement learning algorithms and applications in supply chain management . journal International Journal of Production Rese...

    2022

  45. [53]

    , author Levine, S

    author Schulman, J. , author Levine, S. , author Abbeel, P. , author Jordan, M. , author Moritz, P. , year 2015 . title Trust region policy optimization , in: editor Bach, F. , editor Blei, D. (Eds.), booktitle Proceedings of the 32nd international conference on machine learni...

    2015

  46. [54]

    , author Wolski, F

    author Schulman, J. , author Wolski, F. , author Dhariwal, P. , author Radford, A. , author Klimov, O. , year 2017 . title Proximal Policy Optimization Algorithms https://arxiv.org/abs/1707.06347, :10.48550/ARXIV.1707.06347. note publisher: arXiv Version Number: 2

    2017 doi arXiv

  47. [55]

    , author Esquivel, X

    author Schutze, O. , author Esquivel, X. , author Lara, A. , author Coello, C.A.C. , year 2012 . title Using the Averaged Hausdorff Distance as a Performance Measure in Evolutionary Multiobjective Optimization . journal IEEE Transactions on Evolutionary Computation volume 16 ,...

    2012

  48. [56]

    , author Aliyev, R

    author Seyyedabbasi, A. , author Aliyev, R. , author Kiani, F. , author Gulle, M.U. , author Basyildiz, H. , author Shah, M.A. , year 2021 . title Hybrid algorithms based on combining reinforcement learning and metaheuristic methods to solve global optimization problems . jour...

    2021

  49. [57]

    , author Wang, H

    author Shar, I.E. , author Wang, H. , author Gupta, C. , year 2023 . title Multi- Objective Reinforcement Learning for Sustainable Supply Chain Optimization , in: booktitle 2023 IEEE 19th International Conference on Automation Science and Engineering ( CASE ) , publisher IEEE ...

    2023

  50. [58]

    , author Hosseini-Motlagh, S.M

    author Sharifi, M. , author Hosseini-Motlagh, S.M. , author Samani, M.R.G. , author Kalhor, T. , year 2020 . title Novel resilient-sustainable strategies for second-generation biofuel network design considering Neem and Eruca Sativa under hybrid stochastic fuzzy robust approac...

    2020

  51. [59]

    , author Verma, A

    author Singh, D. , author Verma, A. , year 2018 . title Inventory Management in Supply Chain . journal Materials Today: Proceedings volume 5 , pages 3867--3872 . https://linkinghub.elsevier.com/retrieve/pii/S2214785317329140, :10.1016/j.matpr.2017.11.641

    2018 doi

  52. [60]

    , author Schwind, M

    author Stockheim, T. , author Schwind, M. , author Koenig, W. , year 2003 . title A Reinforcement Learning Approach for Supply Chain Management

    2003

  53. [61]

    , author Barto, A.G

    author Sutton, R.S. , author Barto, A.G. , year 2015 . title Reinforcement Learning : An Introduction . edition 1st ed., publisher The MIT Press , address Massachusetts

    2015

  54. [62]

    , author Nieto-Isaza, S

    author Tariq Afridi, M. , author Nieto-Isaza, S. , author Ehm, H. , author Ponsignon, T. , author Hamed, A. , year 2020 . title A Deep Reinforcement Learning Approach for Optimal Replenishment Policy in A Vendor Managed Inventory Setting For Semiconductors , in: booktitle 2020...

    2020

  55. [63]

    , author Barbosa-Povoa, A.P

    author Tautenhain, C.P. , author Barbosa-Povoa, A.P. , author Mota, B. , author Nascimento, M.C. , year 2021 . title An efficient Lagrangian -based heuristic to solve a multi-objective sustainable supply chain problem . journal European Journal of Operational Research volume 2...

    2021 doi

  56. [64]

    , author Olsson, O

    author Tenggren, S. , author Olsson, O. , author Vulturius, G. , author Carlsen, H. , author Benzie, M. , year 2020 . title Climate risk in a globalized world: empirical findings from supply chains in the Swedish manufacturing sector . journal Journal of Environmental Planning...

    2020

  57. [65]

    , author Goli, A

    author Tirkolaee, E.B. , author Goli, A. , author Faridnia, A. , author Soltani, M. , author Weber, G.W. , year 2020 . title Multi-objective optimization for the reliable pollution-routing problem with cross-dock selection using Pareto -based algorithms . journal Journal of Cl...

    2020

  58. [66]

    , author Terry, J.K

    author Towers, M. , author Terry, J.K. , author Kwiatkowski, A. , author Balis, J.U. , author de Cola, G. , author Deleu, T. , author Goulão, M. , author Kallinteris, A. , author KG, A. , author Krimmel, M. , author Perez-Vicente, R. , author Pierré, A. , author Schulhoff, S. ...

  59. [67]

    , author Doan, L

    author Tran, H.L. , author Doan, L. , author Luong, N.H. , author Binh, H.T.T. , year 2023 . title A Two - Stage Multi - Objective Evolutionary Reinforcement Learning Framework for Continuous Robot Control , in: booktitle Proceedings of the Genetic and Evolutionary Computation...

    2023

  60. [68]

    , author Sauvey, C

    author Turki, S. , author Sauvey, C. , author Rezg, N. , year 2018 . title Modelling and optimization of a manufacturing/remanufacturing system with storage facility under carbon cap and trade policy . journal Journal of Cleaner Production volume 193 , pages 441--458 . https:/...

    2018 doi

  61. [69]

    , author Shi, Q

    author Wang, Y. , author Shi, Q. , author Hu, Q. , author You, Z. , author Bai, Y. , author Guo, C. , year 2020 . title An efficiency sorting multi-objective optimization framework for sustainable supply network optimization and decision making . journal Journal of Cleaner Pro...

    2020

  62. [70]

    (Ed.), year 2013

    editor Watson, M. (Ed.), year 2013 . title Supply chain network design: applying optimization and analytics to the global supply chain . publisher FT Press , address Upper Saddle River, N.J . note OCLC: ocn793220883

    2013

  63. [71]

    , author Tian, Y

    author Xu, J. , author Tian, Y. , author Ma, P. , author Rus, D. , author Sueda, S. , author Matusik, W. , year 2020 . title Prediction-guided multi-objective reinforcement learning for continuous robot control , in: editor III, H.D. , editor Singh, A. (Eds.), booktitle Procee...

    2020

  64. [72]

    , author Chow, A.H

    author Yan, Y. , author Chow, A.H. , author Ho, C.P. , author Kuo, Y.H. , author Wu, Q. , author Ying, C. , year 2022 . title Reinforcement Learning for Logistics and Supply Chain Management : Methodologies , State of the Art , and Future Opportunities . journal Transportation...

    2022

  65. [73]

    , author Sun, X

    author Yang, R. , author Sun, X. , author Narasimhan, K. , year 2019 . title A Generalized Algorithm for Multi - Objective Reinforcement Learning and Policy Adaptation https://arxiv.org/abs/1908.08342, :10.48550/ARXIV.1908.08342. note publisher: arXiv Version Number: 2

    2019 doi arXiv

  66. [74]

    , author Lou, J

    author Zhao, T. , author Lou, J. , year 2025 . title Research on the application of deep learning in low-carbon supply chain management . journal International Journal of Low-Carbon Technologies volume 20 , pages 209--216 . https://academic.oup.com/ijlct/article/doi/10.1093/ij...

    2025 doi

  67. [75]

    , author Kanters, T.V

    author Zintgraf, L.M. , author Kanters, T.V. , author Roijers, D.M. , author Oliehoek, F.A. , author Beau, P. , year 2015 . title Quality Assessment of MORL Algorithms : A Utility - Based Approach , in: booktitle Proceedings of the 24th Annual Machine Learning Conference of Be...

    2015

  68. [76]

    , author Brockhoff, D

    author Zitzler, E. , author Brockhoff, D. , author Thiele, L. , year 2007 . title The Hypervolume Indicator Revisited : On the Design of Pareto -compliant Indicators Via Weighted Integration , in: editor Obayashi, S. , editor Deb, K. , editor Poloni, C. , editor Hiroyasu, T. ,...

    2007

  69. [77]

    , author Thiele, L

    author Zitzler, E. , author Thiele, L. , year 1998 . title Multiobjective optimization using evolutionary algorithms — A comparative case study , in: editor Goos, G. , editor Hartmanis, J. , editor Van Leeuwen, J. , editor Eiben, A.E. , editor Bäck, T. , editor Schoenauer, M. ...

    1998

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.