Goldratt applies his theory of constraints to project management, written as a business novel set in a university business school where an academic and his executive students work out a new approach.
The diagnosis is about how estimates and buffers behave. Individual task estimates include substantial safety margin because estimators are held accountable for their own task. That safety is then wasted through three mechanisms: student syndrome, where work starts at the last possible moment because the buffer is known to exist; Parkinson's law, where work expands to fill the time available and early completion is not reported; and multitasking, where a person switching between several projects makes every one of them slower. Because safety is wasted at every task, the accumulated protection disappears while individual tasks still overrun.
His solution is to cut individual task estimates to roughly half, remove all local safety, and aggregate the removed time into buffers placed at the end of the critical chain and at the points where feeding chains join it. The critical chain differs from the critical path by accounting for resource contention as well as task dependency. Project status is then managed by monitoring buffer consumption relative to chain progress, rather than by tracking whether individual tasks met their dates.
The method has been adopted in some project organisations with reported success, and the evidence is largely practitioner reports rather than controlled comparison. The novel format is thin and Goldratt uses characters to voice objections he then dismisses. The behavioural diagnosis is the strongest part and is widely accepted even by people who do not adopt the method.
Operators managing projects who want a diagnosis of why estimates inflate and schedules slip, and a scheduling method that aggregates rather than distributes safety.
Israeli physicist and management theorist who created the Theory of Constraints and founded the Avraham Y. Goldratt Institute.
The Goal teaches throughput accounting and constraint management inside a plant-turnaround novel: find the bottleneck, subordinate every other decision to it, raise it, then repeat. The method transfers to any process with a limiting step, which is why operations, project and software-delivery teams still run on it.
Chosen by shared topics, the entries that hold them, and the people who recommended both.