Preparing Your Manuscript

This guide explains how to prepare each part of your manuscript so that it can be reviewed and published efficiently. It follows practice recommended by leading open-access publishers and by the reporting, reproducibility and integrity standards recognised across engineering and the computing sciences. Following it closely improves the clarity of your work, speeds up review and reduces the chance of your manuscript being returned before assessment.

Before you start

Confirm that your manuscript fits the journal’s Aims & Scope; that it is original and neither published nor under consideration elsewhere; that all authors have approved it; and that you can provide the declarations described in Ethics and Disclosures. Prepare your manuscript with the TECS template and the required citation style. Assemble your files as a single editable manuscript (text, references, tables and figure captions), with figures supplied either in the manuscript for review or as separate high-resolution files, plus any supplementary material, code or data.

Language and writing style

Write in clear, correct and concise English, using either British or American spelling consistently throughout. Prefer plain, direct sentences; use the active voice where it aids clarity; and keep paragraphs focused on a single idea. Define every abbreviation, symbol and acronym at first use and use it consistently thereafter. Avoid unnecessary jargon, and use accurate, respectful and inclusive language. If English is not your first language, consider professional language editing before submission; language quality does not affect the editorial decision, which is based on the technical merit of the work.

File format and general layout

Manuscript structure at a glance

Assemble the manuscript in the following order. Each element is described in detail below.

Title page

The title page should contain:

Because review is single-blind, you do not need to prepare an anonymised manuscript; author details remain in the manuscript.

Abstract

Provide a self-contained abstract of about 150–250 words that can be read and understood on its own. It should briefly state the problem and motivation, the approach or contribution, the method of evaluation, the principal results (with key quantitative findings where possible) and the main conclusion and its significance. Do not include citations, undefined abbreviations, mathematical notation, figures or tables.

Keywords

Supply 4–6 keywords that describe the content for indexing and discovery. Choose established terms from your field (for example recognised computing or engineering vocabularies and classification schemes), avoid simply repeating words already in the title, and separate the keywords with commas.

Main text

Research articles in engineering and computing commonly follow a structure such as Introduction, Related Work, the proposed method or system, Experiments and Evaluation, Results and Discussion, and Conclusions. Use numbered headings (for example 1, 1.1, 1.1.1) to a maximum of three levels. Reviews, surveys and other article types may adapt this structure to suit their content.

Introduction

Set out the problem and its importance, the limitations of existing approaches, and the specific contribution of the paper. State the research questions or objectives clearly, and summarise the contributions, ideally as a short explicit list.

Related work

Position the work with respect to the state of the art, citing the relevant literature fairly and identifying the gap the paper addresses. Make clear what is genuinely new relative to prior work, including any earlier conference version of the paper (see Online First and Preprints).

Proposed method or system

Describe the method, model, algorithm, architecture or system in enough detail for a competent reader to understand and, where relevant, re-implement it. Define the problem formally where appropriate, state assumptions, and present the design clearly using algorithms, diagrams and analysis. Where you make theoretical claims, state the assumptions and provide proofs or rigorous arguments.

Experiments and evaluation

Describe the experimental design fully: the datasets or benchmarks (with sources, versions and splits), the baselines compared, the evaluation metrics and why they are appropriate, the hyperparameters and configurations, the hardware and software environment, and the number of runs. Report enough detail for the experiments to be reproduced, and follow the Data, Code and Reproducibility guidance in Ethics and Disclosures.

Results and discussion

Present the results objectively, using text supported by tables and figures without duplicating the same data in both. Compare against baselines, report measures of variability and, where appropriate, statistical significance, and include ablation or sensitivity analyses that support the claims. Interpret the results, explain why the approach behaves as observed, and discuss limitations and threats to validity honestly.

Conclusions

Provide a concise conclusion that captures the main contribution and findings and, where appropriate, their practical implications and directions for future work.

Algorithms, notation and mathematics

Reproducibility, data and code

Reproducibility is central to engineering and computing research. Report the datasets, code, models, parameters and computational environment needed to reproduce your results, and, wherever possible, make the data and code openly available in a citable repository with a persistent identifier. Where data or code cannot be shared, explain why. Full requirements, including recommended repositories and the content of the availability statement, are given in Ethics and Disclosures.

Units and measurements

Use SI units throughout, giving other units in parentheses where helpful. Report timing, memory, throughput and other performance measures with the conditions under which they were obtained.

Reporting standards

Where an established reporting guideline applies to your design, follow it and cite it — for example PRISMA for systematic reviews and meta-analyses. For empirical machine-learning and systems research, report datasets, splits, hyperparameters, compute and evaluation protocols in enough detail to support reproduction.

Tables

Figures and image integrity

Supplementary material, code and data

You may provide supplementary files that support but are not essential to understanding the main article — for example proofs, additional experiments, datasets, source code, configuration files or video. Cite each item in the text (for example “Supplementary Algorithm S1”), give each a short title and caption, and supply it in a widely supported format. Source code and data should preferably be deposited in a citable repository, as described in Ethics and Disclosures, rather than only as supplementary files.

References and citations

Declarations and back matter

Place the following declarations after the main text and before the references, stating “None” or “Not applicable” where appropriate but omitting none. Full guidance is given on the Ethics and Disclosures page.

Final checks before submission

Before you submit, run through the Submission Checklist to confirm that your manuscript is complete, correctly formatted and accompanied by all declarations, and that every table, figure, algorithm and reference is cited and consistent.