Abstract
One of the central problems in applying computer methods to chemical synthesis planning is Search Guidance. The ability of a chemist to guide an interactive computer in its search for syntheses and the possibilities for self-guidance in Artificial Intelligence programs are both limited by the form and content of the search information that is made available as the exploration proceeds. A system for Search Modelling is proposed in this paper which can augment existing systems for synthesis planning and serce to gather, analyze and amplify the information generated during controlled exploration. The search management model is specified in a simple descriptive form and two example models are included in the paper. The guidance of search using te information gathered is specified by a rule set in the simple syntax of Condition=>Action pairs. A chemist can interactively modify this rule set. Further, an advanced search model is presented which, by introducing the powerful concept of a Planning Space, allows the search for syntheses to go forward, backward and to leap into the middle under controlled conditions.