Optimal Problem-Solving Search: All-Oor-None Solutions

Optimal algorithms are derived for safisficing problem-solving search, that is, search where the goal is to reach any. solution, no distinction being made among differ-ent solutions. This task is quite different from search for best solutions or shortest path solutions. L:onstraints mav be placed on the order in which sittes may be searched. This paper treats salisficing searches through partially urdered search spaces where there are multiple alternative goals. (1) Simon was c.upported in pait by Research Grant MH-07722 from the National Institule of Mental Health and in port by the Advanced Research Projecls Agency of the Office of the Secretary of Defense (F14620-73-C-0074) which is monitored by the Air Force Office of Scientific Research; Kadane was supported in part by ihe Office of Naval Research under contract number NOC0i4-67-A-0314-002. //