00 / INTRO

CIM · A3 Research Project · 2026 S1

A 60-year arc from punched tape to digital twins.

Computer Integrated Manufacturing is the use of computers to control the entire manufacturing process — from design through production to logistics. In the next ten minutes we trace its history, examine where the field stands today, and look at where it's going.

Presented by [GROUP NAMES] · CIM 2026 S1

Scroll

01 / HISTORY

Wires, paper tape, and the long road to integration.

The CIM vision predates the technology to deliver it by almost two decades. Here's how we got here.

1952

First numerically-controlled mill

MIT demonstrates a milling machine driven by punched tape — the first time a computer-readable instruction set drove a machine tool.

1960s

MRP emerges

Materials Requirements Planning gives manufacturers the first computer-managed view of inventory, demand, and bill-of-materials.

1973

Harrington names the field

Joseph Harrington publishes Computer Integrated Manufacturing, defining the integration thesis that drives the next 50 years of work.

1980s

First-wave CIM falters

MRP II and early CIM systems promise the integrated factory but mostly fail in practice — brittle data exchange and proprietary formats kill the integration story.

1990s

ERP, CAPP, group technology

Enterprise Resource Planning consolidates business systems. Computer-Aided Process Planning and group technology bring structure to manufacturing rules.

2000s

STEP makes data exchange real

ISO 10303 (STEP) lets CAD, CAM, and CAE systems finally exchange geometry and metadata losslessly — the missing piece from the 1980s.

The 80s vision of CIM was right. The technology to make it work shipped twenty years later.

02 / CURRENT STATUS

Industry 4.0 is what CIM was always trying to be.

The integration the 1980s promised is finally shipping — under a new name, on top of a stack of cloud, IoT, and digital-twin infrastructure.

Digital twins

Live virtual replicas of physical assets, driven by IoT telemetry. BMW Regensburg runs its plant on NVIDIA Omniverse; IBM and Siemens ship twin platforms across industry.

Standards stack

ISA-95 governs enterprise-to-control integration. ISO 73933 defines smart manufacturing reference architectures. STEP keeps moving design data.

Connected factory

Private 5G, edge compute, and dense sensor networks (Ericsson, Siemens) make the shop floor a real-time data source — not a quarterly report.

Lighthouse Network

The World Economic Forum's Global Lighthouse Network indexes factories operating at the Industry 4.0 frontier — 90+ sites as of 2025.

Case study · BMW Regensburg

BMW Group runs a full plant-scale digital twin in NVIDIA Omniverse, used for production planning, layout simulation, and operator training before any physical change is made.

03 / FUTURE

AI-native factories, closed-loop PLM, and the human in the loop.

The next decade isn't about more automation — it's about faster loops between every stage.

AI-driven CAPP

Process plans generated and re-optimised in real time from CAD geometry plus live shop-floor data, replacing static rule-based planning.

Generative design loops

CAD ↔ simulation ↔ manufacturability feedback collapses from weeks to minutes — designers iterate against producibility constraints in-session.

Closed-loop PLM

Sustainability, end-of-life, and field telemetry feed back into design decisions — a priority of EU industrial research and emerging product passports.

Human-in-the-loop

Pure lights-out manufacturing stays domain-specific. The realistic frontier is human operators with AI copilots, not empty factories.

The competitive edge is loop speed, not headcount.

04 / CLOSE

From vision to reality, in three layers.

History

60 years of failed integration attempts taught us what doesn't work.

Current

Industry 4.0 finally delivers the integration the 80s promised.

Future

The edge is loop speed, not headcount.

Happy to take questions.

References (15)