ROLE You are a bibliometric and engineering-literature analy...

Prompt

ROLE You are a bibliometric and engineering-literature analyst. Your job is to find the real acceptance pattern of Q1 journals for ANALYTICAL and CODE-ORIENTED papers on steel member stability, by measuring what each paper adds over its own immediate prior work, both analytically and numerically. CONTEXT I am a PhD student working on welded slender (Class 4) I-section members under N+M and N+M+T (axial force, bending, torsion/bimoment), EN 1993-1-1:2022 Annex C.1/C.2, interaction factors k_yy/k_zy and equivalent moment factors C_m/C_mLT, and local-global buckling of mono-symmetric I-sections. I need to know what kind of increment over prior work is enough to be published in Q1 journals (Thin-Walled Structures, Engineering Structures, Journal of Constructional Steel Research, also Structures, Steel Construction, J. Struct. Eng. ASCE, Computers & Structures, Int. J. Mech. Sci.), so that I can decide which of my planned papers to write and where to send them. Earlier attempts failed because they used only abstracts, only the papers that were easy to find, and no measurement of the increment. SAMPLE (build it first, report how you built it) 1. Define 6 topic clusters: A. closed-form / approximate formulas for elastic critical loads (LTB, flexural-torsional, with axial force, mono-symmetric, warping restraint) B. equivalent moment factors C_m, C_mLT, moment-gradient factors C. interaction factors k_yy, k_zy, Method 1 / Method 2 / Annex C.1 recalibration D. bimoment / warping torsion in design (N+M+T, Annex C.2, EN 1993-6 Annex A) E. local-global interaction (analytical, variational, effective section, O.I.C., effective width under non-uniform stress) F. imperfection / residual-stress calibration for GMNIA 2. For each cluster find 12-15 papers 2018-2026 in the journals above. Use Scopus/WoS/Google Scholar/publisher sites. Do NOT pick only highly cited papers. Take at least a third from the last 24 months and at least a third with fewer than 10 citations, so success is not confused with fame. Record how each paper was found. 3. For each included paper also find its "closest prior work" (the paper the authors themselves say they extend, or the closest you can identify) and read both. READING RULE Read the full text (introduction, method, validation, conclusions, tables). Never rely on the abstract. If you cannot access the full text, mark the paper "abstract only" and exclude it from the statistics. EXTRACT, PER PAPER (one row each) - Citation, journal, year, quartile (state source and year of the quartile) - Contribution type: closed-form / FE-campaign recalibration / new formulation / experimental / code background / review - Claimed novelty, quoted in the authors' words - Closest prior work and the DELTA, in two parts: ANALYTICAL DELTA: what changed in the theory or formula? (new term, new boundary condition, new section type, new load case, removed assumption, new derivation route, same formula with new coefficients, or nothing new analytically) NUMERICAL DELTA: what changed in the numbers? Give the accuracy of the new method and of the prior method on the SAME data set if the paper reports it (mean ratio, CoV, max/min, % unsafe, % improvement, error vs exact or FE). If the authors do not report a comparison with the prior work, say so. - Size of validation: number of tests, number of FE models, parameters varied, reference used (tests / shell FE / GBT / exact solution / code) - Whether a reliability analysis (EN 1990 Annex D) is present - Whether the result entered or was proposed for a code - Does it end with a design formula or procedure? (yes/no) - Limitations the authors state; limitations you see (validity range, single span, fork ends, one steel grade, Class 3 only, etc.) - Reviewer-risk notes: what a hostile reviewer could attack ANALYSIS (after the table) 1. Classify the delta of every paper on this ladder and count each level per journal and per cluster: L0 new parameter range only L1 corrected / recalibrated rule L2 new rule with stated validity limits L3 new mechanism or derivation explaining why L4 predictor validated on unseen cases 2. For the numerical delta, give the distribution: how small can the gain over the prior method be (e.g. 3% in mean, 0.01 in CoV) and still be published? Report the minimum, median and 10th percentile of improvement by journal. Say when the paper has no numerical gain at all and what carried it instead (e.g. generality, code adoption, new capability). 3. For analytical-only or partly analytical papers, state what they were compared against (exact solution, FE, GBT, tests, code) and with what accuracy, and whether a design consequence was shown. 4. Identify papers with a "small step" over prior work that were still published in Q1, and papers that look similar but were published in lower tiers. State the features that differ (validation size, code link, reliability, generality). 5. Check my three planned papers against the pattern and say, for each, which level of the ladder it can honestly reach and what minimum evidence the pattern requires: P1: second-order bimoment amplification with axial force inside Annex C.2 (closed form using N_cr,T; GMNIA on welded Class 4 I-sections, 9 cases) P2: closed-form C_m / C_mLT for mono-symmetric beam-columns with psi<0 (benchmark: Method 1, Annex C.1, Chen & Shen 2015, J. Struct. Eng. 2020 Direct Determination Method) P3: elastic branch-exchange (veering) between local and global modes in mono-symmetric welded I-section beam-columns HONESTY RULES - Published papers show only survivors. State this and say what you did to reduce it (e.g. look for preprints, rejected-then-published versions, reviewer reports in open peer review journals, and editorial statements from the journals' author guidelines). - Separate facts you read in the paper from your inferences. Mark inferences. - Do not use percentages of acceptance or tier rules from blogs or AI summaries; use only what you can cite. - If a claim of mine above is wrong or too narrow, say so. OUTPUT 1. Sampling log (how the sample was built, counts per cluster/journal). 2. The per-paper table as CSV (all fields above) plus a markdown version. 3. A one-page pattern summary with the minimum viable increment per journal. 4. A verdict for P1, P2, P3 with the evidence required and the main risk. 5. A list of papers you could not read in full text.