Method Fit (pom-methods)
When to trigger
- The method may not match the operations question (level, uncertainty, causality)
- You must choose between an analytical model, an empirical design, an experiment, or a data pipeline
- A reviewer says "the method does not support the claim" or "this is a methods paper, not OM"
POM mandates no single method — but the method must fit and be rigorous
POM explicitly places no restriction on research methods, while being historically anchored in analytical modeling. Pick the method the question demands, and route to the matching Department.
| Operations question / claim |
Method |
| Optimal policy under cost/service objective |
Optimization (LP/MIP/convex, dynamic programming); characterize the policy |
| Decisions under demand/lead-time uncertainty |
Stochastic modeling, queueing, inventory theory, MDP/ADP |
| Strategic interaction (suppliers, competitors, platforms) |
Game theory (Nash/Stackelberg); prove equilibrium existence/uniqueness |
| Causal operational effect from field data |
Empirical OM: DiD, IV, RD, matching with a clear identification strategy |
| Human operational decision bias |
Behavioral/experimental OM (lab/online); randomization, manipulation checks |
| Systems too complex for closed form |
Discrete-event simulation; validation, warm-up, replications, CIs |
| Prediction feeding an operational decision |
Operations data science (ML / forecasting), tied to a decision/loss |
What each track must defend
- Analytical: assumptions grounded in operations reality; solution concept; structural results; sensitivity/comparative statics; managerial interpretation. Robustness = relaxing key assumptions, not just numerical examples.
- Empirical OM: identification (what makes the effect causal), measurement in decision-relevant units, unit of analysis, external validity to other operations settings.
- Behavioral OM: design, incentives, randomization, manipulation and attention checks, and operational realism (does the lab task map to a real operations decision?).
- Operations data science: validation design, leakage checks, and — critically — the operational value: does a better prediction change a feasible policy (predict-then-optimize)?
- Simulation: parameter provenance, validation against known cases, sensitivity, and a replication package.
The POM bar on method
A method exists to serve an OM contribution judged interesting to practicing managers. If the paper's value is mainly a technical advance with thin OM decision content, a methods journal may fit better. Keep heavy derivations, solver details, and extra robustness for the unlimited online e-companion so the 32-page main document stays focused.
Execution bridge (StatsPAI / Stata MCP)
For the empirical / causal lane, estimate and audit rather than only specify. Full
map: execution-with-mcp. POM spans analytical and empirical OM; apply the chain below to its empirical-OM papers, and note when a contribution is analytical / optimization.
detect_design → recommend → fit with as_handle=true → audit_result to
enumerate the checks the design owes.- Panel / staggered DiD:
callaway_santanna / sun_abraham + bacon_decomposition
honest_did_from_result. IV: effective_f_test + anderson_rubin_ci. RDD:
rdrobust + mccrary_test.
- Experiments: randomization-based inference and
romano_wolf for the many-outcome
family-wise correction reviewers expect.
Match the toolchain to the reviewer pool, and report the effect size the venue
wants. A run end-to-end (synthetic data, real returns) is in the
JF execution walkthrough.
Checklist
Methods pass for Production and Operations Management
Use this as a second-pass capability check. First lock the operational decision, the performance metric, and the implementable lever; then test whether the manuscript addresses POM reviewers who want operational insight tied to production, service, supply-chain, or platform decisions.
- Primary move: Name assumptions, diagnostics, robustness, falsification, and failure modes; do not accept a method section that hides the decisive validity threat.
- Decision ledger: return
claim / evidence / blocker / next edit rows so the next pass can patch the manuscript directly.
- Neighbor test: compare against Management Science for broader OR/MS theory, Operations Research for method-first optimization, MSOM for manufacturing/service operations depth; if the neighboring outlet has the stronger audience claim, recommend re-routing before polishing.
- Verification floor: before submission-ready advice, re-open
resources/official-source-map.md for volatile rules and name the one unresolved fact that could change the recommendation.
Output format
【Method family】optimization / stochastic / game-theory / empirical / behavioral / simulation / data-science
【Operations question】<decision problem>
【Validity risks】assumptions / identification / measurement / leakage / validation
【Practice tie】how the method yields a manager-usable result
【e-companion plan】proofs / extra analyses to move online
【Next step】pom-data-analysis
1---2name: pom-methods3description: Use when selecting or auditing the method for a Production and Operations Management (POM) manuscript — analytical modeling (optimization, stochastic models, game theory), empirical identification, behavioral experiments, simulation, or operations data science. Matches method to the operations question; it does not execute the analysis (pom-data-analysis).4---56# Method Fit (pom-methods)78## When to trigger910- The method may not match the operations question (level, uncertainty, causality)11- You must choose between an analytical model, an empirical design, an experiment, or a data pipeline12- A reviewer says "the method does not support the claim" or "this is a methods paper, not OM"1314## POM mandates no single method — but the method must fit and be rigorous1516POM explicitly places **no restriction on research methods**, while being historically anchored in analytical modeling. Pick the method the *question* demands, and route to the matching Department.1718| Operations question / claim | Method |19|--------------------------------------------------------------|--------------------------------------------------------------------|20| Optimal policy under cost/service objective | Optimization (LP/MIP/convex, dynamic programming); characterize the policy |21| Decisions under demand/lead-time uncertainty | Stochastic modeling, queueing, inventory theory, MDP/ADP |22| Strategic interaction (suppliers, competitors, platforms) | Game theory (Nash/Stackelberg); prove equilibrium existence/uniqueness |23| Causal operational effect from field data | Empirical OM: DiD, IV, RD, matching with a clear identification strategy |24| Human operational decision bias | Behavioral/experimental OM (lab/online); randomization, manipulation checks |25| Systems too complex for closed form | Discrete-event simulation; validation, warm-up, replications, CIs |26| Prediction feeding an operational decision | Operations data science (ML / forecasting), tied to a decision/loss |2728## What each track must defend2930- **Analytical:** assumptions grounded in operations reality; solution concept; structural results; sensitivity/comparative statics; managerial interpretation. Robustness = relaxing key assumptions, not just numerical examples.31- **Empirical OM:** identification (what makes the effect causal), measurement in decision-relevant units, unit of analysis, external validity to other operations settings.32- **Behavioral OM:** design, incentives, randomization, manipulation and attention checks, and **operational realism** (does the lab task map to a real operations decision?).33- **Operations data science:** validation design, leakage checks, and — critically — the **operational value**: does a better prediction change a feasible policy (predict-then-optimize)?34- **Simulation:** parameter provenance, validation against known cases, sensitivity, and a replication package.3536## The POM bar on method3738A method exists to serve an **OM contribution** judged interesting to practicing managers. If the paper's value is mainly a technical advance with thin OM decision content, a methods journal may fit better. Keep heavy derivations, solver details, and extra robustness for the unlimited online **e-companion** so the 32-page main document stays focused.3940## Execution bridge (StatsPAI / Stata MCP)4142For the **empirical / causal lane**, estimate and audit rather than only specify. Full43map: [`execution-with-mcp`](../../../shared-resources/empirical-methods/execution-with-mcp.md). POM spans analytical and empirical OM; apply the chain below to its empirical-OM papers, and note when a contribution is analytical / optimization.4445- `detect_design` → `recommend` → fit with `as_handle=true` → `audit_result` to46 enumerate the checks the design owes.47- **Panel / staggered DiD:** `callaway_santanna` / `sun_abraham` + `bacon_decomposition`48 + `honest_did_from_result`. **IV:** `effective_f_test` + `anderson_rubin_ci`. **RDD:**49 `rdrobust` + `mccrary_test`.50- **Experiments:** randomization-based inference and `romano_wolf` for the many-outcome51 family-wise correction reviewers expect.5253Match the toolchain to the **reviewer pool**, and report the effect size the venue54wants. A run end-to-end (synthetic data, real returns) is in the55[JF execution walkthrough](../../../Journal-of-Finance-Skills/resources/worked-examples/02-execution-walkthrough.md).56## Checklist5758- [ ] Method matches the operations question and the target Department59- [ ] Analytical: solution concept + structural results + assumption-relaxing robustness60- [ ] Empirical: identification strategy and decision-relevant measurement stated61- [ ] Behavioral: randomization, manipulation checks, operational realism62- [ ] Data science: validation, leakage, and decision/operational value63- [ ] Derivations/extra material slated for the e-companion646566## Methods pass for Production and Operations Management6768Use this as a second-pass capability check. First lock the operational decision, the performance metric, and the implementable lever; then test whether the manuscript addresses POM reviewers who want operational insight tied to production, service, supply-chain, or platform decisions.6970- **Primary move:** Name assumptions, diagnostics, robustness, falsification, and failure modes; do not accept a method section that hides the decisive validity threat.71- **Decision ledger:** return `claim / evidence / blocker / next edit` rows so the next pass can patch the manuscript directly.72- **Neighbor test:** compare against Management Science for broader OR/MS theory, Operations Research for method-first optimization, MSOM for manufacturing/service operations depth; if the neighboring outlet has the stronger audience claim, recommend re-routing before polishing.73- **Verification floor:** before submission-ready advice, re-open `resources/official-source-map.md` for volatile rules and name the one unresolved fact that could change the recommendation.7475## Output format7677```78【Method family】optimization / stochastic / game-theory / empirical / behavioral / simulation / data-science79【Operations question】<decision problem>80【Validity risks】assumptions / identification / measurement / leakage / validation81【Practice tie】how the method yields a manager-usable result82【e-companion plan】proofs / extra analyses to move online83【Next step】pom-data-analysis84```
brycewang-stanford