Materials Identification, Classification and Master Data

Introduction

Inventory and materials management begins with a deceptively basic question: what exactly is the item being managed? If an organisation cannot identify, describe, classify and measure an item consistently, it cannot reliably purchase, store, issue, count, value or replenish that item. A single physical material may be described as “bolt,” “hex bolt,” “M10 bolt,” or “maintenance fastener.” These descriptions may refer to the same item—or to different items. Ambiguity creates duplicate stock records, incorrect purchases, excess inventory, stock-outs and safety risks.

Materials identification and master-data management provide the foundation for inventory control. They establish a common language between engineering, procurement, stores, production, maintenance, finance, quality, suppliers and information systems. The discipline combines technical description, classification, coding, cataloguing, standardisation, specification control and governance.

A material master is the structured organisational record that describes an item and supports its life cycle. It may contain the item code, short and long descriptions, technical attributes, units of measure, purchasing data, storage requirements, planning parameters, valuation data, quality requirements and ownership information.

Learning Objectives

By the end of this chapter, the learner should be able to:

1. Explain the purpose and principles of materials identification and cataloguing.
2. Distinguish between coding, classification, description, specification and standardisation.
3. Design practical item codes and technically sound descriptions.
4. Apply variety-reduction and standardisation techniques.
5. Control units of measure, conversions and pack-size information.
6. Define material-master ownership, approval and data-quality rules.
7. Evaluate the risks of poor master data and propose corrective action.
8. Apply these concepts to an operational case.

1. The Role of Identification in Inventory Management

A material record should answer five questions:

• What is the item? Its identity and technical characteristics.
• How is it distinguished? Its unique code and classification.
• How is it measured? Its stock, purchasing and issue units.
• Where and how is it used? Its functional, operational and safety context.
• Who controls it? The responsible owner and approval authority.

Identification is not merely labelling. It is the controlled process of establishing that an item is unique, distinguishable and consistently represented in business processes.

A useful distinction is:

The fundamental rule is one item, one approved identity, one active master record, unless separate records are justified by a genuine difference in function, ownership, valuation, planning or regulatory treatment.

2. Identification, Description and Specification

2.1 Generic and differentiating characteristics

A good description has two parts:

1. Generic noun: what the item is, such as valve, cable, bearing or solvent.
2. Differentiating attributes: characteristics that distinguish it from similar items, such as size, material, rating, grade, standard, connection or manufacturer reference.

For example, “valve” is inadequate. “Valve, ball, 50 mm, stainless steel, flanged, Class 150” is more useful. The exact format should follow the organisation’s naming standard.

Attributes must be relevant to identification. Adding every available detail can make records unwieldy and increase errors. Omitting a critical attribute makes records ambiguous. The objective is sufficient specificity, not maximal verbosity.

2.2 Description standards

A description standard should define:

• approved generic nouns;
• attribute sequence;
• abbreviations and prohibited abbreviations;
• spelling and singular/plural conventions;
• symbols, dimensions and tolerances;
• brand and manufacturer treatment;
• reference to technical standards;
• rules for hazardous, controlled or regulated materials.

Descriptions should normally be objective and independent of temporary location, supplier or buyer. “Pump for Line 3” is weak because the line may change. “Pump, centrifugal, 15 kW, 50 m³/h, 40 m head” identifies functional characteristics more effectively, provided the values are verified.

2.3 Specifications

A specification states the required characteristics, performance and acceptance conditions of an item. It may reference recognised standards such as ISO, IEC, ASTM or national standards, where applicable. Specifications may include:

• dimensions and tolerances;
• material and grade;
• performance rating;
• operating temperature and pressure;
• electrical characteristics;
• compatibility and service conditions;
• inspection and testing requirements;
• packaging, shelf-life and handling requirements.

A specification must not be copied casually from a supplier quotation. Supplier information may be incomplete, commercially biased or inconsistent with the organisation’s engineering requirement. Technical authorities should verify critical attributes.

3. Coding Systems

An item code is a unique identifier assigned to an approved item record. The code is not necessarily a complete description. Its primary purposes are uniqueness, system integration, transaction accuracy and efficient reference.

3.1 Types of coding

Significant or intelligent codes embed meaning. For example, PMP-CEN-015 might indicate a centrifugal pump of a particular size. They can assist recognition but become difficult to maintain when classifications change. Users may also infer meanings that are not formally controlled.

Non-significant or non-intelligent codes are sequential or randomly generated, such as 10045872. Meaning is held in the description and attributes. These codes are stable, scalable and less likely to create misleading assumptions.

Hybrid codes combine a neutral identifier with a controlled category prefix. They can be practical, but the prefix must not be allowed to carry too much technical meaning.

For most modern ERP environments, a neutral unique identifier supported by a strong description and classification is preferable. A code should be:

• unique;
• permanent or carefully retired rather than casually reused;
• sufficiently long for future growth;
• compatible with system limits;
• free from confusing characters where manual entry occurs;
• independent of supplier, warehouse and temporary location;
• governed by documented issuance rules.

3.2 Code assignment and retirement

A code should be assigned only after a duplicate search and technical review. The search should use exact descriptions, synonyms, manufacturer references, specifications and historical records. If an item is no longer used, the record should normally be inactive, not deleted, because historical transactions, audit trails and financial records may depend on it. Reuse of retired codes should be prohibited.

A barcode, QR code or RFID tag is not a substitute for a master-data code. It is a machine-readable representation or carrier of an identifier. The underlying identity and transaction controls remain essential.

4. Classification and Cataloguing

Classification groups similar items according to a controlled structure. A classification hierarchy may contain a segment, family, class and item. For example:

Maintenance materials → Mechanical → Bearings → Deep-groove ball bearings

Classification supports searching, reporting, spend analysis, inventory policy, technical review and variety reduction. It should reflect how the organisation uses and controls materials—not merely how a supplier markets them.

4.1 Classification principles

An effective classification system should be:

• mutually consistent, so similar items are grouped similarly;
• sufficiently comprehensive without creating unnecessary categories;
• understandable to users;
• stable over time;
• extensible for new technologies;
• linked to relevant attributes;
• capable of supporting reporting and procurement analysis.

A classification is not the same as a location. “Warehouse A” is a storage location, not an item class. Similarly, “urgent purchase” is a transaction condition, not a material type.

4.2 Catalogue structure

A catalogue normally includes a controlled list of generic terms, attributes, synonyms, manufacturer references and approved item records. Cataloguing activities include:

1. collecting source records from ERP, spreadsheets, drawings and suppliers;
2. cleansing spelling, units and abbreviations;
3. identifying duplicates and near-duplicates;
4. normalising descriptions;
5. assigning classes and attributes;
6. validating technical equivalence;
7. assigning or linking item codes;
8. publishing approved records;
9. monitoring use and updating controlled data.

Automated matching tools can identify likely duplicates using text similarity, manufacturer part numbers and attribute comparisons. However, similarity is not proof of equivalence. A technical expert must review safety-critical, regulated and performance-sensitive items.

5. Standardisation and Variety Reduction

Standardisation means adopting common materials, dimensions, grades, brands, processes or specifications where they satisfy the operational requirement. Variety reduction means eliminating unnecessary alternatives while preserving required performance, safety and resilience.

Examples include:

• reducing several similar bolt grades to approved standard grades;
• using common bearing sizes across equipment where engineering permits;
• consolidating packaging sizes;
• specifying standard cable types;
• approving preferred equivalent manufacturers;
• replacing obsolete or non-standard descriptions.

Variety reduction can lower purchase prices, simplify training, reduce storage space, improve interchangeability and increase demand aggregation. However, forced standardisation can create failure, warranty or safety risks. The correct test is not “Can the number of items be reduced?” but “Can the required function be achieved reliably with fewer controlled variants?”

A variety-reduction review should examine:

• functional equivalence;
• dimensional compatibility;
• operating environment;
• quality and regulatory requirements;
• total cost of ownership;
• supplier availability and resilience;
• maintenance and replacement implications;
• consequences of failure.

A preferred-item list may classify items as preferred, approved alternative, restricted, obsolete or prohibited. Such status must be visible to buyers, engineers and stores personnel.

6. Units of Measure and Conversion Control

Unit-of-measure errors are a major source of inventory inaccuracy. The material master should distinguish at least:

• base or stock unit: the unit in which on-hand inventory is controlled;
• purchase unit: the unit used on supplier orders;
• issue or consumption unit: the unit used by the user or production process;
• sales or planning unit: where relevant;
• packaging unit: the supplier’s physical package.

If a conversion is fixed and verified, the system may store it. For example, if one carton always contains 24 pieces:

If an order is for 5 cartons, the quantity in pieces is:

Here, Q_cartons is the ordered quantity in cartons and Q_pieces is the equivalent quantity in pieces. The conversion is valid only when the carton quantity is fixed and the item identity is unchanged.

Weight, length and volume conversions require greater care. A conversion between mass and volume depends on density and sometimes temperature. It must not be entered as a fixed factor unless the material and conditions justify it. Similarly, “drum,” “bag” and “box” are not meaningful stock units unless their contents are defined.

Unit governance should specify:

• an approved unit dictionary;
• decimal precision and rounding rules;
• conversion-factor ownership;
• treatment of variable-weight items;
• packaging and net/gross quantity fields;
• controls for unit changes after transactions exist.

Changing a base unit can affect stock balances, open orders, bills of material, costing and historical reporting. Such changes require controlled conversion and impact assessment.

7. Material-Master Data Architecture

A material master normally contains several groups of data:

Not every field should be visible or editable by every employee. Role-based access reduces accidental changes and separates duties. For example, engineering may approve technical attributes, while finance controls valuation fields.

Ownership and stewardship

Data ownership is accountability for meaning, policy and approval. Data stewardship is the day-to-day maintenance, quality checking and coordination. Data custodianship concerns system administration, access, backups and technical controls.

A practical responsibility matrix might assign:

• requester: provides the business need and evidence;
• technical authority: confirms specification and equivalence;
• master-data steward: checks completeness, uniqueness and format;
• procurement: validates supplier and purchasing information;
• stores: confirms storage and issue requirements;
• finance: validates valuation and accounting attributes;
• compliance: confirms regulatory controls;
• system custodian: implements approved changes.

No single department should be able to create an item, approve its technical equivalence and alter financial controls without review.

8. Data-Quality Governance

Master-data quality has several dimensions:

• Accuracy: the value reflects the real item.
• Completeness: required fields are populated.
• Consistency: the same conventions are used across records and systems.
• Uniqueness: duplicate records are avoided.
• Timeliness: changes are reflected when needed.
• Validity: values conform to permitted formats and ranges.
• Traceability: changes and approvals can be audited.

A simple completeness measure can be used for a defined set of mandatory fields:

This measure assumes that all mandatory fields have been clearly defined and that a populated value is not automatically a correct value. A record can have 100% completeness but still contain an incorrect pressure rating.

A duplicate rate may be expressed as:

The denominator and duplicate definition must be documented. Near-duplicates may require technical judgement rather than simple text matching.

Useful controls include:

• mandatory-field validation;
• controlled vocabularies and dropdown values;
• duplicate-search workflow;
• approval thresholds based on risk;
• change logs and version history;
• periodic exception reports;
• inactive-record reviews;
• supplier-reference validation;
• audit sampling;
• data-quality dashboards.

Data-quality governance is a continuous process, not a one-time cleansing project. New items, engineering changes, substitutions, obsolete materials and supplier changes continuously generate data-maintenance needs.

9. The Item-Creation Workflow

A controlled workflow commonly follows these steps:

1. Request: A user submits the business need, application, quantity and supporting documents.
2. Search: The requester or steward searches existing records, synonyms and approved alternatives.
3. Technical definition: The responsible expert confirms the required specification.
4. Classification: The item is assigned to the correct class and attribute template.
5. Description: The naming standard is applied.
6. Duplicate review: Possible matches are evaluated.
7. Cross-functional approval: Relevant technical, procurement, inventory, finance and compliance checks occur.
8. Code assignment: A unique code is issued by the authorised system or steward.
9. Publication: The record is released for relevant business views.
10. Monitoring: Usage, errors, substitutions and subsequent changes are reviewed.

The workflow should distinguish creation from change. Changing a description may be harmless in one case but may alter the identity of an item in another. If a technical change affects fit, form, function, safety, quality or interchangeability, a new item code may be required rather than an overwrite.

10. Applied Case: Duplicate Maintenance Bearings

The following is an illustrative case based on common industrial conditions.

Background

A food-processing company maintained six plants. Its ERP contained 18,400 maintenance-material records. A review found multiple records for bearings used in conveyors and pumps. Examples included “bearing 6205,” “BRG 6205 ZZ,” “6205 SKF,” and “ball bearing 25x52.”

Problem

The records appeared similar but were not automatically equivalent. Some had rubber seals, some metal shields, and some different clearance classes. Buyers selected different records for the same request, while stores held excess stock under multiple codes. The organisation experienced emergency purchases despite substantial total bearing inventory.

Analysis

A cross-functional team extracted descriptions, manufacturer references, technical attributes, equipment bills of material and issue history. It found:

• exact duplicates caused by spelling and abbreviation differences;
• genuine variants incorrectly treated as identical;
• manufacturer names embedded inconsistently in descriptions;
• missing seal and clearance attributes;
• inconsistent stock units between “each” and “box.”

The team classified the bearings by bearing type, bore, outside diameter, width, seal or shield, clearance and approved standard. Engineering reviewed interchangeability for each equipment application.

Approach and solution

The company adopted a neutral item code, a controlled description format and an attribute template. It created a preferred-item list, linked approved manufacturer references and marked obsolete records inactive. A box-to-piece conversion was stored only where packaging was fixed and verified. Critical equipment records were reviewed before substitution.

The master-data steward became responsible for duplicate screening, while engineering retained authority over technical equivalence. Procurement and stores received training on searching by attributes rather than by informal descriptions.

Outcome

The organisation reduced duplicate records, improved search accuracy and consolidated purchasing demand for genuinely equivalent bearings. It did not merge variants merely because their dimensions matched. This protected equipment reliability and demonstrated that variety reduction must follow technical validation.

Learning points

1. Similar text does not prove identical material identity.
2. Manufacturer references are useful search keys but should not replace technical descriptions.
3. Technical equivalence must be approved by a competent authority.
4. Unit-of-measure controls are part of item identity and transaction accuracy.
5. Inactive records preserve historical and audit information.

11. Technology, Automation and Emerging Practice

ERP, product-lifecycle-management, warehouse-management and procurement systems increasingly share material data through integration interfaces. Barcode scanning reduces manual entry, while RFID can support automated identification in suitable environments. Data-quality tools can use natural-language processing to suggest standard nouns, detect spelling variants and identify likely duplicate records.

Artificial intelligence can assist with classification and similarity matching, but it should operate under controlled human review. Risks include false equivalence, biased training data, unexplained recommendations and propagation of an incorrect source record. AI-generated descriptions must not be accepted automatically for safety-critical or regulated materials.

Digital thread and digital-twin initiatives also increase the importance of stable identifiers. A component may be linked to a drawing, asset, maintenance task, supplier certificate, inspection record and operating history. If identifiers are unstable or duplicated, these relationships become unreliable.

12. Governance, Ethics and Compliance

Master data can create safety, environmental and legal consequences. Incorrect chemical classification may expose workers or the environment to harm. An incorrect pressure or electrical rating may cause equipment failure. Misclassification can also produce incorrect tax, customs, financial or regulatory treatment.

Ethical governance requires:

• truthful technical descriptions;
• transparent approval decisions;
• no deliberate creation of duplicate records to bypass controls;
• protection of commercially sensitive supplier information;
• traceable changes;
• appropriate access to hazardous-material information;
• competence-based approval for critical specifications.

Data governance should comply with organisational policies and applicable product, safety, environmental, procurement and financial regulations. The precise legal requirements vary by jurisdiction and industry; therefore, this chapter provides governance principles rather than legal advice.

13. Common Failure Modes and Best Practices

Common failure modes

• allowing every user to create free-text items;
• using supplier names as the primary identity;
• embedding temporary location or project information in permanent codes;
• changing an existing record to represent a different technical item;
• treating packaging units as stock units without conversion control;
• deleting records with historical transactions;
• relying on an automated duplicate match without technical review;
• making all fields mandatory without distinguishing critical from optional data;
• allowing uncontrolled abbreviations;
• failing to assign accountable owners.

Best practices

1. Establish a clear naming and attribute standard.
2. Use one controlled item-creation workflow.
3. Separate neutral identity from descriptive meaning.
4. Maintain a searchable synonym and manufacturer-reference index.
5. Use technical templates by material class.
6. Define owners and approval rights through a responsibility matrix.
7. Control units and conversions centrally.
8. Preserve audit history and retire rather than delete records.
9. Measure quality using accuracy, completeness, uniqueness and timeliness indicators.
10. Review high-risk and high-value materials more rigorously than routine consumables.
11. Link material records to assets, drawings, bills of material and approved suppliers.
12. Treat data maintenance as a permanent operational capability.

Chapter Summary

Materials identification and master data establish the common language required for reliable inventory and materials management. A valid item record combines a unique identity, controlled code, adequate description, relevant specification, appropriate classification and correct units of measure. Coding should support uniqueness and system stability; classification and cataloguing should support search, analysis and procurement; standardisation and variety reduction should eliminate unnecessary alternatives without compromising technical performance or safety.

A material master is cross-functional. Engineering validates technical characteristics, procurement manages purchasing information, stores controls handling and inventory implications, finance governs accounting fields, compliance manages regulated information and master-data stewards maintain quality and workflow. Governance must address accuracy, completeness, consistency, uniqueness, validity, timeliness and traceability.

Technology can improve search, scanning, integration and duplicate detection, but automated recommendations require human accountability, especially for safety-critical items. The central professional principle is that master data is not administrative paperwork: it is operational infrastructure that directly affects cost, availability, quality, safety and decision-making.

Key Terms

Material master: Controlled electronic record containing information about an item used in organisational processes.

Item code: Unique identifier assigned to a managed material or item.

Identification: Process of establishing and distinguishing an item’s identity.

Description standard: Rules governing naming, attributes, abbreviations and formatting.

Specification: Defined technical, performance, quality or acceptance requirements.

Classification: Structured grouping of items with shared characteristics.

Catalogue: Organised collection of approved item records and supporting references.

Intelligent code: Code whose characters carry defined meaning.

Non-intelligent code: Identifier whose characters do not intentionally describe the item.

Standardisation: Adoption of common approved materials, dimensions, grades or specifications.

Variety reduction: Reduction of unnecessary material alternatives while preserving required function.

Base unit of measure: Unit in which inventory quantity is normally controlled.

Conversion factor: Approved relationship between two units of measure.

Data owner: Person or function accountable for the meaning and approval of data.

Data steward: Person responsible for operational maintenance and quality coordination.

Data custodian: Function responsible for technical system administration and protection.

Duplicate record: Separate record representing the same item or an unjustifiably equivalent item.

Preferred item: Approved material selected for normal use where it satisfies requirements.

Traceability: Ability to identify the history, source and approval of a data value or change.

Material equivalence: Demonstrated technical suitability of one item as a substitute for another.

Review Questions

Short Answer Questions

1. Why is a clear material description important for inventory control?
2. Distinguish between an item code and an item description.
3. State two advantages and one risk of variety reduction.
4. What is the difference between a base unit and a purchase unit?
5. Name four dimensions of master-data quality.

Long Answer Questions

1. Explain the principles that should guide the design of a materials-coding system.
2. Describe a complete item-creation workflow, including the roles of technical and administrative stakeholders.
3. Discuss how classification and cataloguing support procurement, stores and inventory analysis.
4. Explain why technical equivalence cannot be established solely through text similarity or matching dimensions.
5. Evaluate the responsibilities of data owners, data stewards and system custodians.

Analytical / Application Questions

1. An organisation has records named “M10 bolt,” “bolt M10,” “hex bolt 10 mm” and “M10 stainless bolt.” Design a review process to determine which records can be merged and which must remain separate.
2. A supplier sells a material in cartons of 24 pieces, but stores issues it individually. Recommend the master-data and transaction controls required to prevent quantity errors.
3. A maintenance department wants to reduce 12 bearing types to four preferred types. Develop an analysis framework that protects equipment reliability while assessing the potential benefits of standardisation.

Multiple Choice Questions with Answers

Question 1

What is the primary purpose of a material master record?

A. To replace all technical drawings

B. To provide a controlled record supporting material-related business processes

C. To record only supplier prices

D. To identify warehouse employees

Correct Answer: B. To provide a controlled record supporting material-related business processes

Question 2

Which description is most technically useful?

A. Valve

B. Important valve

C. Valve, ball, 50 mm, stainless steel, flanged, Class 150

D. Valve for Line 3

Correct Answer: C. Valve, ball, 50 mm, stainless steel, flanged, Class 150

Question 3

Which characteristic is generally preferred for a modern ERP item code?

A. It must contain the supplier name

B. It should be unique and stable

C. It should change whenever the item moves location

D. It should contain every technical attribute

Correct Answer: B. It should be unique and stable

Question 4

Why should a retired item code normally not be deleted?

A. Deletion increases warehouse space

B. Historical transactions and audit trails may depend on it

C. Deleted codes automatically become cheaper

D. Deletion improves technical equivalence

Correct Answer: B. Historical transactions and audit trails may depend on it

Question 5

Which activity is part of classification and cataloguing?

A. Assigning temporary storage staff

B. Normalising descriptions and assigning attribute classes

C. Changing supplier invoices without approval

D. Removing all technical standards

Correct Answer: B. Normalising descriptions and assigning attribute classes

Question 6

One carton contains 24 pieces. How many pieces are contained in 5 cartons?

A. 29 pieces

B. 48 pieces

C. 120 pieces

D. 240 pieces

Correct Answer: C. 120 pieces

Question 7

Which statement about variety reduction is correct?

A. All similar-looking items should always be merged

B. Variety reduction should preserve required function, safety and performance

C. Variety reduction applies only to office supplies

D. Variety reduction eliminates the need for technical approval

Correct Answer: B. Variety reduction should preserve required function, safety and performance

Question 8

Who should normally approve technical equivalence for a safety-critical component?

A. Any warehouse clerk

B. A competent technical authority

C. The first available supplier

D. The finance cashier

Correct Answer: B. A competent technical authority

Question 9

Which data-quality dimension asks whether a record represents the real item correctly?

A. Accuracy

B. Completeness

C. Timeliness

D. Availability

Correct Answer: A. Accuracy

Question 10

Why should AI-generated duplicate suggestions receive human review?

A. AI cannot process any text

B. Text similarity does not prove technical equivalence

C. Human review prevents all inventory transactions

D. AI is used only for financial data

Correct Answer: B. Text similarity does not prove technical equivalence