AMP and Aurora/AMP Planned Space Shuttle Missions for NASA
The Mission Planning Office at NASA planned space missions by allocating resources, assigning facility usage, and calculating launch dates five to ten years into the future. Because the shuttle spacecraft and ground-based facilities were so expensive, any increase in the number of shuttle launches by just one was worth hundreds of millions of dollars, so finding near-optimal schedules was critical.
In addition, NASA carried out “what-if” studies to determine how flight plans must change when ground rules are changed, new facilities are constructed, new vehicles were introduced, or when launches must be delayed. Rapid generation of near-optimal schedules enabled NASA to perform what-if studies efficiently that analyze numerous alternate scenarios.
These complex planning tasks required significant manual effort from experts with many years of mission planning experience. Because of the importance, diversity, and complexity of these scheduling studies, NASA needed a comprehensive and flexible tool to help plan space missions significantly more quickly and efficiently.
Stottler Henke developed for NASA a full-scale Automated Manifest planner (AMP) that enabled NASA staff to create optimized, multi-mission plan orders of magnitude faster than existing tools. Stottler Henke’s knowledge engineers worked with expert planners at NASA to identify and acquire their planning expertise, and Stottler Henke’s artificial intelligence engineers represented this knowledge within the software as rules and objects. Stottler Henke designed the software to be extensible, so planners could enter expert domain knowledge without programming to add, delete, and modify planning heuristics.
AMP was used by NASA to generate short-term and long-term (10-year) schedules of ground-based activities to prepare the space shuttle before each mission and refurbish the shuttle after each mission. AMP had set launch dates automatically, maintained plans, performed advanced “what if” studies, and produced hard copies of plans that can be distributed to other NASA centers. AMP is a NASA Success Story, and was featured in NASA Spinoff 2002.
A successor system to AMP, named Aurora/AMP, was later developed using Stottler Henke’s Aurora™ scheduling system. “The precursor version of Aurora is used daily to support major processing and space shuttle launch decisions; to coordinate our launches with those of Russia, Japan, and the European Space Agency; and to determine NASA’s launch requirements and flight rates,” said NASA Shuttle Processing Manager Tom Overton. “It enables us to generate complex schedules in a few hours, compared to days or weeks required by our previous scheduling systems.”
Aurora Helped NASA Schedule Payload and Vehicle Processing Activities
Preparing vehicles and payloads for launch was an extremely complex process involving thousands of operations for each mission. Because the equipment and facilities required to carry out these operations were extremely expensive and limited in number (often operating at or beyond their capacity), optimal assignment and efficient use were critically important. Overlapping missions that competed for the same resources, ground rules, safety requirements, and the unique needs of processing vehicles and payloads destined for space imposed numerous complex constraints must be satisfied by the schedules.
Traditional scheduling systems used simple algorithms and criteria when selecting the next activity to schedule and when assigning resources and times to each activity. However, schedules generated by these simple and generic decision rules were frequently far from optimal. To solve complex, mission-critical scheduling problems and predict possible problem areas, NASA employed expert human schedulers who used their judgment, experience, and rules of thumb to determine where things should happen, whether they will happen on time, and whether the requested resources are actually necessary.
Stottler Henke developed Aurora, an intelligent planning and scheduling system that enables NASA to solve complex scheduling problems much more quickly by encoding and applying sophisticated, domain-specific, decision-making rules. Aurora users could define attributes for individual tasks, groups of tasks, resources, resource sets, and constraints. These attribute values could be considered by user-supplied or built-in scheduling decision rules that were invoked at key scheduling decision points within single or multi-pass algorithms such as determining which task to schedule next, selecting the overall best time window and resources, or handling the situation where not all of the required resources are available at the required time. Additional attributes of each resource could be considered when making intelligent resource selection decisions in order to generate schedules that are closer to being optimal.
A graphical user interface enabled Aurora users to enter domain-specific knowledge and specify their scheduling requirements quickly and easily. Aurora’s interactive graphical displays enabled the user to visualize and edit the schedule’s resource allocations and the temporal relationships among activities. Scheduling problems, such as unresolved conflicts, were highlighted to attract the user’s attention.
The Aurora™ scheduling system entered operational use at Kennedy Space Center in October 2003. It was used to schedule the use of floor space and other resources at the Space Station Processing Facility (SSPF), where International Space Station components were prepared for space flight.
Aurora was also used to create Aurora/AMP for scheduling Space Shuttle missions (see above).
Aurora was also used by companies to plan complex, large-scale manufacturing operations.
Aurora has been designated by NASA as an SBIR Success Story.
Screenshots of Aurora used for NASA Space Shuttle Processing Facility floor space allocation.
3.23.06 – Stottler Henke to develop software for scheduling astronaut activities aboard next-generation spacecraft.
10.10.03 – Stottler Henke introduces first easy-to-configure intelligent planning and scheduling system.