Corpus 1,272 assessed · 1,173 scored · 643 reproduced ≥75 · 168 flagged ·∅ 74.1/100
← New search

Wireless sensor network design with reliable and long network lifetime.

Sci Rep · 2026
L1 65/100 3/4
Why this verdict

The main results reproduced, with only marginal, non-material deviations.

Reproduced on the brainbox compute brainarbeit.com
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +6
✓ What held up
  • Reported values were directly comparable
What did not (or only partly)
  • 🟡Could not use the authors’ exact input data
  • 🟡A deviation arose in the data or preprocessing
  • 🔴A deviation was attributed to the published material
  • 🟡Reported values were not (fully) derivable from the shared data
  • 🟡The deviation was non-trivial in magnitude
  • 🟡The central claim did not (fully) hold under reproduction
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
65/100
Reproducibility score
0.5 SD below mean
vs. all fields · 1173 studies
🎯 Scores higher than 27% of all assessed papers rank 843 of 1173 scored

A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.

Reproduction agent’s raw note

Described well enough to reproduce. Target = Table 6 SingleCopy (SC) network-lifetime values for the small (9,4) instances (SC=H there), the clearly-specified proven-optimal cells. Method: faithful Python/PuLP re-implementation of SingleCopyModel.cs + inputentry.cs, solved to PROVEN optimality with open solver HiGHS on the paper's own instance files from the Zenodo archive (zenodo 10.5281/zenodo.19135286, sha256 7eb0828...). Since reported values are at gap 0, the optimum is solver-independent, so this is a valid 1:1 reproduction (no Gurobi license needed; brief P16). RESULT: 2 of 3 (9,4) SC values reproduce EXACTLY at proven optimality (31786.30 and 50000.00); 1 does NOT (reproduced proven-optimal 48582.59 vs reported 41395.36). Key audit flag: the article repo and the Zenodo archive ship DIFFERENT instance files for the same Data1/9 slot (verified by SHA256), so 'Data#/9' is not a single canonical instance and the archived Data3/9 need not be the one that produced 41395.36; that reported value is not derivable from any shipped (9,4) instance we solved (flagged for review, NOT asserted as fabrication). NOT attempted (hard 20%): original C#+Gurobi toolchain (license-gated, optima solver-independent anyway); larger instances >=20 sensors (the repo Data1/20 is a different-version instance and HiGHS stayed at ~54% gap; paper Gurobi took 2836s) - dropped per 80/20; DC/H/LH models and the reliability/Monte-Carlo scenarios (Tables 7-9, Fig 9). All download/extract/solve ran inside «our HPC» SLURM jobs on «infra»; only small result values copied to «host».

💻 Code ↗ 🗄 Data: 10.5281/zenodo.19135285

These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.

Assessment versions

Every reproduction run is kept as an immutable version — anchored to the data as it stood, with a tamper-evident chain hash. A rerun (e.g. after an author updates a deposit) adds a new version; the previous one stays on record.

  1. v1 current initial assessment Score 65
    assessed: 2026-06-16 ⛓ 8c769f455e55
✎ I am an author of this paper

Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.

Reason for the rerun

We email you a confirmation link first. The rerun is an objective re-measurement — it cannot change the verdict in your favour, only ask us to look again.

Provenance — full disclosure

When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.

Reproduced
2026-06-16
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-16
no human curator yet
Last updated
2026-08-05

Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.

Deep full-text extraction

Model: opus
Founding hypothesis

Whether the four fundamental WSN design problems (coverage, sink placement/routing, activity scheduling, data routing) together with network reliability can be addressed in a single integrated optimization framework, and which data-transmission strategy best balances long network lifetime against reliability.

Core claims
  • All four fundamental WSN design problems and network reliability can be addressed together in an integrated set of mixed-integer mathematical models. method
  • Three transmission strategies are proposed: Single Copy (SC) for single-path transmission, Double Copy (DC) for multi-copy transmission, and Hybrid (H) where copying occurs in sensors transmitting to central nodes. method
  • The SC strategy is superior for network lifetime, the DC strategy excels in reliability, and the H strategy gives balanced performance in both. finding
  • A Lagrangian Heuristic (LH) enables the H strategy to find solutions efficiently at large scale, using a Dantzig–Wolfe column generation algorithm to solve the Lagrangian subproblem and constructing a feasible solution at each step. method
  • The LH method yields higher network lifetime than the Gurobi solver, especially for medium- and large-scale networks. finding
  • Different scenarios were generated to measure the network reliability of the strategies. method
Experimental setups
Assay System Perturbation Readout Platform
Mixed-integer linear programming optimization (SC/DC/H strategy models) Simulated wireless sensor networks of varying scale none Network lifetime and network reliability Gurobi solver
Lagrangian Heuristic with Dantzig–Wolfe column generation Medium- and large-scale simulated wireless sensor networks none Network lifetime Gurobi solver (for comparison)
Reliability scenario testing Simulated WSNs under various attack/failure scenarios sensor/link failure scenarios Network reliability
Key results
  • SC strategy achieves the longest network lifetime among the three strategies
  • DC strategy provides the highest network reliability
  • H strategy gives balanced lifetime and reliability performance
  • LH method provides higher network lifetime than Gurobi for medium and large-scale networks

Statistical methods review

Model: sonnet

A neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.

This is a mathematical optimization and computational study proposing three mixed-integer linear programming (MILP) strategies—Single Copy (SC), Double Copy (DC), and Hybrid (H)—for integrated wireless sensor network design covering coverage, activity scheduling, sink routing, and data routing problems simultaneously. Evaluation is conducted through computational experiments across 'various scenarios,' comparing strategy performance on network lifetime and reliability metrics. A Lagrangian Heuristic (LH) using Dantzig–Wolfe column generation is additionally proposed and benchmarked against the Gurobi commercial solver by comparing objective function values (network lifetime). No inferential statistical hypothesis tests or p-values are reported; performance assessment is based on direct numerical comparison of optimization outcomes across scenarios.

Replicationunclear Sample sizeDescribed only as 'various scenarios' including small, medium, and large-scale instances; exact scenario count and parameter ranges not stated in the available text GroupsSC strategy vs. DC strategy vs. H strategy; LH heuristic vs. Gurobi solver Pairingna Randomization/blindingnot stated Dispersionnone Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionnone stated
Statistical tests used
Test Applied to n Assumptions
Direct numerical comparison of optimization objective values (network lifetime, reliability) across scenarios Comparison of SC, DC, and H strategies across all generated scenarios not stated
Computational benchmarking: Lagrangian Heuristic vs. Gurobi solver Medium and large-scale network scenarios not stated
Approaches that could also have been used
  • Strategy comparisons (SC vs. DC vs. H) are conducted through direct inspection of objective values across scenarios without statistical testing of performance differences
    Could also: Non-parametric statistical tests (e.g., Wilcoxon signed-rank or Friedman test) could also be applied across the set of scenario instances to assess whether differences in network lifetime are consistent — When multiple problem instances serve as the unit of comparison, a formal test quantifies whether observed differences are systematic across the instance set or driven by a few favorable cases, complementing the direct numerical comparison
  • The Lagrangian Heuristic is benchmarked against Gurobi by comparing solution quality (network lifetime) without reporting a distribution of gaps or confidence bounds on the optimality gap
    Could also: A summary of the optimality gap distribution (mean, median, range) across all test instances could also characterize heuristic quality, as is common in operations research benchmarking studies — Distributional gap reporting allows readers to assess not just central tendency but worst-case and variability of heuristic performance relative to the exact solver
  • Test scenarios are described as 'various' small, medium, and large-scale instances, but the scenario generation process and parameter ranges are not specified in the available text
    Could also: A factorial or Latin Hypercube design of experiments across key parameters (number of sensors, area size, energy budget, coverage demand) could also be used to systematically map the performance landscape — Structured experimental design makes it possible to attribute performance differences to specific problem parameters, aiding generalizability of conclusions
  • Network lifetime and reliability are treated as separate performance dimensions evaluated qualitatively across strategies
    Could also: Multi-objective analysis with a Pareto-frontier characterization could also be used to formally represent the lifetime–reliability trade-off across strategies — A Pareto front makes the trade-off curve explicit and allows quantitative comparison of how much lifetime is sacrificed per unit of reliability gain for each strategy
  • Computational run times are referenced (heuristic solves large-scale instances in 'acceptable time') but no formal computational complexity analysis or timing statistics are reported
    Could also: Reporting wall-clock time with summary statistics (mean, standard deviation) across instances, stratified by problem size, could also provide a reproducible characterization of computational efficiency — Systematic timing data allows practitioners to judge whether the heuristic is practical for their specific scale and hardware, and enables comparison with future methods
Software: Gurobi

Citation network

Where this publication sits in the reproducibility-weighted citation graph — what it is built on, and what is built on it. Citation data from OpenAlex.

Citations
0
Impact: low
Foundation confidence
None of its references are in our reproducibility record yet — its foundation cannot be assessed.
Topics

No assessed neighbours yet — the network grows as more papers are assessed.

What was reproduced

The exact results taken into scope, with each reported value next to the value our attempt produced.

Scope — pmid-41981015

Paper: Çelik E, Keskin ME. "Wireless sensor network design with reliable and long network lifetime." Sci Rep (2026). DOI 10.1038/s41598-026-46014-x. PMCID PMC13083951.

Code: https://github.com/esraaclk64-collab/wsn-rln-article-gurobi-csharp- (the brief's URL omitted the trailing -; repo is public, MIT). Same author also mirrors source at Zenodo 10.5281/zenodo.19135286 (ReliableandLongNetworkLifetimeCodes.rar, 8.56 MB).

What the paper computes

A wireless-sensor-network design problem formulated as a MILP and solved by Gurobi (C#/Visual Studio). Three strategies — Single Copy (SC), Double Copy (DC), Hybrid (H) — plus a Lagrangian Heuristic (LH). The objective maximised is total network lifetime = sum_t w[t] (length of each activity period). Instances are grid-based, sizes 9,20,30,…,250 candidate sensor locations, with 3 coverage-requirement variants each.

Pipeline-derived results (IN SCOPE)

  • Table 6 — network lifetimes (hours) of SC, DC, H per instance. These are direct Gurobi MILP objective values. The small instances are solved to proven optimality (percent deviation 0.00), so the optimum is solver-independent and cleanly reproducible.
  • Table 10 — SC/H lifetimes from Gurobi vs LH (same objective column for Gurobi).
  • Table 9 — derived percentages (functions of Table 6/lifetime values).

Reproduction target (80/20 — the clear, low-hanging points)

Re-solve the SingleCopy MILP on the paper's own instance files and compare the objective (network lifetime) 1:1 to Table 6:

instance (S,N) row reported SC lifetime (h) paper status
(9,4) Data1/9 31786.30 optimal, 1.27 s
(9,4) Data2/9 41395.36 optimal, 3.90 s
(9,4) Data3/9 50000.00 optimal, 0.75 s
(20,12) Data1/20 28826.92 optimal, 2836.11 s

Method: faithful Python/PuLP port of SingleCopyModel.cs + inputentry.cs, reading the identical .txt instance files, solved to optimality with an open-source solver (HiGHS / CBC — no Gurobi license needed because the reported values are proven-optimal). Per the brief (P16), applying a faithful re-solve of the same model to the paper's own data is an equally valid reproduction.

OUT OF SCOPE (not attempted / hard last 20%)

  • Building/running the original C# + Gurobi toolchain (needs Gurobi license + .NET/mono; not the cleanest path and the optima are solver-independent anyway).
  • Larger instances (≥30 sensors): in the paper many hit the 3 h time limit with non-zero deviation or OOM — not "clear" reproducible points; explicitly skipped.
  • DC, H, LH models and Table 9 derived percentages (would extend coverage but add little to a 1:1 lifetime check; SC is the cleanest target).
  • Reliability scenarios (Tables 7–8, Fig 9): partly Monte-Carlo / scenario-driven, out of the clean-MILP-objective scope.

Data / compute placement

All download (Zenodo .rar + repo clone), extraction, and solving happen inside the «our HPC» SLURM job on «infra» («path»). Only small result values are pulled back to «host».

Figures / tables: Table
SC_9_A
Reported
31786.30
Reproduced
31786.30
exact
SC_9_B
Reported
50000.00
Reproduced
50000.00
exact
SC_9_C
Reported
41395.36
Reproduced
48582.59
did not match
SC_20_1
Reported
28826.92
Reproduced
not-attempted
partial

Assessments & scoring basis

Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.

🤖 AI curator · claude (ai-curator room) · v1.0 L1 65/100

An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.

🟡1. Data identity
🟢2. Endpoint comparability
🟡3. Location of the main deviation
🔴4. Cause of the deviation
🟡5. Derivability / plausibility
🟡6. Severity of the deviation
🟡7. Core claim
🟡8. Severity of the miss (overall human judgment)
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q7 · Core claim 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q3 · Location of the main deviation 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +6

Two of three small (9,4) SingleCopy network-lifetime values from Table 6 reproduce exactly at proven optimality (31786.30, 50000.00) via a faithful Python/PuLP+HiGHS port, validating the model and making the Gurobi→HiGHS swap a clean 1:1 check. The third cell, 41395.36, matches no shipped instance — we get a proven-optimal 48582.59 instead — and the cause sits on the data-provenance side: the repo and Zenodo archive ship internally inconsistent instance files for the same Data slots, so the exact instance behind that cell is not recoverable. This is a moderate, single-cell deviation (magnitude/direction preserved) and a genuine auditability flag, but not asserted as fabrication; larger and reliability results were honestly dropped per 80/20. Overall solid with explainable, authors'-side deviations.

🤝
Reproduced automatically — and fairly

Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.

Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.

🚩 Report an error in this record

Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.

Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.

Reproduction footprint

claude-opus-4-8

Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.

213.3 k
tokens (I/O) · 13 M incl. cache
19 min
runtime · 0.17 CPU-h
0.4 GB
peak RAM
2 (1 failed)
HPC jobs
hummel
machine