“STRIPS: A New Approach to the Application of Theorem Proving to Problem Solving” was written by Richard E. Fikes and Nils J. Nilsson of the Artificial Intelligence Group at Stanford Research Institute (SRI) and published in the journal Artificial Intelligence, volume 2, in 1971. The name STRIPS stands for STanford Research Institute Problem Solver. The system was built to plan the actions of Shakey, SRI’s mobile robot, and the version cited here is the proceedings copy from the second International Joint Conference on Artificial Intelligence.
The central problem STRIPS addresses is how a program can find a sequence of actions that transforms an initial situation into one where a goal is satisfied. Earlier theorem-proving approaches tried to reason directly in formal logic and bogged down. STRIPS instead represents the world as a set of logical facts, a “world model,” and searches a space of such models using means-ends analysis: it picks an action that reduces the difference between the current state and the goal, then works to satisfy that action’s requirements.
The lasting contribution is the way STRIPS describes an action, or operator. Each operator has a precondition that must hold for it to apply, an add list of facts the action makes true, and a delete list of facts it makes false. This compact bookkeeping sidesteps the broader frame problem of having to state everything an action leaves unchanged, and it made automated planning tractable. The “STRIPS representation” of actions became a standard that later planners, and the modern Planning Domain Definition Language, were built around.
Why business readers should care: STRIPS is the ancestor of the planning engines that schedule logistics, sequence manufacturing steps, and let software agents break a goal into concrete actions. Its add/delete model of how an action changes the world is still the everyday vocabulary of automated planning more than fifty years later.