| Agenda | Tutorials |
Tutorial M |
Architecting Effective Ground System Automation |
Fees |
$250 USD each |
Date |
Monday – February 24, 2025 |
Time |
1:00 PM – 4:30 PM PT |
Overview |
“…best decision you can make is deciding who should make certain decisions and then empowering those people to make those decisions.” Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future by Ashlee Vance
As observed in Elon Musk’s Algorithm, the last step in developing an effective solution is what needs to be automated. In my architecting experience, I have focused on what level of automation (LOA) should be used for each decisioning process. One must minimize time wasted trying to achieve impossible LOAs across various decision types. Starting with learning dynamic responsive decisioning activities and factors that inform the achievable activity LOA; architects can stop over promising and under delivering decisioning capabilities. Early in Musk’s Algorithm, he recommends 1. Question Requirements 2. Delete obsolete and extraneous processes 3. Simplify rest 4. Accelerate cycle time by increasing production velocity and reducing production switching time. This tutorial focuses on the fifth step to select decisioning activities LOA. In the Industrial Age, almost all problems were structured and autonomous. However, today’s activities LOA is not bimodal, but rather have a 5-level scale from no automation (i.e., manual) to autonomous (i.e., no human interaction). Now each activity’s LOA is assessed for the best return on investment (ROI). This means that to decrease overall cycle time, a balance between speeding up production and reducing production switch time is needed. In a stable predictable context, high LOA (i.e., autonomous) improves operational tempo. In an unpredictable changing context, high LOA is brittle to changes/Black Swans, increases switch, generates spoofing susceptibility, and limits rapid access to complex cognitive power (i.e. experienced experts). Using lower LOAs (e.g., facilitation), allows more adaptive decisioning, improves decision confidence, supports graceful degradation, and enables domain expert leverage. Human operators can respond to anomalous behaviors and provide workarounds for unstructured problems often found in new or assets and/or services. Possible LOA depends upon the type of problem’s decision being made. Any complex system will have many decisioning activities ongoing with various levels of human interaction and machine enabling. Dynamic decisioning uses iterative Observe, Orient, Decide, and Act (OODA) loops at many scopes and scales. Decisioning activities consists of Intelligence Management (Observe & Orient/Sensemaking), Decision Making (Decide), and Decision Implementation (Act). Each OODA loop set of activities has different AOLs and approaches. To realize a balance of any LOA’s benefits and detriments requires a tailored mix of LOAs. And for dynamic changing demands, often human facilitation is used to define various Course of Actions (COAs) that can be learned for higher LOAs. An architecture so balanced will have reliable, high operational tempo with effective complex interactions. The tutorial is composed of the following sections:
|
| Instructor | Dr. Linda Vandergriff, Paul Popejoy, The Aerospace Corporation |
Biography |
Dr. Vandergriff is a Senior Project Lead at The Aerospace Corporation. She holds Doctor of Science in Engineering Management and Systems Engineering and serves as a Complex Venture Architect. With 48 years of experience in classical systems engineering, she has served as chief engineer on two major acquisitions and consultant on numerous electrooptics related acquisitions; develops and teaches classes on complex venture architecting and acquisition; and performs mission level behavioral modeling and satisficing studies. She served as a consultant on INCOSE’s 2005 Complex Systems Engineering tutorial. In her 2005 dissertation “Unified Approach to Decision Support for Agile Knowledge-based Enterprises”, she explored the application of complexity theory to the ever quickening pace of ground systems. She was a contributing editor for the 2009 In Search of Knowledge Management Pursuing Primary Principles Knowledge Management textbook. She also wrote sections of the 2007 Aerospace Corporation’s Mission Assurance Guide and 1999 National Science Foundation’s Photonics Curriculum. In her recent research, she is exploring the role of adaptive systems engineering and the application of OODA loop practices to cope at the edge of chaos. The presentations for this conference are synopsis of this research focused in the areas of automation, model based architecting, and complexity theory insights. |
Description of Intended Students and Prerequisites |
A great introduction for people who are not experts in the subject. It applies a structured approach to automation for ground systems so it should also be of interest to those who have some experience in the field. |
What can Attendees Expect to Learn |
|
| Tutorials |