Computers & Structures Submission Guide
A source-checked guide for turning a computational-structures claim into an auditable submission package.
Readiness scan
Find out if this manuscript is ready to submit.
Run the Free Readiness Scan before you submit. Catch the issues editors reject on first read.
Quick answer: Submit to Computers & Structures, an Elsevier journal, when the computational method, structural or mechanical problem, verification evidence, and practical consequence form one reusable contribution. A simulation applied to one structure is not automatically a computational-methods paper: the reader needs to see what is new, how it was checked, and what transfers beyond the case.
Evidence basis: We reviewed the official Computers & Structures journal page, aims and scope, and Guide for authors on August 31, 2026. The live Elsevier guide controls the upload. The decision artifact below is Manusights editorial judgment, not an acceptance forecast.
Check whether the method and evidence chain are ready for review.
Decide what the reusable contribution is
Contribution shape | Evidence that makes it usable | Hold signal |
|---|---|---|
New numerical formulation or solver | Governing assumptions, implementation, convergence or error evidence, and benchmark comparison are visible | A renamed implementation is offered without a discriminating test |
Structural-analysis method | The method changes a structural inference, design decision, or prediction and is checked against a relevant reference | Software output is treated as validation by itself |
Model-reduction or computational framework | Inputs, training or calibration, baselines, out-of-sample test, and failure boundary are explicit | Accuracy is reported without a fair comparator or applicability limit |
Application-led computational study | The case reveals a transferable mechanism, method, or validation lesson | The only novelty is a new geometry, location, or software workflow |
Worked example: a finite-element method claim
Suppose a manuscript proposes a formulation for nonlinear structural response. The submission-ready version names the governing problem, documents discretization and solver choices, shows a convergence or error check, compares against a credible benchmark, and states where the formulation stops being reliable. The contribution is the verified inference or method, not a contour plot alone.
Common failure patterns to repair
Application substitutes for method novelty. The manuscript shows a detailed model of a bridge, material, or component but never states the transferable computational advance. Put the reusable method claim and the test that distinguishes it near the start.
Verification and validation blur together. Numerical convergence, code checks, benchmark agreement, and agreement with physical observation answer different questions. Label what each check establishes and do not let one stand in for the others.
The boundary of use is invisible. A model is presented as generally accurate while its calibration range, loading regime, material behavior, mesh sensitivity, or computational cost is omitted. Place the limitation beside the result that needs it.
Readiness check
Run the scan against the requirements while they're in front of you.
See score, top issues, and journal-fit signals before you submit.
Assemble the current submission package
The live Guide for authors governs article type, manuscript preparation, figures, declarations, supplements, and publishing policy. It does not provide a fixed main-text word cap or a fixed initial figure cap that should be copied into this guide; confirm those details for the chosen article type at upload.
Final package checklist
- Manuscript and title page matched to the current article type.
- A concise cover letter explaining the computational-structures contribution and why its evidence is reusable.
- Figures, tables, and legends that identify model setup, comparison, uncertainty, and units.
- Data availability statement and a route to code, inputs, meshes, or materials when the claim depends on computation.
- Conflicts of interest, funding statement, author contributions, permissions, and ethics documentation where relevant.
- Supplementary methods, algorithms, parameters, convergence studies, benchmark data, or validation material needed to audit the central result.
Make the verification record match the claim
Verification is not a single checkbox. Before submission, separate the question about whether the numerical implementation solves the stated mathematical problem from the question about whether the model represents the structural system well enough for the intended inference. The first may use mesh refinement, time-step refinement, residual checks, manufactured solutions, or a trusted analytical result. The second may use an experiment, a documented benchmark, an observed response, or an independently established physical constraint. A paper can be careful about one and still be weak on the other.
Claim in the paper | Record a reviewer should be able to find | Boundary to state plainly |
|---|---|---|
A solver is accurate for the stated problem | Discretization, tolerances, convergence behavior, and a reference solution or benchmark | Accuracy may not transfer to another geometry, load case, material model, or time scale |
A model captures a structural response | Model assumptions, parameter origin, calibration evidence, comparison, and residual disagreement | Agreement in one regime does not validate another failure mode or operating condition |
A new method is better than a baseline | Like-for-like inputs, computational budget, error or performance measure, and baseline configuration | A speed or accuracy gain can disappear when the problem class, hardware, or tolerance changes |
A simulation changes an engineering decision | The structural quantity affected, the decision context, uncertainty, and the condition under which the conclusion holds | A numerical trend alone may not establish safety, reliability, or general design guidance |
A useful reviewer path
Pick the central figure or table and write the path a skeptical reader would take: claim, governing assumptions, inputs, implementation, verification check, validation or physical comparison, uncertainty, and limit. Then make sure each step has a visible home in the manuscript or supplement. If a source file is necessary to reproduce a key result, identify what it contains and how it maps to the figure or table. A large archive with no route from claim to file does not make the evidence easier to inspect.
For example, a paper that predicts buckling behavior should identify the constitutive assumption, boundary conditions, imperfection treatment, mesh or discretization check, comparison point, and the range in which the result is being interpreted. A paper that introduces an optimization workflow should make its objective, constraints, baseline, stopping condition, and sensitivity to design choices visible. These are not extra journal requirements; they are the information a reader needs to distinguish a reusable computational advance from a single favorable run.
Decide whether the limitation belongs in the main text
Put a limitation in the main text when it changes how a reader interprets the headline result: for example, a parameter regime, an excluded mechanism, a calibration dependency, a data gap, or a comparison that is not like for like. Reserve the supplement for the detailed derivation, full parameter listing, additional convergence results, and extended validation material that supports the main claim. This division keeps the central evidence legible without pretending that a condensed figure contains the complete record.
Before upload, ask one final question: if the code, mesh, input data, or baseline were unavailable for a week, would another researcher still know exactly what evidence they need to inspect? If not, add a route from the claim to the relevant method or supporting material. That is a stronger submission test than adding a generic computational keyword.
What happens after submission
Computers & Structures does not promise a case-specific review timetable in the sources reviewed. Use this explicitly caveated stage map to prepare the paper, not to predict a decision:
- Submission and technical review: the record, files, and declarations are checked against current requirements.
- Editorial fit review: the journal evaluates whether the computational or structural-analysis contribution is central and sufficiently evidenced.
- Peer review: reviewers inspect assumptions, implementation, verification, validation, interpretation, and presentation.
- Decision or revision: editor correspondence states the action for the individual manuscript.
Review validation, uncertainty, and method-claim boundaries.
Compare the contribution owner before submission
Decision question | Computers & Structures | Computer Methods in Applied Mechanics and Engineering | Engineering Structures |
|---|---|---|---|
Main owner | Computational methods connected to structural analysis, mechanics, or design | Broad computational mechanics method contribution | Structural-engineering contribution and evidence |
Evidence test | A method or model improves a structural or mechanical inference | The numerical method itself has broad mechanics significance | The structural implication remains central beyond the computation |
Hold signal | A standard model is applied without a computational advance | The method is narrow to one structural application | A computational method lacks structural-engineering consequence |
This is a routing boundary, not an acceptance forecast. Verify every journal's live scope before choosing a target.
Submit if
- The computational contribution is distinct from the case study that demonstrates it.
- Verification, validation, and uncertainty evidence match the claim's strength.
- Another researcher could inspect the inputs, parameters, baseline, and result needed to assess the central method.
- The current Guide for authors has been checked immediately before upload.
Think twice if
- The manuscript relies on a commercial or familiar model without demonstrating a transferable advance.
- A calibration result is being generalized beyond the measured regime.
- Code, data, mesh, parameters, or benchmark details prevent the central computation from being checked.
- The reusable contribution belongs more clearly to a broad numerical-methods or application-specific venue.
Frequently asked questions
Use the live official scope to confirm fit. A strong submission makes the computational or structural-analysis contribution, its verification, and its engineering consequence legible together.
Use the current Submit your article and Guide for authors links on the official Elsevier journal page, then verify the current article-type requirements at upload.
Check the live guide and prepare the manuscript, concise cover letter, figures, data and code materials where relevant, declarations, and supplements appropriate to the article type.
This guide does not promise a case-specific review timeline. Use the journal record and editor correspondence for an individual manuscript.
Do not infer private editorial reasoning. Repair visible novelty, verification, validation, reproducibility, and scope gaps before upload.
Check the live official journal page and Guide for authors for current publishing options or charges instead of relying on a static fee summary.
Sources
- 1. Computers & Structures, Elsevier.
- 2. Aims and scope, Elsevier.
- 3. Guide for authors, Elsevier.
Before you upload
Choose the next useful decision step first.
Move from this article into the next decision-support step. The scan works best once the journal and submission plan are clearer.
Use the scan once the manuscript and target journal are concrete enough to evaluate.
Anthropic Privacy Partner. Your manuscript is never used to train any model.