Works from CSV or workbook exports produced from SAP, IBM Maximo, Oracle ERP, Hexagon EAM, Infor, and other ERP, EAM, or CMMS systems. No direct integration or write-back is required — Review data requirements →
PartsCleanse AI: MRO catalog intelligence engine

PartsCleanse AI — MRO Catalog Intelligence Diagnostic

Identify duplicate spare-parts families, manufacturer ambiguity, part-number variation, UOM inconsistency, obsolete records, and MRO catalog-quality risk from exported item master and spare-parts data without writing back into ERP or EAM systems. The workflow starts from exported operational data, produces PartsCleanse AI Catalog Diagnostic Report, and keeps source systems untouched.

Run Free Industrial IQ SnapshotView Sample ReportRequest Diagnostic Pilot
Read-onlyDiagnostic
No ERPWrite-back
Source-filePurge after report generation
HumanReview before action
Evidence summary

Diagnostic evidence path

PartsCleanse AI: This engine path connects the operating problem, exported source data, diagnostic logic, evidence output, review boundary, and next action for Industrial IQ buyers. Diagnose duplicate spare parts, weak descriptions, UOM conflicts, obsolete candidates, and material-master cleanup priority from exported item data without ERP.

Run This Engine
Review ownerThe operating, finance, procurement, maintenance, data, and governance leaders accountable for this diagnostic decision.
Input file contextTypical exported operational files, helpful optional fields, sample data, and mapped CSV or workbook evidence.
Diagnostic evidenceSource-backed findings, confidence tiers, score interpretation, report preview, review actions, and recurring-use path.
Recommended actionInspect the sample report, then run the matching engine with bounded exported operational data.
Executive decision snapshot

Problem, data, output, trust boundary, owner, and next action.

PartsCleanse AI keeps the product review concrete: operating problem, typical exports, output, review owner, trust boundary, and next action stay visible before upload.

Read-only diagnostics. Source files purged after report generation. Human review before action.
Problem diagnosed

MRO Catalog Intelligence Diagnostic

Data needed

item code, description, manufacturer

Output produced

PartsCleanse AI Catalog Diagnostic Report

System boundary

Read-only; no ERP/EAM write-back.

Review owner

CFO, CIO, Procurement, Maintenance, and Materials leaders

Next action

Run Snapshot, review sample report, or scope a pilot.

One platform, eight diagnostic engines

PartsCleanse AI is one engine inside the Industrial IQ platform.

The selected diagnostic is shown inside the broader Industrial IQ operating model: exported operational data, source-backed evidence, confidence tiers, reports, action tracking, and no ERP write-back.

Compare all engines
CFO, CIO, Procurement, Maintenance, and Materials leaders

Catalog Intelligence

Catalog health score

 CFO, COO, Inventory, Materials, and Supply Chain leaders

Inventory Risk Intelligence

Inventory health score

 CPO, Procurement Director, CFO, and Supply Chain leaders

Procurement Leakage Intelligence

Procurement leakage score

 CFO, Finance Head, Procurement, and Board advisors

Working Capital Intelligence

Working capital score

 Asset Integrity, Maintenance, Reliability, and Operations leaders

Asset-to-Part Intelligence

Asset intelligence score

 Maintenance Director, Reliability Manager, COO, and Plant leaders

Maintenance Readiness Intelligence

Maintenance readiness score

 CIO, CTO, COO, Data Governance, and AI Transformation leaders

AI Readiness Intelligence

AI readiness score

 CISO, CIO, Audit, Governance, and Transformation leaders

Evidence Governance Intelligence

Governance readiness score

Product review card

One diagnostic decision, visible before upload.

PartsCleanse AI shows what source data can prove, which findings are ready for review, and what remains buyer-controlled after the report.

Input files to prepare

item code, description, manufacturer, manufacturer part number

First diagnostic question

Can your exported material master reveal duplicate families, unsafe lookalikes, and duplicate-capital exposure in 15 days?

Evidence returned

duplicate family, matching reason, discriminator conflict, confidence tier

Review owner

CFO, CIO, Procurement, Maintenance, and Materials leaders

Trust contract
Export-first Start from buyer-controlled CSV or workbook exports; no production integration is required for the first review.
Evidence before action Findings show source context, confidence, assumptions, limitations, and owner-review status.
Governed boundary No ERP write-back, no uncontrolled remediation, and human review before operational change.
Catalog evidence

PartsCleanse AI makes its source-to-decision path visible before upload.

Duplicate-family evidence, material description quality, and cleanup priority queue.

InputDescription
OutputPartsCleanse AI Catalog Diagnostic Report
Trust boundaryRead-only diagnostic, no ERP write-back, confidence-tiered evidence, and human review before action.
PartsCleanse AI catalog intelligence workflow showing duplicate detection, normalization, and MRO data quality improvement.
PartsCleanse AI focuses on MRO material master and spare-parts catalog diagnostics inside the broader Industrial IQ platform.
Industrial Evidence Graph

PartsCleanse AI converts source records into governed evidence.

PartsCleanse AI follows the same Industrial IQ trace: exported files, field fit, diagnostic lens, confidence marker, human review, read-only boundary, and audit-ready action.

01 Source tile

Exported ERP, EAM, CMMS, inventory, procurement, finance, asset, or work-order records.

02 Evidence trace

Mapped fields, source rows, reason codes, and continuity from file to finding.

03 Diagnostic lens

Catalog Intelligence evaluates the operating question with controlled engine logic.

04 Confidence marker

High, moderate, review-required, or insufficient evidence states use labels and shapes, not color alone.

05 Human review gate

Accountable owners review exceptions, limitations, and next actions before remediation.

06 Read-only boundary

One-way export path. No autonomous ERP, EAM, CMMS, procurement, or inventory write-back.

07 Decision / audit node

Owner, action status, evidence source, report output, and audit metadata stay reviewable.

Visual boundary: this is an original AI2COE information-design system. It represents diagnostic evidence flow, not customer proof, benchmark results, automated remediation, or ERP integration.
First diagnostic proof pack

Test PartsCleanse AI with the smallest credible evidence pack.

Before a buyer shares private data, this proof pack shows the source export, field-mapping gate, source-fit gate, diagnostic signal, evidence output, and repeat path for Catalog Intelligence.

Material or item master CSV

Source export

Start with description, manufacturer, MPN, supplier, UOM, site, value.

Required fields

Mapping gate

Confirm Description. Add Material Id, Manufacturer, Mpn, Supplier where available.

Catalog health score

Source-fit gate

Weak coverage is labeled as an assumption or limitation before scoring.

PartsCleanse AI

Diagnostic signal

Findings show source rows, match reasons, discriminator conflicts, review level, assumptions, and owner status before cleanup action.

PartsCleanse AI Catalog Diagnostic Report

Evidence output

Review duplicate family, matching reason, discriminator conflict, confidence tiers, assumptions, limitations, and owner actions.

Score history and action tracker

Repeat path

Rerun after owner review to compare score movement and open findings.

View sample report Download sample CSV Mapping template Compare alternatives

Control boundary: read-only diagnostic, no ERP write-back, source-file purge after report generation, confidence-tiered evidence, and human review before action.

Output command console

PartsCleanse AI produces a buyer-reviewable output bundle, not a black-box score.

The output UX is designed for executive reading and data-owner inspection: score, evidence table, confidence, report pack, action tracker, score history, export artifacts, and review ownership stay connected.

Open sample output
Catalog health score

Engine score

Score is a diagnostic interpretation, not a certified rating.

duplicate family, matching reason, discriminator conflict, confidence tier

Evidence table

Rows show source context, reason codes, confidence, assumptions, and limitations.

High, medium, low, needs review

Confidence and limits

Findings stay separated by source quality before owner action.

PartsCleanse AI Catalog Diagnostic Report

Executive report pack

Report sections include Catalog health score, duplicate-family evidence, unsafe-match controls, CFO exposure view.

Accept, reject, defer, assign, request more data

Action tracker

Output becomes governed work only after buyer review.

Baseline, rerun, movement, open findings

Score history

Recurring runs show what changed after owner decisions.

HTML, PDF, CSV evidence, mapping, dictionary

Output artifact kit

Data owners and executives can inspect the same report package.

CFO, CIO, Procurement, Maintenance, and Materials leaders

Review owner

The accountable owner reviews evidence before remediation or system change.

HTML sample PDF report Sample CSV Data dictionary Run Snapshot

Output boundary: sample outputs demonstrate structure only. Uploaded-data diagnostics are source-backed, confidence-tiered, no-write-back, and human-reviewed before action.

Product experience

Inspect the upload, mapping, evidence, score, report, and action workflow before private data is uploaded.

These are illustrative Industrial IQ UI previews using sample/demo labels. They show the enterprise workflow buyers should expect: read-only upload, mapped fields, confidence-tiered evidence, executive reporting, and governed action tracking.

Industrial IQ in 2 Minutes
Industrial IQ illustrative upload workflow preview Illustrative product UI preview

Upload operational data

CSV/workbook exports from ERP, EAM, CMMS, inventory, procurement, asset, and maintenance systems. No ERP write-back.

Industrial IQ field mapping preview Illustrative product UI preview

Map fields

AI-assisted column matching exposes required fields, optional fields, source-fit score, and data readiness before the run.

Industrial IQ evidence table preview Illustrative product UI preview

Evidence table

Findings show source row, matched record, reason code, confidence, business impact, and reviewer status.

Industrial IQ diagnostic score preview Illustrative product UI preview

Diagnostic score

Score cards separate duplicate risk, source-fit, capital exposure band, and action priority.

Preview boundary: visuals use sample/demo content and do not represent customer data, approved customer outcomes, or financial commitments.

Product definition

PartsCleanse AI is a MRO Catalog Intelligence Diagnostic.

PartsCleanse AI is the Industrial IQ diagnostic for MRO catalog intelligence. It reviews exported material master, spare-parts catalog, manufacturer/OEM, part-number, UOM, inventory status, and site-level item data to produce source-backed duplicate-family evidence, confidence tiers, and remediation review queues before catalog cleanup or ERP/MDM remediation is approved.

Who should use it: This page is for CFOs, CIOs, procurement leaders, maintenance heads, reliability teams, and ERP/master-data owners who need evidence of MRO catalog duplication, item ambiguity, and catalog-quality risk before funding or approving cleanup.
First 10-second answer
What it diagnosesIdentify duplicate spare-parts families, manufacturer ambiguity, part-number variation, UOM inconsistency, obsolete records, and MRO catalog-quality risk from exported item master and spare-parts data without writing back into ERP or EAM systems.
What it producesPartsCleanse AI Catalog Diagnostic Report, evidence rows, confidence tiers, score interpretation, action queue, and owner review status.
Why it is safeRead-only diagnostic, no ERP write-back, source-file purge after report generation, and human review before action.
Operating problem

Real industrial symptoms this engine is designed to inspect.

Operating problem

The same spare appears under several material numbers with different abbreviations.

Operating problem

Planner search returns too many lookalike items, then misses the part actually in stock.

Operating problem

Manufacturer, OEM, and supplier names are mixed into description fields inconsistently.

Operating problem

UOM conflicts make equivalent records unsafe to merge without specialist review.

Operating problem

Old plant codes or legacy ERP migrations preserve duplicate material families.

Operating problem

Procurement buys a part that maintenance already owns under a different description.

Typical exported data

Start with files the buyer can export safely.

These fields describe the practical evidence pack for the diagnostic. They are not mandatory integration requirements, and they do not authorize ERP, EAM, CMMS, procurement, inventory, finance, or production-system write-back.

Exported field

item code

Exported field

description

Exported field

manufacturer

Exported field

manufacturer part number

Exported field

UOM

Exported field

plant or site

Exported field

commodity class

Exported field

stock value

Exported field

usage or PO references where available

Diagnostic logic

How PartsCleanse AI turns exports into review evidence.

This is a high-level product explanation, not proprietary algorithm disclosure. Findings remain confidence-tiered and require owner review before operational action.

Diagnostic layerHow it works
Map catalog fields Confirm item code, description, UOM, manufacturer, MPN, supplier, plant, and value fields before analysis.
Normalize industrial language Standardize abbreviations and aliases while preserving discriminators such as size, material, rating, model, and UOM.
Group candidate families Identify likely duplicate families and unsafe lookalikes with source-row evidence.
Score confidence Separate high-confidence matches, needs-review records, and blocked matches where conflicts make action unsafe.
Route cleanup backlog Create a steward-ready queue for ERP, procurement, maintenance, and master-data review.
Sample output preview

Illustrative evidence rows show the report format before private data is uploaded.

Rows below are sample/demo examples. Uploaded-data diagnostics replace illustrative rows with mapped source records, confidence tiers, assumptions, limitations, and owner review status.

Finding typeConfidenceSource evidenceRecommended review actionBuyer interpretationLimitation
Duplicate family candidate High Similar description, same MPN, different UOM and material IDs Review before material merge ERP and maintenance owners can test whether one family is safe to steward Lookalike parts remain review-required when specifications conflict.
Manufacturer ambiguity Needs review OEM in description but blank manufacturer field Assign catalog steward Procurement can separate supplier naming issues from true equivalence Missing manufacturer data lowers confidence.
What this product does not do

Credible diagnostics need visible boundaries.

Industrial IQ creates review evidence. It does not become an uncontrolled remediation system.

No automatic merge Does not merge, delete, retire, or overwrite material records.
No ERP write-back Does not write back to SAP, Maximo, Oracle, EAM, CMMS, or MDM systems.
No promised savings Does not treat duplicate exposure as realized savings without owner review.
No engineering approval Does not certify interchangeability or substitute fit.
No blind cleanup Does not hide unsafe lookalikes inside high-confidence groups.
Trust and data lifecycle

PartsCleanse AI moves from exported data to owner-reviewed action.

AI2COE uses exported operational data for diagnostic review. The workflow is read-only, does not write back into ERP, EAM, CMMS, procurement, finance, inventory, or production systems, and source files are purged after report generation. Open Findings, summary metrics, report ownership, continuity metadata, and audit events may be retained where needed for governance and reporting.

View Security Brief
01 Exported data Buyer-controlled CSV or workbook exports start the diagnostic.
02 Read-only diagnostic Industrial IQ inspects mapped fields and never writes back to source systems.
03 Evidence-backed report Findings include source context, confidence, assumptions, limitations, and owner action.
04 Source-file purge Source files are purged after report generation according to disclosed handling boundaries.
05 Governance metadata Open Findings, summary metrics, ownership, continuity metadata, and audit events may be retained where required.
Commercial path

Choose the next action by buyer readiness.

Visitor situationRecommended next stepWhy
I want to see proof first View Sample Report Inspect sample findings, confidence tiers, assumptions, limitations, and owner actions.
I have exported data ready Run Free Industrial IQ Snapshot Upload a bounded file, map fields, and route the question to the right engine.
I need executive review Request Founder-Led Pilot Scope a 15-day diagnostic with accountable reviewers, data readiness, security review, and report expectations.
I need commercial terms View Pricing Review diagnostic and recurring intelligence paths without changing trust or data-retention language.
I need security review Read Trust Center Review no-write-back, source-file handling, confidence tiers, human review, DPA, SLA, and security brief paths.
AI-readable summary

Concise entity definition for search and AI assistants.

PartsCleanse AI is the Industrial IQ MRO catalog intelligence diagnostic. It reviews exported item-master and reference fields to identify duplicate-family candidates, weak descriptions, manufacturer ambiguity, UOM conflicts, obsolete candidates, and cleanup priorities. The workflow is read-only, no-write-back, confidence-tiered, and human-reviewed before any master-data action.

Product decision room

Can your exported material master reveal duplicate families, unsafe lookalikes, and duplicate-capital exposure in 15 days?

Before cleansing the material master, diagnose duplicate families, weak descriptions, UOM inconsistency, field completeness, and duplicate-capital exposure.

duplicate familymatching reasondiscriminator conflictconfidence tiercapital exposure
Decision path Use the hero actions or final buyer journey panel to run the engine, inspect a sample report, see required data, or request pilot scoping.
Buyer intent
Prove duplicate SKU exposurePrepare SAP or Maximo migrationPrioritize catalog cleanupQuantify duplicate capitalCreate a review queue without ERP write-back
What this does not replace: MDM and data-cleansing platforms. It does not replace MDM. It prioritizes the records and duplicate families that should be cleaned first.
Export readiness

Prepare the minimum practical evidence pack before the diagnostic review.

Material master export item code, short and long description, UOM, commodity class, plant/site
Reference fields manufacturer, manufacturer part number, supplier, OEM, catalog class
Value context stock on hand, stock value, usage or PO references when available
After-report action path

Use the report to decide what should happen next.

01 Approve high-confidence duplicate families for steward review.
02 Route unsafe lookalikes to engineering, procurement, or MRO master-data owners.
03 Use the cleanup queue to scope MDM, SAP, Maximo, Oracle, or service-led remediation.
Product workflow

PartsCleanse AI is inspectable before private data is uploaded.

The product flow keeps the buyer's review sequence visible: source data, diagnostic logic, evidence, report, governance boundary, and next action.

1SelectStart with Catalog Intelligence or a buyer pack.
2UploadUse sample data first or map a private CSV/workbook export.
3ValidateCheck required fields, missing values, aliases, and readiness score.
4AnalyzeRun deterministic diagnostics with assumptions and limitations labeled.
5ReviewInspect evidence, review levels, action owners, and trust controls.
6ReportShare report output, score history, and next owner action.
Run Free Industrial IQ Snapshot View Sample Report Review Required Data
Best-fit ICPCFO, CIO, Procurement, Maintenance, and Materials leaders
Minimum dataDescription
Report outputPartsCleanse AI Catalog Diagnostic Report
Trust controlNo ERP write-back, owner review, review levels, and source-row evidence.
Report Preview and Output Contract

PartsCleanse AI Catalog Diagnostic Report tells leadership what happened, why it matters, and what to do next.

Every report separates sample or benchmark assumptions from uploaded-data evidence. It is designed for executive reading, analyst inspection, and owner-assigned review without automatic ERP change.

Catalog health score Source-backed section used by the buyer committee before action is approved.
duplicate-family evidence Source-backed section used by the buyer committee before action is approved.
unsafe-match controls Source-backed section used by the buyer committee before action is approved.
CFO exposure view Source-backed section used by the buyer committee before action is approved.
review queue Source-backed section used by the buyer committee before action is approved.
Open Sample Report Download Sample PDF Sample CSV
Evidence table preview
Duplicate FamilyVisible in the evidence table, score interpretation, or owner review queue.
Matching ReasonVisible in the evidence table, score interpretation, or owner review queue.
Discriminator ConflictVisible in the evidence table, score interpretation, or owner review queue.
Confidence TierVisible in the evidence table, score interpretation, or owner review queue.
Capital ExposureVisible in the evidence table, score interpretation, or owner review queue.
Buyer pack fit

PartsCleanse AI sits inside the eight-engine Industrial IQ platform.

Role-based packs make each engine commercially understandable for the buying committee while preserving the one-platform story.

Buyer packDecision supportedCompanion engines
CFO Pack Fund value realization only after capital exposure, carrying cost, procurement leakage, and review confidence are visible. Working Capital Intelligence, Inventory Risk Intelligence, Procurement Leakage Intelligence
CIO / ERP Pack Prove ERP, data, governance, and AI readiness before integration, automation, or migration decisions. AI Readiness Intelligence, Evidence Governance Intelligence, Asset-to-Part Intelligence
Procurement Pack Convert supplier, PO, duplicate stock, and price-variance evidence into sourcing action without unsupported savings claims. Procurement Leakage Intelligence, Inventory Risk Intelligence, Working Capital Intelligence
Maintenance / Reliability Pack Prove work-order readiness, asset-to-part coverage, critical-spare availability, and false-stockout risk before maintenance or reliability programs scale. Maintenance Readiness Intelligence, Asset-to-Part Intelligence, Inventory Risk Intelligence
18-industry product readiness map

PartsCleanse AI applies across asset-intensive environments without changing the no-write-back model.

The full industry ecosystem remains available in the industry hub. The first scan stays focused on representative contexts and the buyer decision.

Oil & Gas

Catalog Intelligence fit for Oil & Gas

Use exported operational data to test source-backed diagnostic readiness before action.

Mining

Catalog Intelligence fit for Mining

Use exported operational data to test source-backed diagnostic readiness before action.

Manufacturing

Catalog Intelligence fit for Manufacturing

Use exported operational data to test source-backed diagnostic readiness before action.

Utilities

Catalog Intelligence fit for Utilities

Use exported operational data to test source-backed diagnostic readiness before action.

Power Generation

Catalog Intelligence fit for Power Generation

Use exported operational data to test source-backed diagnostic readiness before action.

Chemicals

Catalog Intelligence fit for Chemicals

Use exported operational data to test source-backed diagnostic readiness before action.

Industry fit summary

Use the full engine route when the buying committee needs deeper matrices.

The first scan keeps attention on source exports, evidence output, buyer interpretation, trust boundaries, sample reports, and next action. Deeper scorecards and industry matrices remain available in the platform engine view.

Testing boundary: product and industry coverage is synthetic QA coverage and public-fit guidance, not customer proof, benchmark result, ROI claim, or certification.
Oil & GasMiningManufacturingFood & BeveragePharmaceuticalUtilitiesData CentersAviation MRO / AirlinesHealthcare SystemsRail, Metro & TransitTelecom Network OperatorsPorts, Marine Terminals & ShippingAerospace & Defense Maintenance DepotsWarehousing, Distribution Centers & 3PLCommercial Fleet, Trucking & LogisticsConstruction & Heavy Equipment FleetsHigher Education & Multi-Campus FacilitiesHospitality, Resorts & Gaming
View all industries View sample report
Diagnostic outcome evidence

What buyers inspect when they run PartsCleanse AI.

These cards show the decision frames, inputs, outputs, and proof status a buyer committee should expect. Benchmark and sample frames are labeled; uploaded data replaces assumptions with customer-specific evidence.

Evidence contract
CFO decision frame Assumption-labeled frame

PartsCleanse AI

Oil & Gas upstream · SAP / Maximo pre-migration item master

Assumption duplicate-rate planning input
5 executive report artifacts

"Use the diagnostic to replace catalog-risk assumptions with duplicate-family evidence, confidence tiers, and capital-exposure interpretation."

Finance, maintenance, procurement, and ERP owners

Primary action: use the PartsCleanse AI Snapshot CTA above.
Procurement decision frame Sample diagnostic frame

PartsCleanse AI

Manufacturing multi-plant · Material master, supplier aliases, UOM variants, and OEM references

3 alias / UOM / OEM signals
0 ERP write-back

"The report shows where duplicate item families can drive repeated buying, supplier leakage, and standardization work."

Procurement and master-data governance

Primary action: use the PartsCleanse AI Snapshot CTA above.

Claims discipline: public examples are benchmark/sample frames unless marked as approved customer evidence. Savings or remediation outcomes require uploaded-data diagnostics and human review.

Answer-first product brief

What PartsCleanse AI diagnoses.

PartsCleanse AI diagnoses MRO catalog quality from exported item-master data: duplicate families, weak descriptions, UOM conflicts, missing manufacturer fields, obsolete candidates, and cleanup priority before ERP or MDM remediation.

Recommended first diagnostic: Start with the material or item master export, then add manufacturer, MPN, supplier, UOM, plant, stock value, usage, and PO references where available.
Trust boundary
Read-onlyStarts from exported files; no production integration is required to begin.
No ERP write-backNo ERP, EAM, CMMS, procurement, inventory, or asset record is changed by the diagnostic.
Source-file purgeSource files are processed to generate the diagnostic report pack and then purged.
Human reviewFindings remain confidence-tiered and owner-reviewed before operational action.
Industry operating symptoms

Signals real industrial buyers recognize before they name the product.

Operating symptom

The same spare appears under several material numbers with different abbreviations.

Operating symptom

Planner search returns too many lookalike items, then misses the part actually in stock.

Operating symptom

Manufacturer, OEM, and supplier names are mixed into description fields inconsistently.

Operating symptom

UOM conflicts make equivalent records unsafe to merge without specialist review.

Operating symptom

Old plant codes or legacy ERP migrations preserve duplicate material families.

Operating symptom

Procurement buys a part that maintenance already owns under a different description.

Operating symptom

High-value records have weak descriptions, missing MPNs, or no clear steward.

Operating symptom

Catalog cleanup scope is too broad to fund without a ranked evidence backlog.

Required source files

Practical exports to prepare for Catalog Intelligence.

File groupFields that improve diagnostic confidence
Material master export item code, short and long description, UOM, commodity class, plant/site
Reference fields manufacturer, manufacturer part number, supplier, OEM, catalog class
Value context stock on hand, stock value, usage or PO references when available
Diagnostic method

How the engine turns exports into review evidence.

LayerHow it works
Map catalog fields Confirm item code, description, UOM, manufacturer, MPN, supplier, plant, and value fields before analysis.
Normalize industrial language Standardize abbreviations and aliases while preserving discriminators such as size, material, rating, model, and UOM.
Group candidate families Identify likely duplicate families and unsafe lookalikes with source-row evidence.
Score confidence Separate high-confidence matches, needs-review records, and blocked matches where conflicts make action unsafe.
Route cleanup backlog Create a steward-ready queue for ERP, procurement, maintenance, and master-data review.
Evidence outputs

What buyers receive after the diagnostic run.

OutputWhat it contains
Duplicate-family register Candidate families, source records, reason codes, discriminator conflicts, and confidence tier.
Catalog health score Field completeness, duplicate risk, UOM risk, manufacturer/MPN coverage, and review priority.
Cleanup action queue Owner-routed records for steward review before MDM, SAP, Maximo, Oracle, or service-led remediation.
Business problem

MRO Catalog Diagnostic.

MRO catalog deduplication, field quality, UOM consistency, and duplicate capital exposure.

PartsCleanse AI is evaluated as a diagnostic workflow: upload, preview, field mapping, validation, analysis, source-backed evidence, score, report, action queue, owner review, and score-history movement inside the Industrial IQ portal.

Best-fit buyer rolesCFO, CIO, Procurement, Maintenance, and Materials leaders
Trigger eventERP migration, AI readiness, working-capital pressure, procurement leakage, reliability risk, audit readiness, or recurring operational review.
Input data requiredDescription
Upload workflowUpload CSV, preview fields, confirm AI-recommended mappings, validate required coverage, normalize where needed, then run the engine.
Diagnostic logicFindings show source rows, match reasons, discriminator conflicts, review level, assumptions, and owner status before cleanup action.
Sample evidence tablePublic sample mode exposes mapped evidence rows before the customer uploads private data.
Score outputCatalog health score: lower values mean higher duplicate, field-quality, UOM, supplier-alias, and obsolete-record exposure.
Executive report previewSample report pages show the decision narrative, evidence rows, review levels, assumptions, limitations, PDF export, and next actions before private upload.
Report outputPartsCleanse AI Catalog Diagnostic Report with HTML, CSV evidence, PDF, and report email status for authenticated runs.
Governance controlsNo ERP write-back, no autonomous supplier outreach or system update, owner routing, audit metadata, and clear sample-versus-uploaded-data labeling.
Workflow review pack Upload, runbook, and sample-proof path for Catalog Intelligence

Open for the step-by-step workflow, diagnostic runbook, and public sample proof path.
Upload workflow

Upload -> Validate -> Analyze -> Evidence -> Score -> Report -> Action -> Repeat.

StepLayerCustomer experience
1UploadCSV export enters the parser. Source file retention rules are disclosed.
2MapERP/CMMS aliases are inferred, then corrected or confirmed by the user.
3ValidateRequired fields, completeness, missing values, and confidence reducers are shown before run.
4AnalyzeEngine-specific analyzers generate findings, evidence, and impact estimates.
5GovernFindings receive review levels and owner-decision status before any action.
6ReportExecutive report, evidence table, action tracker, and score snapshot are produced.
Engine diagnostic runbook

PartsCleanse AI operating path from uploaded data to reviewed action.

This runbook makes the engine functional for buyers before a pilot: what to upload, what must pass, what the engine analyzes, what evidence is produced, and what owner decision is required.

StepGateEngine artifactBuyer decision
1 Minimum source Material or item master CSV Start with Description. Best first run adds description, manufacturer, MPN, supplier, UOM, site, value.
2 Source-fit gate Confirm required fields, aliases, completeness, and weak mappings. Context fields such as Material Id, Manufacturer, Mpn, Supplier, Uom improve confidence and reduce assumptions.
3 Operational analysis path PartsCleanse AI Interchangeability classification: exact duplicate, probable duplicate, same OEM reference, substitute candidate, and unsafe match. Golden-record candidate generation with completeness, OEM/MPN, specification, value, and UOM reasons. Supplier alias, canonical UOM, OEM/part-number, taxonomy, and duplicate-prevention evidence.
4 Evidence output PartsCleanse AI Catalog Diagnostic Report Score, findings, evidence rows, confidence tiers, assumptions, limitations, action queue, and score-history snapshot.
5 Acceptance gate Human-reviewed diagnostic Owner accepts, rejects, defers, or requests more data before remediation, optimization, or system change.
6 Repeat path Recurring intelligence Rerun after review actions to compare score movement, open findings, and unresolved evidence.
Control boundary: diagnostic outputs are decision evidence. They do not create ERP write-back, automatic deletion, supplier outreach, stocking-rule changes, or unreviewed remediation.
Public proof before account creation

Customers can inspect samples before uploading their own data.

Every sample result is labeled as sample mode. It demonstrates mapping, validation, scoring, evidence, confidence, report structure, action logic, and recurring-review context without claiming customer-specific exposure.

Sample reportPDFCSVMappingDictionaryDiagnostic page

Input-file detail Field dictionary and diagnostic data pack for Catalog Intelligence

Open when the buyer needs column-level mapping, aliases, and source-context examples.
Input files and field dictionary

Accepted fields, aliases, and mapping expectations.

InputNeedCommon aliasesMeaning
Description Yes description; item_description; material_description; maktx; short_text; part_description; long_text; desc Item, part, asset, work-order, finding, or source-record description used by the engine.
Material Id Recommended material; material_id; material_number; matnr; item; item_number; item_id; sku; part; part_number; stock_code Unique material, SKU, item, or spare-part identifier from the source system.
Manufacturer Recommended manufacturer; mfr; maker; brand; oem; oem_name Manufacturer, OEM, brand, or maker name.
Mpn Recommended mpn; manufacturer_part_number; mfrpn; part_no; oem_part_number; vendor_part Manufacturer part number, OEM reference, supplier part number, or equivalent identifier.
Supplier Recommended supplier; vendor; vendor_name; lifnr; supplier_name Supplier, vendor, vendor alias, or purchasing source.
Uom Recommended uom; unit; unit_of_measure; meins; base_uom Unit of measure such as EA, Each, PCS, Meter, MM, Inch, Set, or Pack.
Quantity Recommended quantity; qty; stock_qty; on_hand; qty_on_hand; unrestricted; labst; stock_on_hand Quantity, balance, order quantity, stock quantity, or demand quantity depending on engine.
Unit Cost Recommended unit_cost; cost; price; moving_average_price; map; valuation_price; standard_price; unit_price Unit cost, average cost, standard price, last purchase price, or valuation rate.
Site Recommended site; plant; werks; location; storeroom; warehouse; depot; facility Plant, site, warehouse, storeroom, region, location, or operating unit.
Currency Recommended currency; currency_code; waers; iso_currency Currency code for cost, value, price, or exposure calculations.
Business impact model

Why this matters to the buyer committee.

Capital exposure

Capital exposure model

Duplicate-family value, confidence-adjusted exposure, recoverable working-capital range.

Operational risk

Operational risk model

False stockout signals, duplicate item creation, maintenance uncertainty, ERP migration blockers.

Diagnostic evidence

Diagnostic evidence model

Duplicate families, review levels, cleanup priority, clean CSV baseline, executive report.

Evidence detail Report evidence and buyer interpretation for Catalog Intelligence

Open when the buyer committee wants the output layers, confidence model, and role-by-role interpretation.
Evidence and confidence

What the report proves.

Output layerExampleWhy it matters
ScoreCatalog health score0-100 signal with risk level and trend-ready snapshot.
Score formulaDeterministic calculationThe report exposes the scoring formula and component inputs; random scores are not used.
Duplicate familyPart family evidenceDescription, manufacturer, MPN, UOM, size/material discriminator, and exposure.
FindingPartsCleanse AI Catalog Diagnostic ReportIssue title, severity, source engine, and owner-facing action.
EvidenceMapped source recordsSource-row references, relevant fields, analyzer reason codes, and review level.
Evidence graphSource -> finding -> evidence -> actionThe result carries an evidence graph for review, report, action, and score-history continuity.
ConfidenceHigh / Medium / Needs ReviewCoverage, completeness, source-field quality, and analyzer agreement.
ActionOwner review itemOwner action, priority, due window, and review status.
Renewal valueRecurring management viewThe report shows exposure identified, review queue size, actions created, and next review cadence.
Buyer interpretation

One diagnostic, multiple executive decisions.

BuyerDecision questionEvidence source
CFOduplicate capital exposure: fund cleanup sequencing before capital targets are committedCFO exposure view
COOfalse stockouts and planner searchability: protect operational availability while cleanup is plannedcatalog risk report
Procurementduplicate item creation and supplier confusion: standardize buying evidence before supplier actionprocurement review queue
Maintenancespare identification risk: improve searchability without unsafe consolidationmaintenance evidence table
ERP / data governancecleanup priority and migration readiness: sequence SAP, Maximo, or MDM remediation safelymaster-data backlog
Trust controls

This engine is governed before operational action.

Trust control

No ERP write-back

Findings are recommendations and review items, not autonomous ERP changes.

 Trust control

Evidence traceability

Every finding links back to source records, mapped fields, and reason codes.

 Trust control

Evidence confidence

High, medium, low, and needs-review logic is visible before action.

 Trust control

False-positive control

Unsafe lookalikes are separated from high-confidence findings.

 Trust control

Owner review

Owners can accept, reject, defer, assign, or request more data.

 Trust control

Audit trail

Report exports, reviewer actions, and status changes remain traceable.

FAQ

Questions buyers ask before running PartsCleanse AI.

What problem does PartsCleanse AI solve?

It diagnoses duplicate MRO records, weak descriptions, UOM conflicts, missing manufacturer fields, and cleanup priority so buyers can decide what to review before MDM, ERP cleanup, or catalog remediation.

What files should we upload first?

Start with a material or item master export. Manufacturer, MPN, supplier, UOM, plant, stock value, usage, and PO references improve confidence and business interpretation.

Does PartsCleanse AI require ERP integration?

No. The first diagnostic starts from exported CSV or workbook files and does not require SAP, Maximo, Oracle, IFS, EAM, or CMMS integration.

Does it write back to ERP or merge records automatically?

No. It produces source-backed findings and review queues only. ERP changes, record merges, and item retirement stay under buyer governance.

What output does the buyer receive?

A catalog health score, duplicate-family register, confidence-tiered evidence table, cleanup priority queue, executive report, and action tracker items.

Who should own the review?

ERP/data governance, materials management, procurement, maintenance, reliability, and finance should review together because interchangeability, value, and buying behavior must all be validated.

How is it different from MDM or a cleansing service?

MDM and services may execute stewardship and remediation. PartsCleanse AI runs first to prove which records matter, where confidence is strong, and what should be reviewed safely.

What is the safest first step?

Run a bounded catalog Snapshot with sample or exported item-master data, inspect the evidence table, then decide whether a paid diagnostic or remediation scope is justified.

Recommended next step

Move from product interest to buyer-ready evidence.

PartsCleanse AI should help a buying committee answer one practical question: what can your exported data prove, which findings are confidence-tiered, who must review them, and what action is safe after the report. Choose the next step based on buyer readiness, not a generic demo sequence.

Run Free Industrial IQ Snapshot View Sample Report See Required Data Request Diagnostic Pilot
Boundary: sample and benchmark outputs are not customer-specific findings. Uploaded-data diagnostics remain no-write-back, confidence-tiered, source-backed, and human-reviewed before action.
Buyer journey

Choose the next action by buyer readiness.

Choose the next action based on where the buying committee is now: proof review, data readiness, pilot scoping, or security validation.

Open evaluation guide
Stage: Need proof first View sample diagnostic reports

Inspect evidence rows, confidence tiers, limitations, scores, and owner actions before sharing private data.

View Sample Reports Stage: Have an export ready Run an Industrial IQ Snapshot

Start with exported operational data or sample data and route the issue to the right diagnostic engine.

Run Snapshot Stage: Need committee alignment Download the buyer evaluation guide

Use the finance, operations, technology, procurement, maintenance, and security checklist for internal review.

Download Buyer Guide Stage: Active initiative Request a founder-led pilot

Use this path when ERP migration, inventory action, procurement leakage, or AI readiness needs a scoped diagnostic.

Request Pilot Stage: Security review Review the security brief

Validate no ERP write-back, source-file purge, human review, access controls, DPA/SLA path, and retention boundaries.

Review Security Brief

Trust boundary: No ERP write-back. Source files purged after report generation. Human review before action. Sample reports use demonstration data until replaced by uploaded-data diagnostics.

Diagnostic evidence path

Choose the next step that matches your buying stage.

Industrial IQ is designed for evidence-first buyers. Review sample proof, run a bounded Snapshot, align the buyer committee, or request a founder-led diagnostic pilot when the operating problem is ready for review.

Read-only diagnostics · No ERP write-back · Source files purged after report generation · Human review before action
Still evaluating View Sample Reports

See what the report looks like before sharing internal data.

 Have a data export Run Catalog Intelligence

Start with an export-first diagnostic path and no ERP write-back.

 Need committee alignment Download Buyer Evaluation Guide

Give finance, operations, procurement, ERP, security, and maintenance the same evaluation frame.

 Ready for review Request Founder-Led Pilot

Ask for a founder-led pilot review when the problem has an owner and source data is available.

Product evaluation lens

Use this page to evaluate PartsCleanse AI as part of the 8-engine Industrial IQ platform.

Product evaluation separates the operating problem, required source files, diagnostic logic, evidence outputs, and review boundary so buyers understand what the engine does and what remains human-owned.

Best-fit reader

CFO, CIO, Procurement, Maintenance, and Materials leaders

Evidence to prepare

Source exports mapped to the engine, required fields, optional context, assumptions, exclusions, and confidence-tier evidence.

Output to expect

Engine-specific findings, score context, report sections, open findings, action-tracker items, and next review steps.

Trust boundary

Read-only diagnostics, no ERP write-back, source-file purge after report generation, and human review before action.

Compare all 8 enginesView sample reportsDownload data templatesRun a Snapshot
AI2COE Copilot