{"schema_version":1,"experiment_id":"eoa_inverse_innovation_exp09_archetype_breadth150_20260804","cell_id":"additive_measure_space_design__accounting_auditing","arm":"BREADTH_PROBE_ONE_SHOT","candidate_id":"additive_measure_space_design__accounting_auditing__P1","proposal_index":1,"version":0,"title":"Partition-Stable Audit Testing Coverage Ledger","problem":"An audit team estimates how much of a transaction population has been tested by adding the amounts or counts selected by separate procedures, even though journal-entry testing, revenue testing, confirmation work, and fraud-focused testing may cover the same underlying transactions. Each procedure total can be correct while their sum overstates distinct coverage because population identity, admissible subsets, units, overlaps, and exclusions are not governed by one coverage measure.","actors":["Audit engagement partner accountable for the assurance conclusion","Audit manager maintaining the population and testing plan","Audit staff executing procedures and recording tested items","Entity accounting staff supplying transaction extracts and identifiers","Engagement quality reviewer or internal inspection reviewer","Audit committee receiving coverage representations"],"observable_state":"The audit file contains multiple procedure worksheets whose tested-item counts or amounts are summed into an overall coverage figure; identical transaction-exposure units appear in more than one worksheet, exclusions and unmatched records are handled inconsistently, and regrouping procedures changes the reported total despite unchanged underlying work.","consequence":"The engagement may present an inflated or partition-dependent view of distinct transaction exposure tested, obscure untested portions of the population, and allocate additional work using a misleading coverage total.","affected_objective":"Produce a structurally auditable, partition-stable measure of distinct transaction exposure covered by completed audit procedures without treating coverage as sufficient evidence of audit quality or conclusion validity.","intervention":"Create a shadow coverage ledger before using aggregate testing coverage. Declare the in-scope transaction population and reporting period; establish a stable identity for each transaction-exposure unit; define which subsets generated by audit procedures, accounts, periods, and risk strata are measurable; assign each unit a nonnegative coverage mass such as one identity-stabilized transaction or absolute currency exposure; set the empty subset to zero; compute the union of completed-procedure subsets rather than summing overlapping worksheet totals; record unmatched, disputed, zero-mass, and out-of-scope items separately; test alternative disjoint partitions for equal recomposition; and attach the measure definition and limitations to every downstream coverage statement.","structural_mapping":[{"archetype_element":"Measurable Universe Scope","domain_realization":"The frozen ledger population for specified entities, accounts, transaction dates, and extract version covered by the engagement calculation."},{"archetype_element":"Measurable Subset Family","domain_realization":"Subsets defined by documented transaction identifiers and approved filters for procedures, accounts, periods, entities, and risk strata; ambiguous or unreproducible filters enter boundary review."},{"archetype_element":"Empty-Set Zero Rule","domain_realization":"A completed procedure matching no in-scope transaction-exposure units contributes zero coverage mass, regardless of hours spent."},{"archetype_element":"Nonnegative Size Assignment Rule","domain_realization":"Coverage mass is either a count of unique transaction identities or nonnegative absolute currency exposure; signed ledger balances are not used directly as measure mass."},{"archetype_element":"Disjoint Additivity Rule","domain_realization":"Distinct coverage equals the sum of masses for disjoint transaction subsets; overlapping procedure selections are combined by union before aggregation."},{"archetype_element":"Normalization or Scale Anchor","domain_realization":"Coverage is reported in declared units and, if expressed as a proportion, divided by the total measurable mass of the frozen in-scope population."},{"archetype_element":"Null-Set and Negligibility Policy","domain_realization":"Zero-amount entries, unmatched identifiers, disputed duplicates, and excluded records remain visible and cannot be interpreted automatically as zero audit relevance."},{"archetype_element":"Partition Consistency Register","domain_realization":"A table compares whole-population mass with recomposed masses under procedure, account, period, and risk-stratum partitions and records gaps or overlaps."},{"archetype_element":"Measurability Boundary Queue","domain_realization":"Transactions lacking stable identity, reproducible filter membership, or compatible units are quarantined for manager resolution rather than silently counted or discarded."},{"archetype_element":"Integration and Downstream Use Contract","domain_realization":"Reports state that the measure describes distinct population exposure touched by completed procedures, not evidence quality, control effectiveness, detection probability, or sufficient appropriate audit evidence."}],"mechanism_mapping":[{"mechanism_slug":"measure_space_specification","role":"Defines the frozen audit population, admissible subsets, coverage mass, scale, exclusions, and permitted interpretations before aggregation.","counterfactual_removal":"Without the specification, teams can combine figures derived from different populations, identities, or units and still label the result overall coverage."},{"mechanism_slug":"partition_sum_table","role":"Recomposes coverage across disjoint procedure, account, period, and risk partitions to expose overlaps, gaps, and boundary drift.","counterfactual_removal":"Without the table, a union total could appear plausible while remaining dependent on an undocumented partition or incomplete population mapping."},{"mechanism_slug":"finite_or_countable_additivity_test","role":"Checks that disjoint tested subsets sum to the same mass as their union under the declared rule.","counterfactual_removal":"Without the additivity test, the ledger does not establish that its coverage total is stable when audit work is regrouped."},{"mechanism_slug":"null_set_policy_register","role":"Separates zero mass from unmatched, excluded, disputed, or operationally important cases.","counterfactual_removal":"Without the register, exceptional transactions can disappear from the calculation and be mistaken for irrelevant cases."},{"mechanism_slug":"measure_invariance_review","role":"Repeats the calculation under approved alternative procedure and account partitions while holding universe and units fixed.","counterfactual_removal":"Without invariance review, a coverage figure may change merely because the same work was classified differently."}],"causal_chain":["Separate audit procedures select overlapping subsets of one transaction population.","Procedure worksheets summarize their selections independently, so locally correct totals contain an unmarked intersection.","Adding those totals counts transaction-exposure units once per procedure rather than once per distinct unit.","The resulting coverage figure changes with procedure boundaries and can conceal the mass of the untested complement.","A declared universe, stable membership rule, and nonnegative coverage mass make each procedure selection a measurable subset of the same space.","Union-before-sum aggregation and partition additivity tests remove overlap inflation and expose gaps, ambiguous membership, and incompatible units.","The resulting coverage statement remains stable under approved repartitioning and carries an explicit contract limiting what audit conclusions may be drawn from it."],"baseline":"The baseline is a spreadsheet-based engagement summary that copies tested counts or amounts from individual procedure workpapers, applies ad hoc deduplication when obvious duplicates are noticed, and reports the sum or ratio without a common population specification, boundary queue, partition test, or null-case policy.","nearest_rivals":["Transaction-ID deduplication across workpapers, which removes exact duplicate identifiers but does not by itself define the population, size unit, admissible subsets, or treatment of unmatched and zero-mass records.","An audit coverage matrix mapping procedures to assertions or controls, which describes conceptual coverage but does not assign additive mass to distinct transaction subsets.","A stratified audit sampling plan, which partitions a population for selection but does not reconcile overlap across multiple completed procedures into one union measure.","Ledger-to-workpaper reconciliation, which checks completeness and agreement of extracts but need not make aggregate testing coverage partition-stable."],"remaining_contrastive_claim":"Conditional on stable transaction identity and a valid nonnegative exposure unit, this intervention makes aggregate tested-population coverage a measure of the union of procedure subsets and tests whether it survives repartitioning; unlike simple deduplication or a coverage matrix, it also governs scope, measurability boundaries, normalization, null cases, and downstream interpretation. It does not claim that greater measured coverage implies better evidence or a justified audit opinion.","authority_safety":{"decision_authority":"The engagement partner retains authority over audit scope, materiality, sufficiency of evidence, and the audit opinion; the audit manager may authorize only the shadow coverage calculation within the approved engagement population.","authorized_first_step":"The audit manager may authorize a read-only retrospective reconstruction using frozen extracts and completed workpapers, with no change to the audit opinion, client ledger, sampling decisions, or filed report.","excluded_actions":["Changing the entity's accounting records or transaction identifiers","Replacing professional judgment about evidence sufficiency with the coverage measure","Using absolute exposure mass to infer misstatement direction or net balance","Silently resolving ambiguous identities, exclusions, or nonmeasurable records","Reopening, expanding, or curtailing audit procedures solely from pilot output","Communicating a revised assurance conclusion to the client, audit committee, regulator, or public"],"halt_rollback":"Stop if source extracts cannot be frozen, identity collisions cannot be resolved without altering records, incompatible currencies or units cannot be separated, or confidential data cannot remain within approved systems. Preserve the original workpapers, label all outputs experimental, revoke downstream access to the shadow metric, and discard the derived mapping subject to engagement retention rules."},"negative_tests":{"strongest_counterevidence":"Across independently constructed procedure, account, period, and risk-stratum partitions, the existing baseline already yields the same distinct union coverage, documents all overlaps and exclusions, and preserves unmatched and zero-mass cases under a common population and unit definition.","problem_falsifier":"The candidate problem is not present if the engagement never combines coverage across procedures, or if the procedure subsets are demonstrably disjoint and tied to one frozen population with stable identities and compatible units.","intervention_falsifier":"The intervention fails if its union total changes under logically equivalent approved partitions with the universe and mass rule held fixed, if material records cannot be assigned reproducible membership, or if reviewers cannot reconstruct the total and complement from the specification and register.","risks":["A stable additive figure may create false confidence even though coverage mass does not represent evidence quality or detection probability.","Absolute currency exposure can overemphasize high-volume entries and suppress the importance of small but risky or unusual transactions.","Identity-resolution errors can merge distinct transactions or split one transaction into several units.","Restricting the measurable family can exclude contested records and make the computed portion appear cleaner than the full population.","The mapping ledger can expose sensitive transaction relationships or expand access to confidential audit data.","Building the measure may add documentation burden without improving a decision if aggregate coverage is not used downstream."]},"next_evidence_step":"On one closed engagement selected by the engagement partner, perform a read-only shadow reconstruction for one frozen transaction population and no more than three overlapping audit procedures. Predeclare count mass and one compatible absolute-currency mass, generate the union and untested complement, populate the boundary queue, and test recomposition under procedure and account partitions. Have one reviewer not involved in construction reproduce the totals from the specification. End after this bounded reconstruction and record whether overlap, nonmeasurable records, partition instability, or downstream interpretive ambiguity was observed; do not modify the audit file's conclusions or current procedures.","prior_art_status":"UNSEARCHED","diversity_from_prior_proposals":"Not assessed against other proposals because runtime isolation prohibits inspecting them; this candidate is derived solely from the supplied archetype and accounting-and-auditing domain card.","revision_record":{"parent_version":null,"progress_targets_addressed":["One-shot initial candidate only"],"conceptual_changes":[],"operational_changes":[],"evidence_changes":[],"claim_changes":[]}}