Lexicographic Priority Rule¶
Selection rule — instantiates Greedy Stepwise Commitment
Ranks each choice by a fixed hierarchy of criteria, consulting a lower criterion only to break ties left by the ones above it — never trading a worse top criterion for a better lower one.
Lexicographic Priority Rule compares candidates the way a dictionary orders words: on the first criterion alone, then on the second criterion only among the candidates that tie on the first, and so on down a fixed hierarchy. There is no weighting and no trade-off — a candidate that leads on the top criterion wins no matter how far it trails on every criterion below. That absolute, non-compensatory priority is the one idea that makes it this rule and not a weighted score: it encodes which considerations may never be traded against each other, and demotes the rest to pure tie-breakers.
Example¶
An emergency department must decide who is seen next as patients keep arriving. It applies a strict priority order: first the acuity level — a five-level clinical read of how urgently care is needed, as in the real Emergency Severity Index — and only among patients at the same acuity level, earliest arrival time. A patient assessed at acuity 2 is always taken before any acuity 3, even one who has waited hours; wait time is consulted solely to order patients who share an acuity level.
The result is a queue that is defensible and instantly explainable — "sicker first, then longest-waiting" — without anyone having to agree how many minutes of waiting equal one acuity level. That exchange rate is exactly the trade the rule refuses to make, which is what lets it hold up under scrutiny when a weighted triage score would be argued over.
How it works¶
- Fix the criteria in priority order before any choice is made; the ordering is the design decision that carries all the weight.
- Select the candidate that is best on the first criterion. If several tie — and only then — compare those on the second criterion; repeat down the hierarchy.
- Because lower criteria act only inside ties, a difference on a high criterion, however small, dominates any difference, however large, on a lower one. The rule never sums or averages criteria.
Tuning parameters¶
- Criterion ordering — which consideration is supreme. The single highest-leverage dial: reorder it and the entire ranking changes.
- Tie band — whether "tie" means exactly equal or within a tolerance on a criterion. A wide band hands more decisions down to lower criteria; a zero band lets the top criterion decide almost everything.
- Hierarchy depth — how many criteria are listed before falling through to an arbitrary tie-break; deeper hierarchies resolve more cases but get harder to justify.
- Final tie-break — FIFO, random, or a coin-flip when every listed criterion ties.
When it helps, and when it misleads¶
Its strength is transparency: it needs no exchange rates between incommensurable criteria, so it is ideal where some priorities are meant to be absolute — safety over cost, legal precedence, clinical urgency — and where a weighted score would be disputed or gamed.
Its central failure mode is that it is non-compensatory: it discards the magnitude of every criterion below the deciding one, so a hair's-breadth lead on criterion 1 can block a candidate that is overwhelmingly better on criteria 2 and 3.[n1] And because the ordering is fixed up front, any error in it is baked into every decision the rule ever makes. The classic misuse is to pick the desired winner first and then arrange the criteria order so it comes out on top. The discipline that guards against this is to justify the ordering from genuine, stated priorities, and to widen the tie band wherever a trivial top-criterion difference should not be allowed to decide.
How it implements the components¶
local_priority_score— the score here is not one number but an ordered vector of criteria; the rule reads it left to right instead of summing it.selection_and_tie_break_rule— its heart: take the best on the top criterion, and descend to the next criterion only to break an exact tie.diversity_or_fairness_quota— a strict priority class (clinical acuity, legal precedence, protected need) is enforced here as an ordering, where a higher-priority claimant always precedes a lower one, rather than as reserved slots.
It does not represent or generate the candidates it ranks — that state and generator belong to Nearest-Neighbor Route Extension — nor maintain them in a live re-prioritizing engine, which is Priority-Queue Step Selection; this is a pure comparison policy.
Related¶
- Instantiates: Greedy Stepwise Commitment — supplies the per-step selection-and-tie-break logic the pattern needs.
- Sibling mechanisms: Priority-Queue Step Selection · Sorted Candidate Sweep · Nearest-Neighbor Route Extension · Shortest-Processing-Time-First Rule · Trap-Sentinel Escalation · Highest-Marginal-Gain-First Rule · Dijkstra-Style Frontier Expansion · Earliest-Deadline-First Dispatch · Greedy Assignment Pass · Greedy Set-Cover Heuristic · Kruskal-Style Edge Acceptance
Editorial Notes¶
Form Classification¶
Form family: Rule, Policy & Commitment
Rationale: Lexicographic Priority Rule operates as a standing rule, threshold, contractual commitment, or policy constraint governing future conduct because it ranks each choice by a fixed hierarchy of criteria, consulting a lower criterion only to break ties left by the ones above it — never trading a worse top criterion for a better lower one.
Independent corroboration: The frozen evidence defines Lexicographic Priority Rule as 'Ranks each choice by a fixed hierarchy of criteria, consulting a lower criterion only to break ties left by the ones above it — never trading a worse top criterion for a better lower one', so its operative form is Rule, Policy & Commitment.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Operations Research
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Lexicographic optimization and priority rules are canonical operations-research methods for ordered, non-compensatory objectives.
Related originating lineages:
- Economics & Finance — Lexicographic preferences independently developed as a model of non-compensatory choice.
- Mathematics — Lexicographic order supplies the formal comparison structure on which the decision rule depends.
Review resolution: Both independent reviews assign primary provenance to operations_research. The queued secondary differences (alternate_origin_disagreement, origin_mode_disagreement) are reconciled by retaining economics_finance, mathematics only as formative or independently established lineage(s), not merely as application domains. origin_mode=cross_disciplinary_synthesis records the provenance relationship, while domain_reach=multi_domain separately records applicability breadth. confidence=high preserves the more cautious assessment, and encyclopedia_synthesis=false records whether either reviewer identified a corpus-specific synthesis.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
A lexicographic rule decides how to compare, and nothing else — it presumes some other component has already produced the candidates and their criteria values. It is most natural as the selection core inside a larger greedy loop: keyed into Priority-Queue Step Selection or driving a Sorted Candidate Sweep.
[n1] In decision theory a lexicographic preference ranks options one attribute at a time and never trades a better lower attribute for a worse higher one. Its defining property — and its standard criticism — is that it ignores the intensity of every attribute except the one that happens to decide. ↩