shigeo-shingo's detailed account of smed (Single-Minute Exchange of Die) — his systematic methodology for reducing machine changeover times. Productivity Press, 1985.
Overview
The book documents 19 years of development, from Shingo's first insight in 1950 (at Toyo Kogyo's Mazda plant) through refined implementations across dozens of companies. The three-stage SMED methodology is presented with extensive case studies:
1. Separate internal setup (machine stopped) from external setup (machine running) 2. Convert internal setup operations to external ones 3. Streamline all remaining operations
Key Arguments
Shingo's most dramatic example: reducing a 1,000-ton press changeover at toyota-motor-corporation from 4 hours to 3 minutes. The economics are transformative — SMED is the enabler of small-batch production. When changeover is expensive, economic logic demands large batches to amortize the cost; when changeover costs approach zero, one-piece-flow becomes possible.
Shingo traces the 19-year development: the initial insight at Toyo Kogyo (1950), the conceptual breakthrough at Mitsubishi Heavy Industries Hiroshima (1957, distinguishing internal vs. external setup), and the achievement of single-minute changeover at Toyota (1969). This long development arc illustrates that SMED was not a single invention but a gradually refined methodology.
Significance
SMED connects directly to continuous deployment in software development — the same logic applies: when "deployment" (changeover) is painful and slow, teams batch up changes; when deployment is automated and fast, continuous delivery becomes possible. This is the manufacturing insight that the-machine-that-changed-the-world and later lean software thinkers would translate into the digital domain.