Reasoning about Operationality for Explanation-Based Learning