Geng Fraud Report: "Complex Cold Chain Logistics Model and Solution Algorithm under the Dual-Carbon Background" by Chen Yudie et al.

Academic fraud report · Geng Detector

Summary

Verdict: Highly suspicious paper, rated orange-level, published in Systems Engineering — Theory & Practice (DOI: 10.12011/SETP2024-2027). Six anomalies were identified from the text alone. (1) References [17] and [23] are exact duplicates of the same Bektaş & Laporte 2011 paper, suggesting careless merging or padding. (2) Table 12 in Section 4.8 carries the copied header from Table 11 ("Comparison of separate and simultaneous pickup-delivery") although its columns describe single- vs. multi-compartment vehicles — clear template-reuse residue. (3) The MPGA-VNS algorithm reportedly achieves 0.00% gap to best-known solutions on 28 classic CVRP instances including A-n80-k10, performance unmatched even by top solvers such as LKH-3 or HGS. (4) Cost coefficients c1, c2, c3 are redefined inconsistently between Table 1 (carbon price, fixed cost, fuel price) and Section 4.1.1 (product unit prices in CNY/t). (5) Damage cost in Eq. (10) uses departure load Qp^i instead of arrival load, contradicting physical reality. (6) Heavy overlap with the authors' prior 2023 paper indicates minimal incremental contribution. Confidence: text-based only; no pixel-level image forensics possible.

Verdict

Highly suspicious (🟠). Multiple strong text-based red flags were detected, but no pixel-level image analysis was possible because the original PDF images were not available for inspection. Final determination requires the authors' raw data and a response from the editorial office.

Key findings

Duplicate reference [17] and [23] — both cite Bektaş & Laporte (2011), Transportation Research Part B 45(8): 1232–1250.
Table 12 header is copied verbatim from Table 11 ("独立送取货和同时送取货对比表") while its columns actually compare multi-compartment vs. single-compartment vehicles, indicating template splicing.
The MPGA-VNS algorithm is reported to hit 0.00% gap on 28 CVRP benchmarks, including A-n32-k5 through A-n61-k9 and A-n80-k10, an implausibly perfect result.
Cost parameters c1, c2, c3 are defined as carbon price, fixed cost, and fuel price in Table 1 but are used as unit product prices (4000, 2000, 3000 CNY/t) in Section 4.1.1.
Damage-cost formula (10) applies the departure load Qp^i instead of the arrival load, contradicting the physical process described.
The paper overlaps heavily with the authors' 2023 paper "'双碳'背景下联合配送冷链物流模型及其求解算法" in 控制与决策, with the same MPGA-VNS algorithm and similar structure.

Evidence highlights

Reference duplication: [17] = [23] = Bektaş T, Laporte G. Transportation Research Part B 45(8): 1232–1250, 2011.
Table 12 (p. 20) column labels: 多隔间车辆, 单隔区车辆 1, 单隔区车辆 2; caption still reads "独立送取货和同时送取货对比表".
Table 5 (p. 13): Best-known gap reported as 0.00% across the entire A-n32-k5 to A-n80-k10 benchmark set.
Symbol conflict: c1 = 4000 元/t vs. Table 1 definition of c1 as "单位碳价格 (元/kg)".
Self-citation [18]: Chen Yudie, Gan Hongcheng, Cheng Liang. 控制与决策, 2023 — same algorithm family, near-identical model.

Notes

DOI: 10.12011/SETP2024-2027.
Limitations: analysis is text-based only; image-level forensics (e.g., duplicated or spliced figures) were not feasible.
Suggested follow-ups already initiated in the original report include contacting the authors for raw CVRP logs and submitting inquiries via PubPeer and to the journal editorial office.