Theory of Constraints

Eliyahu M. Goldratt published The Goal in 1985, a business novel that became an unexpected success. Hundreds of plant managers replicated the actions of its protagonist, Alex Rogo, and achieved dramatic results. Yet two obstacles emerged: Improvements would plateau because companies lacked a process of ongoing improvement, and the book provided specific solutions rather than the thinking processes managers need to generate solutions independently. Theory of Constraints (TOC) addresses both gaps, asking: What thinking processes enable people to invent simple solutions to seemingly complicated situations? And how can organizational psychology assist the implementation of those solutions as an ongoing process?

Goldratt opens Part One by arguing that people grossly underestimate their own intuition; as long as it remains unverbalized, they act contrary to what they already know. Every system exists for a purpose, and every action should be judged by its impact on the system's global goal. A system's constraint is anything that limits it from achieving higher performance relative to that goal. Goldratt presents five focusing steps: (1) Identify the system's constraints; (2) Decide how to exploit them; (3) Subordinate everything else to that decision; (4) Elevate the system's constraints; (5) If a constraint has been broken, return to step one, but do not allow inertia to become a system constraint. The fifth step is critical: When a constraint is broken, policies derived from it persist even though their reasons have disappeared. Most companies, Goldratt asserts, are limited by policy constraints rather than physical ones. He reframes the five steps as three questions: (1) What to change? (2) To what to change to? (3) How to cause the change?

The third question is fundamentally psychological. Any improvement is a change; any change is a perceived threat to security; and any such threat produces emotional resistance that can only be overcome by a stronger emotion. Creating fear through escalating market demands is self-defeating because it requires constant insecurity and fades over time. Goldratt identifies a more powerful alternative: the "emotion of the inventor," the ownership a person feels when figuring something out independently. He uses The Goal as evidence: Readers responded with enthusiasm, typically inventing solutions before reading them and sharing the book widely. Such engagement demonstrates the effectiveness of the Socratic method, in which a teacher guides through questions rather than supplying answers. In the novel, Jonah, Alex's external mentor, refuses to give answers because doing so would block Alex from inventing them.

A practical obstacle remains: proving to an audience that a problem is theirs, since people naturally deflect responsibility. Goldratt introduces the Effect-Cause-Effect method, which relies on a situation's intrinsic logic rather than examples. He traces three stages every science passes through: classification, where data is organized into categories; correlation, where patterns enable prediction without asking "why"; and effect-cause-effect, where a cause is hypothesized and its predicted effects are verified. He maps these stages onto manufacturing: Manufacturing Resource Planning (MRP) represents classification; the Toyota Production System, Toyota's method for controlling production flow, and Just-In-Time (JIT), a production approach that makes goods only as needed, represent correlation; and The Goal represents effect-cause-effect, deriving conclusions from measuring goals through Throughput, Inventory, and Operating Expense. Goldratt warns that correlations without causal understanding mislead, citing the late-1970s belief that low Japanese inventory drove superior performance. This belief triggered a Western inventory-reduction race, though work-in-process and finished-goods reductions, not raw material cuts, were responsible.

Part One's final chapter presents the Evaporating Clouds method for constructing non-compromise solutions. Goldratt argues that whenever a compromise seems necessary, a non-compromise solution always exists. Precise problem definition requires stating the desired objective, identifying two requirements that must be satisfied, and locating the conflicting prerequisites. Rather than optimizing a compromise, the method challenges the hidden assumptions behind each logical connection. Through the Economic Batch Quantity problem, which generated thousands of articles over fifty years, Goldratt demonstrates three challenges: testing local objectives against global goals, questioning whether setup costs are truly fixed (the assumption JIT attacks through setup reductions), and exposing that "batch" conflates two meanings, the transfer batch (units moved between resources) and the process batch (units processed before resetting). The chapter includes a dialogue from the Theory of Constraints Journal in which a consultant reveals that a company's actual lead times have dropped far below its quoted times, creating a vicious cycle of client changes and firefighting.

Part Two argues that the arrangement in The Goal, where Alex and Jonah are separate people, guarantees long-term stagnation: As a plant becomes unique, an external consultant's general intuition grows inadequate, and managers must learn to verbalize their own intuition. Goldratt describes the Jonah Course, a ten-day program that teaches Effect-Cause-Effect and Evaporating Clouds through simulations and requires written implementation plans, without which organizations drift into stagnation. External consultants should scrutinize plans, not devise them, because they lack the organization's inertia.

Goldratt analyzes organizational psychology's impact on improvement. If even one function stagnates while others improve, it blocks Throughput for the entire organization, forcing all pressure into cost reduction and layoffs. Functions that improved most appear to have excess manpower and lose people first. In hierarchical organizations, functional heads competing for promotion resist improvements that disproportionately benefit a rival's metrics. The lesson: All functions must agree on a common path before any demonstrates individual results. To build consensus, Goldratt describes the Executive Decision Making Workshop, a two-day session that produces genuine group agreement. He defines the minimum viable unit as a "division" containing all functions and argues the division head and comptroller (head of finance) should be the first Jonahs. Education must eventually reach the lowest-level managers who feel total responsibility for their areas. For non-divisional managers, Goldratt advises finding an ally in a different function and jointly approaching the division head, since cross-functional agreement commands attention.

Chapter 5 examines the relationship among TOC, JIT, and Total Quality Management (TQM). All three attack the same erroneous assumption: the conventional ranking that places Operating Expense first. All three replace it with Throughput first, Inventory second, and Operating Expense third. TQM uses Statistical Process Control (SPC), a method of reducing variation through process measurement. JIT reduces fluctuations through setup reduction and flow-lines. TOC reduces both through predetermined finite schedules. The deeper realization is that most inventory exists because of erroneous policies, not genuine need, driving a paradigm shift from a "cost world" to a "Throughput world," where a tiny fraction of variables, the constraints, determines almost all results. Goldratt contends that JIT and TQM opened the door to this world but stopped at its border, while only TOC provides the map. He closes by warning that powerful solutions change the environment and can obsolete themselves. A final section describes the Jonah Line, an ongoing support structure of consultant visits, knowledge updates, and semi-annual conferences.

An appendix reprints two excerpts from The Goal. In the first, Alex meets Jonah at O'Hare airport; through questioning, Jonah demonstrates that a claimed 36% productivity improvement from robots reflects no real gain. In the second, Alex leads a Boy Scout hike that becomes an analogy for dependent events and statistical fluctuations. A slow hiker named Herbie governs the troop's throughput, and a dice-and-matchstick experiment confirms that in balanced systems with dependent events, throughput declines while inventory rises. Alex places Herbie at the front and lightens his pack, doubling speed while keeping everyone together, illustrating the focusing steps.