Fragmented Data Access Resolved with Integrated, Permissions-Aware Enterprise Search Across MES, ERP, QMS, CMMS

Overview

A diversified manufacturer struggled to find and use operational knowledge because data lived in many systems with uneven access controls. Engineers and supervisors waited on ad hoc extracts, and sensitive fields occasionally surfaced in shared spreadsheets. Intelligex implemented a permissions-aware enterprise search that indexed Manufacturing Execution System (MES), Enterprise Resource Planning (ERP), Quality Management System (QMS), and Computerized Maintenance Management System (CMMS) content, mirrored source permissions, and redacted sensitive fields. Teams found what they needed quickly without creating new copies, and governance strengthened with full auditability and role-based guardrails—without replacing core systems.

Client Profile

• Industry: Diversified manufacturing across discrete assembly and light process
• Company size (range): Multi-site, multi-division operations
• Stage: Mature MES/ERP/QMS/CMMS footprint; mixed file shares and SharePoint
• Department owner: Operations & Manufacturing
• Other stakeholders: IT/Security, Data Governance, Quality, Maintenance/Reliability, Supply Chain, Finance, Legal/Compliance, Site Leadership

The Challenge

Information needed to solve day-to-day problems was scattered. A line supervisor might need a recent nonconformance, a maintenance history on a critical asset, and the current routing from ERP—all in different systems with different logins and search behavior. People asked for extracts and screenshots, which spawned uncontrolled copies. Sensitive elements such as supplier pricing, bank details, personally identifiable information in complaints, or export-controlled drawings occasionally leaked into share drives and email threads. Data owners tightened access as a result, which slowed problem-solving further.

Existing search tools worked within each application but not across them. Attachments inside QMS records or maintenance history PDFs in CMMS were hard to find without exact IDs. MES held operator guides and work instructions in varied repositories. ERP stored specifications and vendor documents with different naming conventions by division. A typical investigation turned into a chain of emails and remote sessions just to locate the right record, while governance teams worried about the spread of untracked exports.

Security was non-negotiable. The company followed least-privilege principles and wanted to align with zero trust concepts; see NIST SP 800?207 for an overview. Any solution had to mirror existing permissions, avoid becoming a back door, and provide field-level masking for sensitive data. Teams asked for a single, governed way to search, preview, and request access, integrated with identity and audit trails.

Why It Was Happening

Root causes were fragmentation and inconsistent controls at the last mile. Each system implemented its own access model and search, so users either learned every interface or relied on someone else to retrieve what they needed. Attachments and scanned PDFs were not indexed uniformly, metadata was sparse or inconsistent, and synonyms across sites (asset names, defect terms, supplier aliases) made discovery hit-or-miss. When people exported extracts to accelerate their work, control and lineage were lost.

Ownership was diffuse. IT and Security owned identity and infrastructure, Operations owned content in MES and CMMS, Finance and Supply Chain owned ERP, and Quality owned QMS. There was no shared search catalog or redaction layer. As a result, teams either over-shared to be helpful or under-shared to be safe, and neither approach scaled.

The Solution

Intelligex implemented a permissions-aware enterprise search platform that crawled and indexed content across MES, ERP, QMS, CMMS, and file repositories, while mirroring source access controls. Identity integrated with single sign-on, search results respected role and site constraints, and sensitive fields were masked according to data classification rules. Users searched once, filtered by line, site, asset, or document type, and previewed content without creating new copies. Requests for elevated access followed a governed workflow. The approach aligned with zero trust principles and left systems of record unchanged.

• Integrations: Read-only connectors to MES (e.g., Siemens Opcenter, Rockwell FactoryTalk ProductionCentre, SAP ME), ERP (e.g., SAP, Oracle), QMS (e.g., ETQ Reliance, MasterControl, TrackWise), CMMS (e.g., IBM Maximo, Fiix), and content stores (SharePoint, network drives). Enterprise search stack reference: Elastic Enterprise Search.
• Identity and permissions: Mapped users and groups from Active Directory/Azure AD to source-system roles. Mirrored access control lists (ACLs) and site scoping. Enforced least privilege with attribute-based constraints where needed (site, product line, project).
• Sensitive data controls: Field- and pattern-level redaction for supplier banking, contract prices, personal data in complaints, and export-controlled identifiers. Masking applied in snippets and previews with unmask under approved roles.
• Indexing and relevance: Full-text and metadata indexing for records and attachments (PDFs, images with OCR). Synonym dictionaries for local naming and multilingual terms. Filters for asset, line, SKU, supplier, document type, and status.
• Search experience: Web and Teams add-in with quick filters and saved searches. Hover previews, governed downloads, and one-click deep links back to the source record in the system of record.
• Audit and governance: Immutable logs of queries, result views, and downloads. Data classification tags in the index. Access requests routed to data owners with e-signatures and reason codes.
• Security boundaries: Read-only connectors, network segmentation, and encryption in transit and at rest. No writes to source systems; edits occurred only in the system of record.
• Standards alignment: Design informed by zero trust principles per NIST SP 800?207 and information security management practices consistent with ISO/IEC 27001.

Implementation

• Discovery: Cataloged systems, repositories, and data owners. Mapped current access models, sensitive-field patterns, and common search tasks by role. Identified high-value use cases and risky export practices.
• Design: Defined the search schema and metadata, ACL mirroring, site and role scoping, and redaction patterns. Built synonym dictionaries and filters by domain (quality, maintenance, supply chain). Agreed on audit retention and access-request workflows.
• Build: Configured connectors to MES/ERP/QMS/CMMS and content stores. Implemented indexing with OCR for attachments, relevance tuning, and redaction. Integrated SSO and group mapping. Stood up the web UI and Teams add-in with deep links.
• Testing/QA: Ran in shadow mode with a pilot group. Verified that permissions mirrored source systems, redaction masked correctly, and relevance met operational needs. Exercised audit logs and access-request routing with data owners.
• Rollout: Phased by site and function. Started with Operations and Maintenance, then added Quality and Supply Chain. Kept legacy bookmarks and shared folders as a controlled fallback until teams adopted saved searches and deep links.
• Training/hand-off: Delivered short role-based sessions on searching, filtering, previews, and requesting access. Updated SOPs for data access and export. Transferred ownership of synonym lists, redaction patterns, and filter sets to Operations and IT Security under change control.

Results

Engineers and supervisors found answers without waiting on extracts. A single query returned relevant QMS deviations, MES context, maintenance history, and the current ERP spec with the right filters applied. Previews reduced the spread of exported copies, and deep links took users straight to the authoritative record for edits. Sensitive fields stayed masked unless the viewer’s role and site scope permitted access, and every view and download was auditable.

Governance improved in parallel. Data owners saw fewer one-off access requests because ACLs were mirrored consistently and access requests followed a clear workflow with e-signatures. Compliance teams pulled audit logs and sampling evidence without chasing people. Security gained confidence that the index did not become a back door, and operations moved faster with fewer misuses of data.

What Changed for the Team

• Before: Each team searched a different system or asked for extracts. After: A single, permissions-aware search spanned MES, ERP, QMS, CMMS, and file stores.
• Before: Sensitive fields leaked into spreadsheets. After: Field-level redaction and governed previews prevented inappropriate exposure.
• Before: Name and code differences hid relevant records. After: Synonym dictionaries and filters surfaced results across local naming conventions.
• Before: Access rules varied and were unclear. After: Source ACLs were mirrored, and access requests followed a routed workflow.
• Before: Audits required reconstructing who saw what. After: Query, view, and download events were logged immutably with user and context.
• Before: Bookmarks and attachments went stale. After: Deep links led to systems of record, reducing copies and rework.

Key Takeaways

• Make search permissions-aware; mirror source ACLs and scopes so discovery does not bypass controls.
• Redact at the field and pattern level; previews should mask sensitive elements while preserving utility.
• Index attachments and standardize metadata; OCR and synonyms turn buried content into usable knowledge.
• Keep edits in systems of record; search should point to the source, not create new copies.
• Log everything; searchable audit trails simplify compliance and build trust.
• Roll out in phases; start with high-value roles and refine relevance and redaction patterns before broad adoption.

FAQ

What tools did this integrate with? Connectors indexed content from MES platforms such as Siemens Opcenter, Rockwell FactoryTalk ProductionCentre, or SAP ME; ERP systems like SAP or Oracle; QMS tools such as ETQ Reliance, MasterControl, or TrackWise; CMMS like IBM Maximo or Fiix; and SharePoint and network drives. Identity and permissions mapped from Active Directory/Azure AD. The search stack followed enterprise patterns; for reference, see Elastic Enterprise Search.

How did you handle quality control and governance? Source ACLs were mirrored, and access was constrained by role, site, and project where applicable. Sensitive fields were masked using pattern and field-level redaction. All queries, result views, and downloads were logged. Access requests routed to data owners with e-signatures and reason codes. Design aligned with zero trust principles outlined in NIST SP 800?207 and information security management practices consistent with ISO/IEC 27001.

How did you roll this out without disruption? The index ran in shadow mode with a pilot group first. Permissions and redaction were validated against source systems, and relevance and filters were tuned with Operations and IT Security. Rollout proceeded by site and function, with legacy bookmarks and folder paths retained as a controlled fallback during early cycles.

How were sensitive fields redacted and who could unmask them? Redaction patterns covered items like supplier banking, unit prices, personal identifiers in complaints, and export-control markers. Masking applied in snippets and previews; unmask required roles with explicit permission and was recorded with reason codes. Full documents remained accessible only through the system of record under the same controls.

How were permissions mapped and kept current? Group and role mappings synchronized from Active Directory/Azure AD and the source applications on a schedule. When site moves or role changes occurred, access reflected those changes without manual reconfiguration. Periodic access reviews with data owners ensured mappings remained accurate, and changes followed documented change control.