Title :
IxTeT: an integrated approach for plan generation and scheduling
Author :
Laborie, Philippe ; Ghallab, Malik
Author_Institution :
CNRS, Lab. d´´Autom. et d´´Anal. des Syst., Toulouse, France
Abstract :
In a traditional approach to solve complex planning problems, plan generation and scheduling are addressed at two different steps of the resolution: plan generation answers the question “what to do?” and scheduling, the question “how to do it?”. This paper describes a planning system, called IxTeT, that handles together and in an opportunistic way plan generation and scheduling, on the basis that the two questions are very much interdependent. Thanks to a rich formalism to represent time, resources, state of the world and changes, our planning system can address a large panel of realistic problems. The search relies on the resolution of flaws on the current plan: it is controlled by a least-commitment opportunistic strategy coupled with an original abstraction hierarchy. A heuristic search driven by a near-admissible algorithm ensures at the same time the global efficiency of the system and the good quality of the solution in terms of flexibility
Keywords :
computer applications; heuristic programming; planning; scheduling; search problems; IxTeT; complex planning problems; heuristic search; integrated approach; least-commitment opportunistic strategy; near-admissible algorithm; original abstraction hierarchy; plan generation; scheduling; Automation; Delay effects; Dictionaries; Humans; Job shop scheduling; Process planning; Resource management; Strips; Time factors;
Conference_Titel :
Emerging Technologies and Factory Automation, 1995. ETFA '95, Proceedings., 1995 INRIA/IEEE Symposium on
Conference_Location :
Paris
Print_ISBN :
0-7803-2535-4
DOI :
10.1109/ETFA.1995.496801