Skip to content
LeanSuite - Lean Manufacturing Software

Book summary · Operations management · Advanced

Factory Physics: summary and review

By Wallace J. Hopp and Mark L. Spearman. First published 1996.

The short answer

Factory Physics is a university textbook that treats a plant as a system with measurable behavior. Its central argument is that output, work in process and lead time are tied together by relationships that hold whatever you call your improvement method, and that variability always gets paid for with inventory, time or spare capacity. The authors turn this into formulas an engineer can run before touching the floor. It is the best book for learning why lean rules of thumb work and where they fail. It is heavy going, so read it with a pencil and your own line data.

Read it if

  • You are an engineer or planner who must defend a WIP cap, buffer size or lot size with numbers.
  • Your lean effort cut inventory and service then got worse, and you want to know why.
  • You plan production and want to see what MRP, kanban and CONWIP each cost.

Skip it if

  • You want a first book on lean. This one assumes you know the tools already.
  • You will not collect process-time and arrival data. The formulas are only as good as those inputs.
  • You need something a leadership team will finish. Its authors' shorter book for managers suits that better.

When it pays off: After several kaizen events and inventory cuts in at least one line, when someone asks how low WIP can safely go or how big a buffer should be. It also pays off before you size a new line or set a service-level target.

The core ideas, in plain language

  1. 1

    Little's Law checks your own numbers

    In a stable line, average WIP equals throughput times average cycle time. Know two and you can test the third, which quickly shows whether a quoted lead time or a WIP count can be true. Chen and Ye call it the one result in the book with a broad, well-studied proof.

  2. 2

    Variability is always buffered

    Spearman's framing: when demand and production fall out of step because of variability, the system carries a buffer of inventory, time or capacity. You choose which, never none. In his example, a build-to-order company with almost no inventory or time buffer carries about 20 percent spare capacity.

  3. 3

    Utilization has a steeply rising price

    Average queue time at a station is roughly a variability factor times a utilization factor times process time. The utilization factor is u divided by (1 minus u), so moving a station from 85 to 95 percent load more than triples it. Cutting variability usually buys more than squeezing utilization.

  4. 4

    Pull means a cap on WIP

    In a 2004 journal paper on pull, Hopp and Spearman define pull as explicitly limiting WIP, and say make-to-order versus make-to-stock is a separate question. CONWIP applies the cap with one card count per line.

  5. 5

    Lean as the cheapest way to buffer variability

    The authors read lean as production with minimal buffering cost. Removing sources of variability lowers the buffer you need; cutting buffers without that just moves the pain. Spearman describes a client whose line starved after WIP cuts; output rose over 20 percent when WIP went back up.

How to use it on Monday morning

  1. 1

    Test Little's Law on one line

    Pick one routing. At the same time each shift for five shifts, count pieces at stations and in queues, and note pieces finished per hour. Divide average WIP by hourly throughput to get implied cycle time and compare it with the schedule or ERP. A large gap means the system data is wrong or WIP is hiding somewhere, such as a rework rack.

  2. 2

    Find the real pacing station

    For each station, work out effective capacity from cycle time, parallel units, availability and yield. The lowest paces the line. Compare its load with takt and look at the queue in front of it; a heavily loaded station with a long queue is what the VUT relation predicts.

  3. 3

    Measure spread, not just averages

    At that station, record arrival and process times for about 30 consecutive jobs, noting stoppages, changeovers and waits for material as they happen. You want a ranked list of what makes the station irregular, not another average.

  4. 4

    Name the buffer you carry at each stage

    Walk the line with the supervisor and write down, for each stage, whether pieces, waiting time or idle capacity protects it. Mark where one buffer was cut without adjusting the others. Bring that page to planning before anyone changes a WIP limit or lot size.

What to be careful about

  • It is a textbook and reads like one

    The book runs 720 pages and builds its argument with queueing and probability models. For leaders, the authors with Edward Pound and Jeffrey Bell wrote Factory Physics for Managers (McGraw-Hill, 2014), which the publisher pitches as a route through lean, Six Sigma and constraint theory.

  • The laws are narrower than they sound

    Chen and Ye (Journal of the Operations Research Society of China, 2015) tested the variability, utilization, rework, lead time and CONWIP efficiency laws beyond simple serial lines. They found sufficient conditions for some laws and counterexamples for others, and urge caution for non-Markov, non-tandem systems such as wafer fabs. They do not frame this as criticism. For re-entrant routings or heavy batching, treat the formulas as direction, not forecasts.

  • It needs data many plants lack

    The variability factor depends on the spread of arrival and process times at each station. Without them you get the shape of the problem, not a number. A week of paper collection is feasible, but expect rough first estimates.

  • Lean readers sometimes take it as an attack

    Spearman argues against dogmatic WIP cutting but credits Toyota with buffering on purpose: two 12-hour shifts scheduling 10 hours of work, a 20 percent capacity buffer. That comes from his 2007 interview and we could not confirm it independently. His client examples are also his own accounts.

Factory Physics compared with The Goal

The Goal teaches one idea, the constraint, through a novel you can finish in a weekend. Factory Physics supplies the relationships among WIP, throughput, cycle time and variability that a story cannot. Read The Goal first; pick Factory Physics when you must size a buffer or defend a WIP number to a skeptical planner.

Read the summary of The Goal.

Tools and templates on LeanSuite that go with it

Book details

Bibliographic details of Factory Physics
TitleFactory Physics
AuthorWallace J. Hopp and Mark L. Spearman
First published1996. First edition published by Irwin (Chicago) as Factory Physics: Foundations of Manufacturing Management; second edition 2000; third edition 2008
Edition shownThird edition, hardcover (copyright 2008, Waveland printing 2011), Waveland Press, 2011
ISBN-13978-1577667391
Pages720

Details checked against public records on 7 October 2026. Other editions have other ISBNs. We carry no affiliate links.

Key terms

Lean glossary

FAQ

Questions about Factory Physics

More lean books

Back to the lean books hub for reading paths and filters.

Rather see it on your own problem?

Bring the problem you are stuck on, and we will show you on a call how LeanSuite handles it.

Schedule Demo

Pass it on

We want LeanSuite to be the best place on the internet for lean help. If this was useful, send it to someone on your team or in your network who needs it.