I have been in a leadership position regarding the evolution of manufacturing plant information generation and management since the first concepts of a manufacturing execution system and real time data management were developed. I sometimes wonder how we got to where we are and how the modern pieces fit together in an AI world and I thought you might be interested. Most of this material was provided by a simple search on ChatGPT. Michael McClellan
Below is a concise, manufacturing-centric historical outline that traces how information systems evolved from basic data capture to today’s Digital Thread paradigm. The emphasis is on why each layer emerged, what information it managed, and how it changed manufacturing decision-making.
1. Pre-Computer & Early Computer Era (1950s–1960s)
Manual Data Gathering & Accounting Systems
Primary focus: Cost control and inventory accounting
- Paper travelers, route cards, time clocks, and manual inspection logs
- Batch reporting; data was historical and lagging
- Early mainframes used for payroll and inventory valuation
Key characteristics
- Information captured after the fact
- No operational feedback loop
- Manufacturing knowledge resided in people, not systems
Impact
- Enabled basic financial control
- Little influence on shop-floor execution or engineering decisions
2. MRP Era (1970s)
Material Requirements Planning
Primary focus: Planning material availability from demand
- Explosion of Bills of Material (BOMs)
- Time-phased planning (lead times, lot sizes)
- Driven by forecasts and MPS (Master Production Schedule)
Systems
- Standalone MRP systems on mainframes
Limitations
- Assumed infinite capacity
- No real-time feedback from manufacturing
- Engineering changes were slow to propagate
Impact
- First structured linkage between product structure and manufacturing supply
- Data still siloed and plan-centric
3. MRP II Era (1980s)
Manufacturing Resource Planning
Primary focus: Synchronizing materials, capacity, and cost
- Added capacity planning, shop-floor control, and costing
- Introduced “closed-loop” planning (feedback to plans)
Systems
- MRP II suites integrating planning and execution data
Limitations
- Still batch-oriented
- Weak engineering integration
- Manufacturing data primarily summarized, not contextual
Impact
- Manufacturing recognized as a system, not a department
- First operational feedback loops
4. ERP Era (1990s)
Enterprise Resource Planning
Primary focus: Enterprise-wide transaction integration
- Unified finance, HR, procurement, inventory, manufacturing
- Standardized data models and processes
Systems
- SAP, Oracle, Baan, JD Edwards
Limitations
- Manufacturing execution abstracted into transactions
- Poor representation of product intent and process detail
- Engineering and manufacturing largely decoupled
Impact
- Financial and supply-chain visibility improved
- Manufacturing became a transactional node, not a knowledge source
5. PLM & Engineering Data Management (1990s–2000s)
Product Lifecycle Management
Primary focus: Managing product definition and change
- CAD, PDM, configuration management
- Engineering BOM (EBOM) ownership
- Formal change control (ECOs)
Systems
- Dassault ENOVIA, Siemens Teamcenter, PTC Windchill
Limitations
- Weak manufacturing process context
- Limited operational feedback
- Manufacturing BOM (MBOM) often disconnected
Impact
- Digital authority for product definition
- Engineering rigor improved, but downstream visibility limited
6. MES & MOM Era (2000s–2010s)
Manufacturing Execution & Operations Management
Primary focus: Real-time execution and control
- Work instructions, dispatching, quality, genealogy
- Traceability and compliance
- Connection to automation (PLC, SCADA)
Systems
- MES, MOM platforms
Limitations
- Often isolated from PLM and ERP
- Contextual data trapped at plant level
- Limited lifecycle continuity
Impact
- Manufacturing becomes data-rich
- Shop-floor knowledge digitized but not fully leveraged
7. IIoT & Analytics Era (2010s)
Connected Assets & Advanced Analytics
Primary focus: Machine-level data and performance insight
- Sensor data, condition monitoring, OEE
- Time-series data at scale
- Cloud platforms emerge
Systems
- IIoT platforms, data lakes
Limitations
- Data volume without product context
- Difficult to relate asset behavior to product design or requirements
Impact
- Shift from descriptive to predictive analytics
- Highlighted the need for contextual integration
8. Digital Thread Era (Late 2010s–Present)
End-to-End Product & Manufacturing Continuity
Primary focus: Lifecycle coherence and decision traceability
- Persistent digital linkage from:
- Requirements → Design → Manufacturing → Operation → Service
- Bi-directional information flow
- Context preserved across domains
Key Concepts
- Authoritative data sources by domain
- Semantic models and digital twins
- Closed-loop learning (design informed by manufacturing and field data)
Systems Involved
- PLM, MES/MOM, ERP, IIoT, analytics platforms
- Integration via APIs, event streams, and data models
Impact
- Manufacturing becomes a knowledge generator
- Decisions are traceable, explainable, and improvable
- Foundation for AI-driven engineering and operations
Summary View (Conceptual Progression)
| Era | Primary Question Answered |
| Manual | “What happened?” |
| MRP | “What should we buy?” |
| MRP II | “Can we make it?” |
| ERP | “What did it cost?” |
| PLM | “What is the product?” |
| MES | “How is it being built?” |
| IIoT | “How is it performing?” |
| Digital Thread | “Why did this happen—and how do we improve it?” |
Key Insight
The Digital Thread is not a single system. It is the culmination of decades of incremental information capture, progressively shifting manufacturing from transactional reporting to contextual, lifecycle intelligence.