EXPLICABLE PLANNING AND REPLANNING FOR HUMAN-IN-THE-LOOP DECISION SUPPORT AS NASA'S VISION OF HUMAN SPACEFLIGHT AUTONOMY IS REALIZED THE AUTONOMOUS SYSTEM IS EXPECTED TO PROVIDE DECISION SUPPORT IN MANY COMPLEX SITUATIONS. AT THE SAME TIME THE OPERATING PERSONNEL ALIKE IN THE LOOP OF THE DECISION MAKING WILL PROVIDE SUPERVISION ON THE AUTOMATED SOLUTION SYNTHESIS BY GIVING FEEDBACKS TO INITIATE ADJUSTMENTS WHEN NECESSARY. TO MAKE SUCH SYSTEMS USEFUL IT FIRST REQUIRES THE SOLUTIONS OR PLANS SYNTHESIZED TO BE COMPREHENSIBLE TO THE HUMAN OPERATORS. COMPREHENSIBLE PLANS ARE EASILY UNDERSTANDABLE BY THE OPERATORS IN THE LOOP AND HENCE ENABLE THEM TO QUICKLY GRASP THE GIST OF THE PLANS IN THE FACE OF VARIOUS TASK REQUIREMENTS CONSTRAINTS AND PREFERENCES THAT THE OPERATORS MAY HAVE. THIS CAPABILITY OF DECISION SUPPORT SYSTEMS ALLOWS THE HUMANS IN THE LOOP TO MAINTAIN PROPER SITUATION AWARENESS DURING A MISSION WHICH IS ALSO CRITICAL WHEN THEY NEED TO UPDATE THE CURRENT SOLUTION IN CASE THE SITUATION CHANGES. TO REALIZE SUCH DECISION SUPPORT HUMAN DECISION MAKERS WILL NEED A SYSTEM THAT CAN (I) PREDICT THE EXPLICABILITY (OR UNDERSTANDABILITY) OF PLANS BASED ON THEIR CHARACTERIZATIONS GIVEN A DOMAIN AND SYNTHESIZE COMPREHENSIBLE PLANS BASED ON THIS PREDICTION (II) GENERATE EXCUSES WHEN A PLAN DOES NOT EXIST FOR A TASK (E.G. WHEN CERTAIN TASK CONSTRAINTS CANNOT BE SATISFIED) AND GENERATE EXPLANATIONS WHEN PARTS OF THE PLANS ARE EXPECTED TO BE DIFFICULT TO UNDERSTAND (III) REPLAN WHEN THE CURRENT SOLUTION NEEDS TO BE UPDATED TO INCORPORATE HUMAN FEEDBACKS WHILE BEING COMMITMENT SENSITIVE AND GENERATE DIVERSE PLANS IN THE FACE OF PARTIALLY KNOWN HUMAN PREFERENCES. ALL THESE ACTIVITIES MAY BE PERFORMED MANY TIMES THROUGHOUT A MISSION OF HUMAN SPACEFLIGHT AND MUST BE TIGHTLY INTERTWINED IN A DECISION SUPPORT SYSTEM. THE ASSOCIATION BETWEEN THE CHARACTERIZATIONS OF PLANS AND HOW EASILY THEY ARE COMPREHENDED IS USED TO BALANCE PLAN OPTIMALITY (E.G. IN TERMS OF COST) WITH PLAN EXPLICABILITY. AFTER ALL A DECISION SUPPORT SYSTEM SHOULD MAKE ITSELF UNDERSTOOD BY AVOIDING GENERATING INNOVATIVE OR INGENIOUS SOLUTIONS; INSTEAD PLANS THAT ARE MORE EXPLICABLE (OR EXPECTED) BY THE HUMANS SHOULD BE PREFERRED. PLAN EXPLICABILITY INFORMS THE GENERATION OF EXCUSES AND EXPLANATIONS BY ANSWERING WHEN AND WHERE THEY SHOULD BE PROVIDED. EXCUSES AND EXPLANATIONS MAY IN TURN TRIGGER REPLANNING WHICH WOULD INVOKE THE PLAN GENERATION PROCESS. ADDITIONALLY THE REPLANNING PROCESS MUST BE AWARE OF THE PREVIOUSLY MADE COMMITMENTS (E.G. A PREVIOUS GOAL MAY BE HALF ACHIEVED) AND BE ABLE TO GENERATE A DIVERSE SET OF PLANS TO ACCOMMODATE THE PARTIALLY KNOWN HUMAN PREFERENCES. THE TECHNICAL OBJECTIVE OF THIS WORK IS TO DEVELOP IMPLEMENT AND EVALUATE A DECISION SUPPORT SYSTEM FOR HUMAN SPACEFLIGHT THAT SUPPORTS THE ABOVE ACTIVITIES. THE GOAL OF SUCH SYSTEM IS TO FACILITATE A NATURAL HUMAN-MACHINE INTERACTION WHICH IS ALSO DESIRABLE IN A VARIETY OF APPLICATIONS. THE PROPOSED SYSTEMS TERMED READS (RECOMMEND EXPLICABLE PLANS THAT ARE EASILY INTERPRETABLE EXPLAIN PLAN RECOMMENDATIONS VIA EXCUSE AND EXPLANATION GENERATION ACCOMMODATE COMMITMENTS AND HUMAN PREFERENCES FOR DECISION SUPPORT) REPRESENTS A CRITICAL CAPABILITY THAT IS REQUIRED IN SUCH SYSTEMS.