← Risk register SOC 15-2031 · reviewed 2026-08-11

Operations Research Analysts

108,510 US workers · median $88,940/yr · Tech

EXPOSED

The daily work — pulling and cleaning data, coding optimization and simulation models in Python/R, running scenario analyses, and writing up findings in decks and memos — sits squarely in the zone current AI handles at usable quality. What persists is upstream: correctly framing a messy business problem as a solvable model, choosing which constraints and objectives actually reflect what the organization wants, and defending a recommendation that reroutes a supply chain or reshapes a staffing plan. There is no license and no signature requirement, so nothing regulatory slows the substitution; the modal analyst's output is an artifact, not a relationship.

10-year outlook: Headcount likely flattens or shrinks as one senior analyst plus AI covers what a team of three did, with the surviving roles concentrated in stakeholder-facing model design and decision ownership rather than model building.

Score — 34/100 resistance

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 9 + 3 + 2 + 8 + 12 = 34.

Task resistance 9/20

Mixed — a routine tier and a judgment tier. Writing the LP formulation, tuning solver parameters in Gurobi/CPLEX, building discrete-event simulations, and generating sensitivity tables are all things an LLM now does at first-draft quality; the 9 rather than a 5 reflects that the problem-framing step — deciding that a nurse-scheduling complaint is actually a demand-forecasting failure, and that the constraint the client insists on is the one to relax — still requires sitting in rooms with operations staff who cannot state their own objective function.

Embodiment 3/20

Fully desk- and screen-based. A 3 rather than 0 accounts for the occasional plant walkthrough, warehouse time-and-motion observation, or ride-along to see why the routing model's assumptions don't match how drivers actually load the truck — real, but a few days a quarter against a full-time seat in front of a solver and a BI tool.

Liability shield 2/20

No licence, no signature requirement. There is no state license, no PE-style stamp, and no statutory sign-off on an optimization recommendation; the 2 rather than 0 is only for the narrow slice working under DoD/FFRDC clearance requirements or SOX-adjacent model-governance regimes, where the barrier is the clearance and the model-validation paperwork, not the analyst's credential.

Trust premium 8/20

Some relationship component. An 8 rather than a 3 recognizes that internal analysts who have earned the plant manager's or the CFO's confidence get their recommendations adopted while equally correct models from outside consultants get shelved — but the deliverable is still a deck and a model file that can be handed to a successor, so the relationship accelerates adoption rather than constituting the product.

Judgment & accountability 12/20

Meaningful discretion. A 12 sits above procedural work because you choose the objective — minimize cost, or minimize worst-case delay, or keep the union contract intact — and that choice is a value judgment nobody hands you in a spec; it stops short of 14+ because the executive who approves the network redesign owns the consequence, and your name appears on the analysis, not the decision.

Scored twice. An independent second run returned 34/100 — EXPOSED, agreeing with the verdict above.

This score sits on a verdict boundary. At 34/100 it is one point from COOKED. Re-scoring moves results by a point or two, so here the score is more informative than the label.

Confidence: high · reviewed 2026-08-11 · how scoring works

Tasks already automatable

What survives

Active moats: judgment, trust

How to future-proof this job

Field report — do you do this job?

Has AI actually changed your work?

Self-reported and unverified — a sentiment signal, not a survey. One response per person per occupation; you can change your answer.

From people who do this job

Nobody has filed one yet. If you do this work, you know things the rubric can't see.

What has actually changed in your work?

Concrete beats general: a tool that arrived, a task that moved, a headcount decision you watched happen. Don't include anything that identifies you or your employer if that would put you at risk.