pom-theory-development

Model & Theory Development (pom-theory-development)

When to trigger

• You have an operations question but no formal model or sharp mechanism
• Your assumptions are mathematically convenient but operationally unmotivated
• A reviewer says "the model is not well grounded in practice" or "the mechanism is unclear"
• Empirical results are interpreted as bare correlations with no operations mechanism

POM's dominant track: analytical modeling — done rigorously

POM is historically anchored in analytical/mathematical modeling (operations-research / management-science-style optimization, stochastic modeling, and game theory), so theory development most often means building a model, not deriving verbal hypotheses. Make every modeling choice defensible:

• Decision primitive & objective. Name the operational decision (capacity, inventory, pricing, scheduling, sourcing, routing, staffing) and the objective (cost, profit, fill rate, waiting time, welfare).
• Assumptions tied to operations reality. Justify each assumption (lead times, demand distribution, information structure, rationality) against an operational constraint — not analytical convenience.
• Solution concept. For optimization, characterize the optimal policy (e.g., base-stock, (s, S), threshold) and structural properties (monotonicity, convexity). For game-theoretic models, state the equilibrium concept (Nash, Stackelberg), and prove existence/uniqueness.
• Propositions over hand-waving. State results as numbered propositions/theorems; full proofs go to the e-companion, with intuition in the main text.

Empirical, behavioral, and data-science tracks

• Empirical OM: derive hypotheses from an operations mechanism, then identify it with archival data; specify the estimand and the causal logic before estimation.
• Behavioral OM: ground predictions in decision biases (e.g., newsvendor pull-to-center, bullwhip from misperception of feedback) and design the experiment to isolate the mechanism.
• Operations data science: state the prediction/decision task and the operational loss the model improves (e.g., predict-then-optimize), not accuracy for its own sake.

The practice-relevance gate

Every mechanism must answer: what does a practicing operations manager do differently? Convert results into decision levers (a policy, a contract term, a staffing rule), not just comparative statics. POM weights this alongside rigor.

Checklist

• [ ] Decision primitive, objective, and assumptions stated and operationally justified
• [ ] Analytical: solution concept, structural results, equilibrium existence proven (proofs in e-companion)
• [ ] Empirical/behavioral: mechanism → hypothesis → identification/experiment chain explicit
• [ ] Result expressed as a managerial lever, not only a sign of a derivative
• [ ] Mechanism generalizes beyond the focal instance/dataset

Anti-patterns

• Method novelty substituted for an OM theory contribution.
• Assumptions chosen for tractability with no operational story.
• "Curvature" or a significant interaction reported with no operational mechanism.
• Practice implications that merely restate the result without an actionable decision.

Operating 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: Return a claim-evidence-risk ledger; every recommendation must point to a manuscript location or missing artifact.
• 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

【Decision primitive】capacity / inventory / pricing / routing / staffing ...
【Mechanism】queueing / contracting / incentives / learning / behavioral friction ...
【Result form】proposition/theorem (proof→e-companion) | hypothesis (→identification)
【Assumption risk】assumption + operational defense
【Practice lever】decision changed by the result
【Next step】pom-literature-positioning or pom-methods