ijoc-workflow

name	ijoc-workflow
description	Use when deciding which ijoc-* sub-skill to invoke next, or when sequencing a manuscript from idea through rebuttal for an INFORMS Journal on Computing (IJOC) submission. Routes — it does not replace — the specialized skills.

IJOC Workflow Router (ijoc-workflow)

Overview

This is the router. It tells you which ijoc- skill to use at the current stage* of a manuscript aimed at the INFORMS Journal on Computing (IJOC) — the INFORMS journal at the intersection of operations research / management science and computer science. IJOC judges a paper on its computational and/or methodological advance: algorithm design and analysis, exact and heuristic optimization, integer/stochastic/robust programming, simulation, computational probability, machine learning for OR, data-science methodology, and the software/computational tooling that makes OR work. A clever application alone does not clear the bar; the computing must be the contribution.

Operational tells that you are at IJOC and not a sibling: the author selects one of the journal's 10 technical areas at submission via ScholarOne (mc.manuscriptcentral.com/ijoc); the Area Editor desk-rejects roughly 40% of submissions and the associate editor's identity is never disclosed to authors; review is single-blind (reviewers see author names); and IJOC runs a distinctive reproducibility regime — accepted papers deposit code/data in the IJOC Software and Data Repository on GitHub (the INFORMSJoC organization), which snapshots a tagged release and mints a separate code DOI (…​.cd) alongside the article DOI. Editor-in-chief: David L. Woodruff (UC Davis), succeeding Alice E. Smith (whose term ended 2025-12-31); confirmed via INFORMS (检索于 2026-06-22；以官网为准). Re-verify volatile specifics on the official INFORMS pages.

When to trigger

• The user asks "what should I do next?" on an IJOC-bound paper
• A draft's binding constraint (scope fit, algorithmic claim, experiments, exhibits, deposit) needs naming
• Work is ping-ponging between method design, computational experiments, framing, and the response letter
• An IJOC decision letter arrived and the user must switch into revision mode

Routing table

Current symptom	Next skill
Unsure the paper is computing-first or which of the 10 areas fits	ijoc-topic-selection
The algorithm/model is described but its formulation, correctness, or guarantees are loose	ijoc-theory-development
Contribution vs. OR/MS and CS prior art is fuzzy or oversold	ijoc-literature-positioning
Method choice, baselines, or experimental protocol are not pinned down	ijoc-methods
Computational experiments need instances, fair tuning, timing, and statistics	ijoc-data-analysis
The "what is new computationally" sentence is not sharp	ijoc-contribution-framing
Performance profiles / runtime-vs-size plots are dense or misleading	ijoc-tables-figures
Prose buries the method; abstract/intro do not land the advance	ijoc-writing-style
Page limit, LaTeX template, area choice, GitHub deposit prep	ijoc-submission
Want to understand AE/AE-blind timeline, desk-reject odds, or sibling transfer	ijoc-review-process
Received an R&R; need a response-letter and re-experiment plan	ijoc-rebuttal

Default order

1. ijoc-topic-selection — confirm computing-first scope and the right area
2. ijoc-theory-development — formulate the algorithm/model; state correctness/complexity
3. ijoc-literature-positioning — stake the advance vs. OR/MS and CS frontiers
4. ijoc-methods — method choice, baselines, experimental design
5. ijoc-data-analysis — instances, tuning, timing, statistical comparison, scaling
6. ijoc-contribution-framing — sharpen the one-sentence computational claim
7. ijoc-tables-figures — performance profiles and scaling plots
8. ijoc-writing-style — make the advance land (abstract + intro last)
9. ijoc-submission — page limit, template, area, GitHub deposit preflight
10. ijoc-review-process — calibrate the AE-blind cycle; consider sibling fit
11. ijoc-rebuttal — after the R&R

ijoc-writing-style is a late polish; do not rewrite the intro before the algorithm and experiments settle.

Anti-patterns

• Treating IJOC like Operations Research (broad OR theory) or Management Science (broad MS) — IJOC wants the computing/methodology to be the news
• Confusing it with Mathematical Programming Computation (pure computational MP) or INFORMS Journal on Optimization (optimization theory) — IJOC is the OR↔computing interface across simulation, ML, probability, and tooling, not optimization alone
• Polishing exhibits or prose while the algorithm or its experiments are still moving
• Deferring the GitHub code/data deposit to "after acceptance" — design reproducibility in from the start
• Submitting to the wrong technical area and inviting a fast Area-Editor desk reject

Routing by paper archetype

IJOC spans distinct computing archetypes; the binding constraint differs by archetype. Read the archetype, then enter the chain at the right link.

Archetype	Likely first bottleneck	Enter at
exact algorithm / new formulation (B&C, decomposition)	correctness + complexity + benchmark instances	ijoc-theory-development → ijoc-data-analysis
(meta)heuristic / matheuristic	fair tuning, baselines, statistical wins not luck	ijoc-methods → ijoc-data-analysis
ML-for-OR / learning-to-optimize	generalization claim + OR baseline + reproducibility	ijoc-methods → ijoc-data-analysis
simulation / computational probability	DGP, seeds, variance reduction, validation	ijoc-methods → ijoc-data-analysis
software / computational tooling	what is the methodological news beyond the tool	ijoc-contribution-framing → ijoc-submission

Worked routing example (illustrative)

A user says: "My branch-and-cut solver beats the commercial baseline on my instances, but a referee says the win could be tuning and the instance set is cherry-picked." That is two IJOC pushbacks — unfair experimental comparison and non-representative benchmark — both owned by ijoc-data-analysis, with the formulation/cut-validity defense in ijoc-theory-development. Route to ijoc-data-analysis first: re-run on standard library instances (e.g., MIPLIB-class), tune both solvers on a disjoint set, report a performance profile and a Wilcoxon test, and disclose hardware and time limits. Only once the win survives do you return to ijoc-tables-figures and ijoc-rebuttal.

Minimal decision snippet

if decision_letter_arrived:        -> ijoc-rebuttal
elif ready_to_submit:              -> ijoc-submission
elif exhibits_or_profiles:         -> ijoc-tables-figures
elif experiments_or_benchmarks:    -> ijoc-data-analysis
elif method_or_baselines:          -> ijoc-methods
elif formulation_or_guarantees:    -> ijoc-theory-development
elif claim_or_positioning_fuzzy:   -> ijoc-contribution-framing / ijoc-literature-positioning
else:                              -> ijoc-topic-selection