DESIGN FOR AFFORDABILITY IN DEFENSE AND AEROSPACE SYSTEMS USING TRADESPACE-BASED METHODS by Marcus Shihong Wu B.Eng. Electrical and Electronics Engineering Imperial College London, 2012 Submitted to the Department of Aeronautics and Astronautics and Engineering Systems Division in Partial Fulfillment of the Requirements for the Degrees of Master of Science in Aeronautics and Astronautics Master of Science in Technology and Policy at the Massachusetts Institute of Technology June 2014 © 2014 Massachusetts Institute of Technology. All rights reserved. Signature of Author: Department of Aeronautics and Astronautics, Engineering Systems Division, May 22, 2014 Certified by: Adam M. Ross, Research Scientist, Engineering Systems; Lead Research Scientist, Systems Engineering Advancement Research Initiative; Thesis Co-Advisor Certified by: Donna H. Rhodes, Principal Research Scientist and Senior Lecturer, Engineering Systems; Director, Systems Engineering Advancement Research Initiative; Thesis Supervisor Certified by: Daniel E. Hastings, Cecil and Ida Green Education Professor of Engineering Systems and Aeronautics and Astronautics; Academic Advisor Accepted by: Paulo Lozano, Associate Professor of Aeronautics and Astronautics; Chair, Graduate Program Committee, Department of Aeronautics and Astronautics Accepted by: Dava J. Newman, Professor of Aeronautics and Astronautics and Engineering Systems; Director, Technology and Policy Program ABSTRACT Program failures have plagued the defense and aerospace industry for decades, as unanticipated cost and schedule overruns have rendered the development of systems ineffective in terms of time and cost considerations. This raises the need to holistically include performance, cost and schedule considerations during the early-phase design of systems to perform valuable tradeoffs that derive more feasible and affordable solutions. This paradigm is the design for affordability. This design for affordability conundrum is targeted at defense and aerospace systems, which have complex mission requirements and stakeholder involvement that are susceptible to changes and perturbations over time. Without a systematic framework, the design for affordability process can potentially become cognitively challenging to system architects and lead to unsatisfactory results. To resolve affordability, it can first be defined as the property of becoming or remaining feasible relative to resource needs and resource constraints over time. Affordability can then be treated as an ility that drives the design of more affordable yet technically sound architectures. Tradespace-based methods are introduced to drive affordability and incorporate these holistic considerations into the design process. They facilitate the systematic and disciplined search for affordable solutions to the system, program and portfolio of interest. Multi-Attribute Tradespace Exploration (MATE), Epoch-Era Analysis (EEA) and the Multi-Attribute Expense (MAE) function were modified for affordability analysis. Their feasibility was demonstrated through application to two design case studies. Results from both case studies demonstrated the dynamic tradeoffs among performance, cost and schedule parameters. Tradespace-based methods can thus be applied to the progressive design of systems, programs and portfolios using either a bottom-up or top-down approach to deliver affordable solutions in these cases. Affordability is not only an engineering problem; it is also a policy and management problem. Therefore, affordability can be approached through perspectives beyond engineering design. New policies and refined management practices can be used alongside tradespace-based methods for affordability analysis to ensure the continued delivery of affordable systems for the future. Thesis Supervisor: Dr. Donna H. Rhodes Title: Principal Research Scientist and Senior Lecturer, Engineering Systems; Director, Systems Engineering Advancement Research Initiative ACKNOWLEDGEMENTS "Consider it pure joy, my brothers, whenever you face trials of many kinds, because you know that the testing of your faith develops perseverance. Perseverance must finish its work so that you may be mature and complete, not lacking anything." - James 1:2-4 This quote from the Holy Bible summed up not only my entire experience in writing this thesis, but also the time I have spent in MIT and the United States. 2 years. It has only been two years. As I lay back in my chair and typed away at this section, I took a quick glance at the digital clock on my screen. It was 4:58am. All was dark and quiet around me, yet all felt surreal. I could feel nothing more but leaps of emancipation that were growing stronger and stronger through the night. Deep down, I knew the source of that irrepressible sense of liberation. It was the sweet taste of victory. It was the sweet taste of success. It was the glowing enlightenment of knowledge. It was the warmth of completeness. It was the hope for better days ahead. It was, everything. There has never been a dull moment on my journey. Each day brings new challenges, and I am expected to deliver new solutions the next day. “Taking a drink from the fire hose” is still an understatement of the unique learning experience I received at MIT. Since Day 1, I had set my sights on learning as much as I could and getting the best experience possible out of this prestigious technical institution. At first, I wanted to take a more encompassing approach to learning about the world. That led me to the Technology and Policy Program, which I always consider my home base. I wanted to learn about engineering topics beyond the scope of mathematics and programming, and see how the skills I have acquired initially as an Electrical Engineering student could help shape the world in a way I wanted it to be. Then, I decided to do more and I applied to the Department of Aeronautics and Astronautics to pursue a second, concurrent masters. I have always been interested in space systems engineering but never had the opportunity to pursue further. Simply being at MIT gave me that chance and I was elated that I finally got to pursue what I always dreamt of. At the same time, I also became a Research Assistant at the SEAri, the only laboratory I enquired about and gave me a chance prior to my enrollment in MIT. By the end of 2012, I was on the way to completing two masters and doing my research. All was fine, or so it seemed. But by the turn of 2013, MIT showed its true face, and that hit me really hard. In Spring 2013, I learnt never to take 5 graduate classes at the same time again, and never to take on so many major design projects in one sitting. However, the horror story was just beginning. Despite my asphyxiating class and research schedule, I felt motivated or even compelled to take on the job of President of the MIT Singapore Students Society. It was a rewarding experience to treat my fellow countrymen here in Boston with servitude and smiles, but it was physically and mentally draining. If this looked bad, I only got myself into worse situations. I again felt motivated and compelled to take on the job of Chairperson of the Ashdown House Social Committee. The job scope was no different than that for the Singapore Society, but just for an entirely different crowd. 2 masters programs, 2 leadership roles, and 1 major research commitment. Woe (Wu) is me indeed. But then, the sun began to shine brightly on me. In the midst of all this work and responsibilities, I met a girl. A really great girl called Ruth Choi, from a setting and a background that I never imagined to be associated with. She began to show me that there were indeed greater things out there, not in engineering, but in life. And I knew that she was the girl of my dreams. To cap off all the interesting things I had experienced or subjected myself to, I declared her to be the love of my life. We took our blossoming romance through the summer and winter of 2013 and shared many happy memories along the East Coast. By Spring 2014, I proposed to her and voila, we were engaged and looking forward to getting happily married that July. It has been an incredible journey with so many life-changing twists and turns that brought me through a range of emotions for days on end. By Fall 2013, schoolwork got less intense, but research work began escalating. My job as President and Social Chair also kicked into full force. Graduation was in sights, but many major hurdles still lay ahead. It was evident that I never learnt from the lessons of the past, and I chose to take a very intensive math class instead of the other easier options that were made available to me. I needed the class to graduate, I needed to run my simulation models, I needed to write my thesis, I needed to finish my jobs of President and Social Chair, and I needed to plan for the wedding. And above all that, I wanted to be with Ruth. Fast forward to present day, it may be peculiar as to why I have gone such a long way in describing my journey and not yet thank a single person in this section titled “Acknowledgements”. But I do this to say that this journey is not simply about the things that happened to me. It is about the people who I have met along the way and it is these people that define what my journey really is about. These people have been significant to me in my two years and I am who I am today because of them. Without them, I would not have had the strength to become more mature and complete in the face of trials and tribulations. Without them, I would not have learnt to persevere and grit my teeth through hard times. Without them, I would never have learnt to keep faith in the things that meant most to me. Without them, I would not have found joy in whatever I was doing. The first person I will like to truly thank will be my research advisor Dr. Adam Ross. Adam is an exceptional scientist and his meticulous and intelligent work ethos is exemplar to all. Rarely does an error or even a little blip slip by without his attention, and that drives me to continuously strive towards clarity and near perfection in my work. With his guidance and teachings here in SEAri, I have mastered many important systems engineering concepts and reshaped the way I see the world. Now, I see affordability in everything and I even operationalized the principles of dynamic performance, cost and schedule tradeoffs in wedding planning. An affordable location that provided a wide variety of food, drink and entertainment was found and affordable solutions to the wedding dress, gifts, invitations and photography were derived. Multi-stakeholder negotiation with Ruth, my parents and future in-laws was also conducted. Almost every aspect of the wedding is on track and within budget after I had time and cost as independent variables in my wedding planning design problem. Now, I feel ready for the real world. Of course, systems engineering is not as trivial as that and I have encountered a lot more complex design problems. But from my experience here in SEAri and the work ethics I hope that I have inherited from Adam, I want to be able to take the world by its horns and get things done the right way. The next person I will like to thank will be my other advisor Dr. Donna Rhodes. Donna is the face of SEAri in many ways and she was the one who corresponded with me in my first email exchanges with SEAri. Together with Adam, she brought into me the world of new-age systems engineering and I was never happier to have been able to acquire the very skills that this laboratory was built upon. She has provided me with valuable advice throughout my time here and it has been a pleasure and a fruitful learning experience to be working with her and Adam. I will then like to thank my academic advisor Professor Dan Hastings, who always brings me positivity in his words whenever he shares his experiences in space systems design and engineering. I will not forget how he stepped into my 16.89 Space Systems Engineering class just seconds before I was to present my tradespace analysis segment. He enjoyed the work I produced and that meant a lot to me at that time. I cannot sign off my time here in SEAri without sincerely thanking my wonderful lab mates who are also my fellow graduate students, beginning with those who have graduated. In order of seniority, I will first like to thank Nirav Shah. My encounter with Nirav was a short one as he was gone by the winter of 2012. However, his knowledge of systems engineering was to me well above any other student in SEAri and he could answer almost every question with aplomb. Many other students will also share the same sentiments as me. I recently met him at CSER 2014, and he was still the same: full of wit, full of passion, and full of knowledge. Without him, many of my questions or misconceptions would have remained unaddressed. I will also like to thank Paul Grogan, who very recently passed his dissertation defense. Paul has provided me and everyone else in the lab with new perspectives on gaming simulation and multi-stakeholder interactions. I am also part of the SIRG research group with him and my encounters with him often left me in awe of his work throughput rate. He has programmed many simulations from scratch and conducted many experiments, which I truly enjoyed and learnt a lot. For those who are still currently in SEAri, it has been fun sharing every “SEArious” moment with you. I will like to thank Matthew Fitzgerald, for being the big brother of the student group who takes care of almost everything in the lab, ranging from technical troubleshooting of ViSLab and the computers, as well as watering the plants. I value Matt for his valuable opinions on everything and it is evident that he has thought through the questions that I am just starting to discover. He was the first to give me a student’s perspective of SEAri and I enjoyed his immense contributions to the group. Next, I will like to thank my fellow batchmates, Nicola Ricci and Michael Schaffner. I managed to psyche Nico into applying for TPP and I am glad it is a decision we both do not live to regret. Sharing the office space with Nico was a valuable experience, as we often discussed each other’s work. I got to learn a lot more about options and he got to learn a lot more about affordability. We also got to discuss a lot of TPP homework together in Spring 2013, and never would I have been able to get through that semester without knowing someone who is in the same boat as I am. I also enjoy his signature lines of “Ordinal versus Nominal”, “Static versus Dynamic”, and “So it is ~”. Michael is my partner in crime when we dissected the topic of affordability over these 2 years. I will not forget how we did our first joint presentation, throwing high-5’s as we switched from section to section. We also co-authored a paper together, which essentially threw out the research questions that directed our subsequent research thrusts and our theses. Michael is also a good coffee buddy and everything seems more cheerful when he is around. Together with Nico, there has never been a dull moment in the lab. My new lab mates whom I got to know a lot more this year have contributed greatly to my learning as well. Paul La Tour, for all his experiences and advice in designing satellite systems. Without his help, I would not have had the inspiration to perform some of the modeling techniques shown in this thesis. He also briefly helped validate my FSS model and results and that left me feeling more reassured of my assumptions. Michael Curry, for his technical prowess that I often see evidence of running on the computer screens in SEAri and for his advice of modeling satellite systems. Ben Putbrese and Hunter Zhao, with whom I shared an office briefly, but with memorable moments when we shared our thoughts on life. Yeong Li Qian, for being the other awesome Singaporean in SEAri and also the best baker that I ever knew for a friend. Thanks for taking on the role of President of the Singapore Society. It is a challenging task but I would not have asked him to do it if I did not have the confidence that he could make it his. Other Singaporeans have also been significant here during my time here in Boston and MIT. Kenneth Loh, Tan Siah Hong, Ng Sheng Rong, Karthick Murugappan, Xue Kun, An Jingzhi, Kang Zi Han, Lee Yin Jin, Liu Yun, Jonathan Teo, Ng Huey Jeen, Lim Shi Min and Joy Chua are all good friends who have made my transition from Singapore/UK to USA comfortable and enjoyable. Thanks for keeping me sane and happy even without knowing it. This brings me to my experience in Ashdown House. Grace Gu, for being such an awesome friend without whom I would have run out of ideas and initiatives for more social events. Also, simply for just being a great friend and a supportive buddy. Thomas Mahony and Jenny Schloss, my ex-roommate and his now fiancé, who were great company during my first year in Ashdown. Christopher Foy, for all your wittiness and funny moments that never failed to keep me entertained. And how could I forget my TPP friends, many of whom I have shared a laugh with and there were truly memorable experiences that we were all in together. Brandon Karlow, for being the friendliest person I know around here and my best man for the wedding. I shared many classes with him and I would not have gone through all of them without his words of advice and encouragement. We are practically good buddies on almost every scale. Hisham Bedri and Ekene, for being such cool dudes that I will enjoy hanging out and talking about anything under the sun. Jordan Foley, for helping me out on a number of topics, most notably the naval ship design and for bringing pride to the US Navy. Erin Leidy, for being such a great companion during classes and never failing to cheer anyone up with her bright red hair and vivacious nature. Nina Schuchman, for being such a bubbly friend as well, and together with Brandon and Erin, we form the gang of four that has gone out on countless dinners and drinks together. And also, for being there at some of my lowest moments. I did not have the luxury of time to know everyone in AeroAstro and get off on a good note with them. But one person I have worked with in some of my classes really stands out. Giuseppe Cataldo, whom I worked with in 16.851, 16.89 and 16.895, for being the most reliable and knowledgeable astronautical engineer and space scientist I know, and for always being an inspiration to me, and for hosting me in DC and at NASA Goddard Space Flight Center. Ioana Josan-Drinceanu, Kathleen Voelbel, and Dustin Hayhurst: the Communications Subsystems team in 16.89 that I really enjoyed working in. I never forgot the laughter, the fun, and the pains we went through to get our modular codes working (or so it seemed). Finally, my Mum and Dad, who always love me dearly no matter what happens. It has been tough on them for the last 7 years, during which I have spent most of it away from home due to my time in military service, at Imperial College London in the United Kingdom, and at MIT in the United States. I can still feel the love from them (and my cat) through our regular Skype calls, and we never fail to set up an entertaining chat that we can relish for days. So thank you very much Mum and Dad, without whom I definitely would not be where I am or who I am today. It is only in recent years that I realized how much you have cared for me when I was younger, and I now look back in shame at the moments when I behaved like a spoilt brat or really disappointed them with my bad behavior. Thank you very much for everything. Now, that I am getting married, I have also become part of Ruth’s family. Here in the United States, I have another Mum and Dad from the Choi family. Throughout my visits to Philadelphia in the previous summer, Thanksgiving, Christmas, New Year’s and Valentine’s Day weekend, I felt the family love that I have always been yearning for as long as I have been away from Singapore. Never have I felt so loved by another family before. It definitely reassures me that a lot of love and hospitality do exist in the world out there. Thank you very much for always praying for me, and getting me safely through all my obstacles. To my beloved Ruth Choi, no amount of words can truly express how thankful I am to have you during this period and in my life. You have stood by me all this while and showered my stresses, sorrows and perpetual lack of sleep with your love and grace. Never could I have soldiered through the days without your words and hugs of comfort. You have definitely made me a better person, and it was through you that I rediscovered my faith. Finally, to God Almighty, my Savior Jesus Christ, and the Holy Spirit, thank you for always watching over me. Do let me through this final hurdle and lead me on your path to Glory. Amen. Thank you very much. I dedicate this thesis to all of you. You all mean very much to me. BIOGRAPHICAL NOTE Marcus Shihong Wu was born to a humble family in Singapore, and lived in a cozy government flat tucked away in the only quiet corner of an otherwise bustling city district. He was an only child and did not have many people to look up to except for his parents. However, that was more than sufficient, as his parents sowed the seeds of perseverance, determination and love in him. As a civil servant and a nurse by profession, his parents were doing enough to earn their keep and provided the little family with enough food, sustenance and entertainment. His mother took a first hand in directing her son’s education and before he enrolled in first grade, Marcus was able to read a broad array of children’s books in both English and Mandarin. He was also blessed with good memory and had the knack of repeating word for word what other people were saying and the dialogue lines he heard on television. He began to exhibit streaks of perfectionism, as he would spend hours writing the same Chinese characters until their form was consistent over an entire page. He also liked the arrangement of objects in certain ways, often symmetric or monotonic, and would make all effort to get them the way he wants. Apart from these traits, he was a shy boy who did not enjoy going to the playground and mixing with other kids. He was not the typical boy, as he did not enjoy games very much and often liked to be alone reading his comic books. However, he began to find joy in sports, with soccer and basketball being his favorites. Although he never exhibited much dexterity in any of them, he always enjoyed watching or simply being associated with them. Despite a promising start, Marcus did not do too well in his first two years of school. He was not fantastic in his classes, especially English and Mathematics. However, something unexplained dawned upon him in the third year and from out of nowhere, Marcus topped his class that was ranked close to the bottom. To his surprise, he found out that he even topped his school. He never looked back since and he would go on year after year to achieve stellar academic results. He eventually graduated top of his elementary school. He later enrolled in a prestigious high school called Raffles Institution, which has a rich tradition of grooming many government officials, military leaders, and scientists in the country. It was there that he really began to excel. Apart from excelling in his studies, he also became a leader in his co-curricular activities. It was during his adolescent years that he began to come out of his shell and started making himself heard. He rose to prominence with a close to perfect score in the GCE O Level Examinations and everything seemed on track for him to succeed. However, Marcus got too involved in his sideline activities and began to neglect his schoolwork at a critical period of time. Towards the end of high school, his grades dipped significantly and it was only a miracle that he managed to finish within the top 10% of the school for the GCE A Level Examinations. There were over a hundred people ahead of him and they appeared more well-rounded candidates than him in both academics and leadership. Receiving scholarships were typically regarded as the pinnacle of a local education experience, and the most well rounded students would receive the most prestigious scholarships. Marcus was struggling at this point. He was not on the radar of many prominent scholarship agencies and his chances worsened with his not-so-ideal experiences in the military. Nonetheless, he put on a brave front and applied for a number of scholarships. One night after a day of intense military training, he was surprised to receive a text stating that the Defense Science and Technology Agency (DSTA) of Singapore had offered him an overseas scholarship. He was elated and he promptly seized the chance to become a DSTA scholar and pursue life overseas for a few years. However, there was a little problem. He had to study Electrical and Electronics Engineering (EEE), instead of his preferred major – Aeronautical Engineering. Nonetheless, he went ahead with applying to several UK and US colleges and received a number of positive responses. In the end, he decided to pursue EEE at Imperial College London. He did what any other naïve and excited 19-year-old boy would do – he chose London, not engineering. As someone who chose to major in a subject not for academic interest but purely to seek entertainment outside Singapore, Marcus struggled in his first two years at college. He was still doing well and came in the top 5% for those two years, but there were major incidents in his budding academic journey that shook his confidence. He found great difficulty in programming and had a harder time understanding algorithms as compared to everyone else. Also, he spent a year on a project, which was eventually declared a failure and left with a poor grade being the embodiment of his disappointment. However, the seeds of perseverance and determination began to grow, and Marcus did his utmost best to overcome his weaknesses and failures. In his senior year at Imperial College London, Marcus finally understood why he chose EEE. He began to enjoy all his classes and loved how all the topics were interconnected. Although he excelled again in his final set of undergraduate classes, his big break came through his yearlong capstone project. He had the fortune of being assigned to one of the best supervisors, who took him deep into a journey of memristive devices and quantum mechanics. Marcus, when left on his own, surprisingly began to flourish. He began applying his new knowledge to the development of analogue electronics and bio-inspired devices. He started developing several software tools to analyze the non-linear behaviors of these devices and unknowingly became very adept at programming. He spent day and night working on this project, and he was rewarded with the top prize for best undergraduate project. For the obstacles he had encountered before, this achievement was his crowning moment and Marcus once again found himself at his best. In between, he applied to several US colleges for graduate school. Without harboring a single strain of hope, he was pleasantly surprised to hear that he was accepted into MIT. He was to enroll in the Technology and Policy Program, and was about to step away from the world of nanodevices to confronting broader engineering challenges. He also managed to enroll in the Department of Aeronautics and Astronautics to pursue a second, concurrent Masters. Currently, he hopes to be able to complete all class and thesis requirements in order to graduate. After which, he is planning to get married in Philadelphia to the love of his life - Ruth Choi. Marcus will be commencing his work with DSTA back in Singapore by August and he hopes to be able to apply all he has learnt at MIT to his career and his life. Despite his overly packed schedule, Marcus still finds time to follow his favorite team Manchester United, watch his favorite TV shows, listen to music, and spend time with people who matter to him most! TABLE OF CONTENTS Abstract ...........................................................................................................................................3 Acknowledgements .........................................................................................................................5 Biographical Note .........................................................................................................................11 List of Figures ...............................................................................................................................17 List of Tables .................................................................................................................................23 1 Introduction ............................................................................................................................25 1.1 Motivation: The Ariane VI Conundrum ...........................................................................25 1.2 Prevalence of Cost and Schedule Overruns ......................................................................27 1.3 Considering Performance, Cost and Schedule for Affordability ......................................30 1.4 Finding the Best Approach to Affordability .....................................................................31 1.5 Affordability through Better Systems Engineering ...........................................................32 1.6 Research Questions ...........................................................................................................35 1.7 Research Methodology and Thesis Outline ......................................................................36 2 Literature Review of Affordability .......................................................................................37 2.1 Motivation ............................................................................................................................37 2.2 The Meaning of Affordability ..............................................................................................37 2.2.1 Lexicographic Analysis .................................................................................................37 2.2.2 Common Usage and Application ...................................................................................39 2.2.3 Significance in the Defense and Aerospace Industry ....................................................41 2.3 Affordability in the Context of Engineering Design ............................................................45 2.3.1 Affordability and the Goal of Design ............................................................................45 2.3.2 The Operating Climate for Affordability ......................................................................47 2.4 Current Ways of Understanding Affordability .....................................................................54 2.4.1 Working Definitions ......................................................................................................54 2.4.2 The Affordability Triangle ............................................................................................56 2.4.3 Should-Cost Review and Statistical Methods ...............................................................60 2.4.4 Sand Chart Tool .............................................................................................................63 2.4.5 Cost and Time as Independent Variables ......................................................................65 2.4.6 Effective Portfolio Management ...................................................................................67 2.5 Limitations of Existing Affordability Methods ....................................................................69 2.5.1 Lack of Time Centricity ................................................................................................69 2.5.2 Failure to Recognize Complexities of Scale .................................................................70 2.5.3 Lack of Cost Breakdown Structures ..............................................................................70 2.5.4 Treating Affordability as a Constraint ...........................................................................71 2.5.5 Lack of Value-Centric Perspective ................................................................................71 2.6 Summary ...............................................................................................................................71 3 Tradespace-based Methods For Affordability Analysis .....................................................73 3.1 Motivation .........................................................................................................................73 3.2 The Tradespace Exploration Paradigm .............................................................................74 3.3 Capturing Value through Tradespace Exploration ............................................................76 3.3.1 Value Creation ...........................................................................................................77 3.3.2 Multi-Attribute Utility Theory for Value-Centric Design .........................................77 3.4 Multi-Attribute Expense to replace ‘Cost’ ........................................................................79 3.5 Multi-Attribute Tradespace Exploration for Affordability Studies ..................................80 3.6 Affordability as an Ility .....................................................................................................83 3.7 Affordability in Systems of Systems .................................................................................84 3.8 From System to Program to Portfolio and Back ...............................................................85 3.9 Constraint Levels for Utility and Expense ........................................................................90 3.10 Epoch-Era Analysis for Affordability .............................................................................92 3.11 Demonstration of Tradespace-based Methods in the Affordability Analysis for a Space Tug system and program ............................................................................................................95 3.11.1 Demonstration of System Level Analysis ................................................................95 3.11.2 Demonstration of Program Level Analysis ..............................................................98 3.11.3 Single-Epoch Affordability Analysis for a Space Tug Program ............................100 3.11.4 Multi-Epoch and Single-Era Analysis for a Space Tug program ..........................101 3.12 Summary .......................................................................................................................104 4 Federated Satellite Systems System Analysis ...................................................................107 4.1 The Federated Satellite Systems Paradigm .....................................................................108 4.2 FSS Architectural Assumptions ......................................................................................109 4.3 System Definition ............................................................................................................116 4.4 System Level Design Variables and Epoch Variables ....................................................120 4.5 Hypothetical FSS Development Missions ......................................................................124 4.6 Epoch Construction and Sequencing for System, Program and Portfolio Level Analysis 126 4.7 System Level Design-to-Value Mapping ........................................................................129 4.8 Generation of System Tradespaces and Design Identification .......................................132 4.9 Lognormal Distributions and Confidence Intervals ........................................................136 4.10 Discussion of System Level Results .............................................................................147 4.11 Differences in using MAE and Cost .............................................................................148 4.12 Summary of System Level Analysis .............................................................................149 5 Federated Satellite Systems Program Analysis .................................................................151 5.1 Program Definition ..........................................................................................................151 5.2 Program Level Design Variables ....................................................................................154 5.3 Program Level Design-to-Value Mapping ......................................................................157 5.4 Generation of Program Tradespaces and Design Identification .....................................159 5.5 Expense and Utility Trajectories for Constellation Designs ...........................................169 5.6 Discussion of Program Level Analysis Results ..............................................................176 6 Federated Satellite Systems Portfolio Analysis .................................................................181 6.1 Portfolio Definition .........................................................................................................181 6.2 Determining Weights and Preferences for Portfolio Analysis ........................................184 6.3 Portfolio Cost, Performance, and Satellite Profiles ........................................................189 6.4 Portfolio Analysis Results ...............................................................................................204 6.5 FSS Development Strategy .............................................................................................206 7 Integrating Tradespace-Based Methods with Industry Practices & Policies for Affordability ................................................................................................................................211 7.1 Motivation .......................................................................................................................211 7.2 Benefits and Concerns of Using Tradespace-Based Methods for Affordability Analysis 213 7.2.1 Benefit 1: Systematic, Disciplined and Convenient Approach to Design ..............213 7.2.2 Benefit 2: Layered and Scalable Methodology for Error and Complexity Reduction 215 7.2.3 Benefit 3: Overcoming the Limitations of Existing Methods ....................................217 7.2.4 Drawback 1: Inherent Subjectivity of Stakeholder Preferences and Multi-Attribute Utility 220 7.2.5 Drawback 2: Variations in Model Fidelity and Tractability ....................................225 7.2.6 Drawback 3: Merits and Drawbacks of Qualitative and Quantitative Measures .....226 7.2.7 Drawback 4: Problems with Model Validation ........................................................227 7.2.8 Drawback 5: Availability of Expertise .....................................................................228 7.2.9 Drawback 6: Availability of Decision Makers .........................................................229 7.2.10 Drawback 7: Computational and Schedule Constraints .........................................229 7.2.11 Summary of Implementation Issues ..........................................................................230 7.3 Applying Tradespace-based Methods in Defense and Aerospace Acquisition ..............231 7.3.1 US Department of Defense Acquisition Framework ...............................................231 7.3.2 NASA Project Life Cycle and Trade Study Framework ..........................................235 7.3.3 Affordability Engineering Framework with Tradespace-based Methods ................237 7.4 Affordability through Effective Policies and Management ............................................243 7.4.1 Implementing New Systems Engineering and Acquisition Policies ........................243 7.4.2 Managing Multiple Stakeholders .............................................................................244 7.4.3 Better Communication and Work Flow Organization .............................................244 7.4.4 Integrating Practices from Industry and Academia ..................................................244 7.4.5 Continuously Educate and Train the Acquisition Workforce ..................................245 7.5 Summary .........................................................................................................................245 8 Conclusion & Future Work .................................................................................................247 8.1 General Conclusion .........................................................................................................247 8.2 Research Contributions ...................................................................................................247 8.3 Future Work ....................................................................................................................256 8.3.1 Enhancement to the FSS case study .........................................................................256 8.3.3 Applying the top-down approach .............................................................................257 8.3.4 Enhance the formulation and aggregation of expenses ............................................257 8.3.5 Trading Affordability with other Ilities ....................................................................258 8.3.6 Multi-Era Analysis ...................................................................................................258 8.3.7 The Concept of Product-Service Systems ................................................................258 8.3.7 Affordability, Profitability and Sustainability ..........................................................259 8.3.8 Extending the Scope of Affordability ......................................................................259 8.4 Epilogue ..........................................................................................................................261 References ...................................................................................................................................263 Appendix 1: System Tradespace Model ..................................................................................271 Appendix 2: Program Tradespace Model ...............................................................................287 Appendix 3: Portfolio Level Analysis ......................................................................................305 1 INTRODUCTION 1.1 Motivation: The Ariane VI Conundrum In July 2013, the European Space Agency (ESA) announced that it had selected a basic design for a new launch vehicle, the Ariane VI, which will be powered by two solid-fuelled lower stages and incorporate the liquid-fuelled upper-stage currently being developed as an upgrade for the existing Ariane V vehicle (BBC, 2013). The basic design was chosen after conducting a series of trade studies for six months and feasibility work approved by ESA member states. However, despite being the newest member of the Ariane family, the Ariane VI will have less lifting capacity than the Ariane V and it will only be able to carry a payload of 3 - 6.5 tons to the high orbits occupied by telecoms satellites. The latest variant of the Ariane V, however, will actually be able to carry up to 11.5 tons after its upgrade. Furthermore, the Ariane VI will only be able to launch just one spacecraft at a time, not the two that are routinely lifted by the Ariane V. The Ariane VI seems to have several performance limitations, which may deeply impact the nature of future ESA space missions. Such a development will appear confounding at first sight, as it seems that the latest class of launch vehicles is taking a step backwards in terms of performance. Designing and launching the Ariane VI already consumes substantial time and monetary resources from ESA and such expenditures would typically be justified under traditional standards only if there were a significant improvement in performance levels. However, this was not observed in the Ariane VI. This poses several questions as to how and why this might have happened, and whether it was accidental or intentional. ESA representatives offered their explanations for this new twist in design and they stated that the primary driver for the new configuration is the quest to reduce costs of manufacture and operation. The Ariane V, despite being highly reliable and successful, is priced above its competition and will probably not be sustainable for future space missions. As a result, ESA representatives feared that global demand for Ariane launch vehicles would decrease over time unless a cheaper approach is adopted (BBC, 2013). Therefore, ESA has set a target to try to produce and launch the Ariane VI for no more than about 70m euros (£60m/$90m). This is much cheaper than the current launch cost of the Ariane V, which is estimated to be around 145m euros (£60m /$200m) in 2013 (Parabolic Arc, 2013). However, reducing costs is not a straightforward task. Producing a new launch vehicle would involve the collaboration of many ESA member states and there is much to be done during early-phase design to ensure that the Ariane VI can operate reliably. Alain Charmeau, the CEO of Astrium Space Transportation, which leads the Ariane industrial consortium, expressed his concern regarding the challenge of adopting a cheaper approach and reducing costs: "Astrium now has to capture the ball that has been sent to us today by the agency.” “We will have to make Ariane 6 a very competitive launcher. It's really a complete change in Europe. It's the first time ever that we will try to develop a rocket thinking about production price and not just performance." Through producing the Ariane VI, ESA hopes to achieve more economic returns by reducing the scale of the production consortium spread across the European continent, and by including fewer, less complex components in the build itself. ESA also envisions that once the Ariane VI enters service and proves its reliable performance, it can replace both the more expensive Ariane V and the Russian medium-class Soyuz launcher. If development is approved by 2014, the Ariane VI could make its first flight by 2021-22. Eventually, operating just the Ariane VI alone can help ESA fulfill a wide range of customer needs and provide substantial profits in the future. Therefore, the Ariane VI may pale in comparison in terms of performance, but it is projected to make space transportation more affordable in the future. The Ariane VI conundrum serves to highlight several important issues. It appears that performance has no longer become the top priority during the design process. It is clear that the definition of “better” is no longer confined to performance levels of the system. In order to remain competitive in today’s defense and aerospace industry, cheaper approaches must be explored and there has to be more emphasis on price or cost in addition to performance. Apart from cost, there is also the notion of adhering to the development schedule, as every year delayed equates to another year of operating the more expensive Ariane V. There is now a need to place more priority on cost and schedule throughout the design process and determine which aspects would allow new designs to remain attractive and competitive. Antonio Fabrizi, Director of ESA launchers, offered his opinion of how this could be done: “We don't reduce the costs via technologies; there are no breakthrough technologies that help us to make revolutionary launchers that can provide performance at low cost” It appears to suggest that other ways must be devised and deployed to allow the conduct of tradeoffs between performance, cost and schedule during the design process. 1.2 Prevalence of Cost and Schedule Overruns The approach taken by ESA in designing the Ariane VI is motivated by the need to reduce the cost and time spent developing the system. It is just one of the many design cases where cost and schedule are explicitly taken into consideration during design at the expense of performance levels. Previously, emphasis was placed mostly on maximizing performance and less attention was given to managing cost and schedule attributes. As a result, many cost and schedule overruns are experienced during system and program development. This predicament has been observed to be prevalent throughout the US defense and aerospace industry for decades. In 2012, nearly half of the US Department of Defense’s (DoD) 96 largest acquisition programs (GAO, 2012) have failed to meet the cost growth and schedule standards that were established to identify troubled defense programs (Schwartz, 2010, 2013). In fact, Figure 1-2 shows that the number of programs that met these criteria have been decreasing in recent years. This is an alarming trend as it indicates a reduced buying power for the military. It was further reported that the total acquisition cost of DoD’s Fiscal Year 2011 portfolio of 96 major defense acquisition programs grew by more than $74.4 billion, or 5%, in the past year. About $31.1 billion of that amount can be attributed to factors such as inefficiencies in production, $29.6 billion to quantity changes, and $13.7 billion to research and development cost growth (GAO, 2013). DOD’s largest weapon system acquisition program – the Joint Strike Fighter program – accounted for most of the cost growth (GAO, 2013). However, it is just one of the many programs to experience management and execution problems as a result of cost and schedule overruns. Despite active reductions in weapon unit quantities and reduced performance expectations, the cost overruns on such Major Defense Acquisition Programs (MDAPs) have grown to more than $300 billion over original program estimates. These overruns have led to delays in program developments and even cancellations. Other notable defense programs that experienced cost and schedule overruns were the Army’s Comanche armed reconnaissance helicopter, the Navy’s DDG-1000 next-generation surface combatant, and the Air Force’s Transformational Satellite Communications System (TSAT) (Cancian, 2010). The Comanche program commenced in 1982, but increasing unit costs resulted in a 10-year delay in schedule and its eventual cancellation in 2004. The $6.9 billion initially allocated for the procurement of 120 Comanche helicopters over 5 years could have been directed towards upgrading 350 AH-64 attack helicopters to deliver greater warfighter capability, but was instead used to purchase 800 other helicopters (Cancian, 2010). Similarly, the DDG-1000 program was cancelled in 2009 due to high costs and mission limitations, and funds were instead used to procure additional units of the older DDG-51 model. Unnecessary expenditures and schedule delays could be averted if the Navy initially decided to purchase 13 units of the DDG-51 class for its $23 billion investment in only 3 DDG-1000 units. TSAT was also cancelled in 2009 due to rising costs and schedule slips. The Air Force might have used the $3.5 billion initial investment in TSAT to purchase 7 units of the existing Advanced Extremely High Frequency (AEHF) satellites to avoid gaps in coverage (Cancian, 2010). The failure to deliver these defense systems as a result of cost and schedule overruns can seriously compromise the US military’s warfighting capabilities. These high-profile failures have therefore accentuated the need to reduce cost overruns and schedule delays. These failures are also abundant in the aerospace industry, with the most notable being the James Webb Space Telescope (JWST). Referred to as the “Next Generation Space Telescope” and a top priority in the National Aeronautics and Space Administration (NASA) science decadal survey, JWST is a large deployable, infrared-optimized space telescope that has been designed to succeed the Hubble Space Telescope. JWST is to conduct a 5-year mission to find the first stars and trace the evolution of galaxies from their beginning to their current formation. However, like the defense programs described, the development of JWST also experienced significant increases to project costs and schedule delays (GAO, 2013). Prior to being approved for development, cost estimates of the JWST project ranged from $1 billion to $3.5 billion with expected launch dates ranging from 2007 to 2011. In March 2005, NASA increased the JWST’s lifecycle cost estimate to $4.5 billion and slipped the launch date to 2013. It was found that the cost growth was due to a 1-year schedule slippage, which was caused by a delay in the decision to use the Ariane V launch vehicle. Further schedule slippages followed due to budget profile limitations in later fiscal years. Further cost growth was later incurred as a result of changes to requirements and architectural decisions. Most recently, NASA announced that the project has been re-baselined at $8.835 billion - a 78% increase to the project’s lifecycle cost from the initial baseline - and would be launched in October 2018 - a delay of 52 months. (GAO, 2013 – JWST) Cost growth and schedule growth in other NASA Earth and Space Science missions conducted were also significant. Shown in Figure 1-5, a number of missions have experienced at least a 40% increase in both percent cost growth and percent schedule growth (NRC, 2010). The percent cost growths for some missions were also broken down across the phases during which they were incurred. These phases are “Start to PDR”, “PDR to CDR” and “CDR to Launch”. At NASA, the Preliminary Design Review (PDR) is typically conducted to demonstrate that the preliminary design meets all system requirements with acceptable risk and within the cost and schedule constraints. The Critical Design Review (CDR) occurs later and it serves to demonstrate that the design has matured sufficiently to support proceeding with full-scale fabrication, assembly, integration, and test. From Figure 1-5, it appears that majority of the cost growths occurred from “CDR to Launch”. This trend is important as it highlights that most cost growths occur later in development and it may suggest that little is done during the earlier phases to prevent the occurrence of these growths. Therefore, it is imperative to consider cost and schedule parameters on top of performance levels right at the very beginning of design. Holistic considerations of these elements at program inception can potentially reduce the overruns of overruns in future. 1.3 Considering Performance, Cost and Schedule for Affordability Across all cases of program management failures observed in the defense and aerospace industry, unanticipated cost and schedule overruns have rendered systems and programs unaffordable. Project developments are thus becoming more expensive and longer than initial estimates as a result of both cost and schedule growths over time. The failure to adequately consider cost and schedule at the beginning of design as well as the ineffectual management of tradeoffs among the three key elements have collectively contributed to the current circumstances. The rudimentary tradeoffs among the three elements are shown in Figure 1-6, which shows how performance, cost and schedule are closely interconnected. When a system costs more or is under performing, system architects are immediately prompted to implement a greater cost margin and spend more to make up for these shortfalls. A similar action is taken if the system costs more or is behind schedule. When a system is now behind schedule or underperforming, system architects will apply a schedule margin and extend the development schedule. With more states that the system being developed can exist in, many margins are implemented and their cumulative effects will eventually result in cost and schedule overruns. Therefore, there is a need to perform these trades better to prevent such overruns and be able to design systems that remain affordable over time. This is the principle of affordability. In response to the prevalence of cost and schedule overruns, the US government issued a new initiative - “Mandate affordability as a requirement” (DAU, 2013). Dr. Ashton Carter, then Under Secretary of Defense for the United States, outlined this affordability initiative to improve efficiency in spending to ensure that the country will be able to afford the systems it acquires. Dr. Carter defined affordability as an approach “to manage programs for weapons or information systems without exceeding our available resources”. If programs were not designed with the notion of affordability, more time, effort and money would be wasted on cancelled programs. In the face of budgetary and mission uncertainties, affordability is needed more than ever to achieve the optimal balance of performance, cost and schedule elements. Therefore, affordability has now become a design requirement. It now remains to find out the most preferred and suitable approach to implementing affordability. 1.4 Finding the Best Approach to Affordability To find the best approaches to affordability, the problem has to be tackled at its roots. Over the years, many investigations were conducted to identify the major causes of cost and schedule growth in defense and aerospace programs. A 2009 report by the Institute for Defense Analyses (IDA) narrowed the causes to two main categories: weaknesses in management visibility, direction and oversight; and weaknesses in initial program definition and costing (IDA, 2009). The first broad category covers a general lack of discipline in management, and this may imply lax or inappropriate implementation of policies, excessive reliance on unproven management theories and acquisition strategies, and poor contractor selection processes. The second broad category encompasses failures in systems design and early-phase planning, as well as unrealistic cost estimates. Specific causes within this category may include failure in eliciting or anticipating stakeholder requirements, usage of immature technologies, shortfalls in systems engineering methods, as well as inefficiencies resulting from schedule compression and concurrency. IDA further reviewed a number of programs and discovered their cost growths can be attributed to an overlap of several weaknesses from both categories. The weaknesses underlying the cost growths for 11 selected programs are shown in Table 1-1. Tabulating the frequency of occurrence for each weakness reveals that the lack of appropriate systems engineering methods is a common underlying factor of cost growth in major acquisition programs. Other dominant factors include failures in the requirements process, as well as schedule compression and concurrency. As such, these results indicate that much can be done to enhance the overall systems engineering framework used for designing and managing a program. Therefore, systems engineering can be the main platform upon which affordability can potentially be implemented and designed for. Applying more advanced systems engineering methods thus constitute the best approach to affordability. 1.5 Affordability through Better Systems Engineering Affordability can be implemented in the defense and aerospace industry through better systems engineering methods. Systems engineering is a discipline that has been cultivated from the very same industry and it has evolved as systems become increasingly complex. Nowadays, defense and aerospace systems are composed of a myriad of interacting subsystems that independently and collectively must satisfy a complex set of performance requirements. These requirements can change over the system development cycle and can even evolve during system operation to satisfy new challenges and mission requirements. Therefore, the design, development and operation of such systems can be an arduous task. Systems engineering emerged from these needs to manage complexities and it provides system designers and engineers the capability to ensure that the design process proceeds smoothly and that the system can fulfill most or if not all requirements within budgetary constraints. Therefore, systems engineering is a methodical, disciplined approach for the design, realization, technical management, operations, and retirement of a system (NASA, 2012). It is a way of looking at the “big picture” when making technical decisions and achieving stakeholder functional, physical, and operational performance requirements in the intended use environment over the planned life of the systems (NASA, 2012). To address affordability issues using systems engineering methods, it is first important to note that large commitments of technology applications, system configuration and system performance characteristics, obligation of resources, and potential lifecycle cost all occur at the early stages of a program. The system design conducted during these early stages is known as early-phase design. Referring to Figure 1-7, it is at early-phase design when decisions on conceptual and preliminary design are made and they will have a great impact on the cost of activities later on. However, system-specific knowledge is often limited during early-phase design, but decisions will still have to be made to further development progress. The defense and aerospace industry has been applying a variety of systems engineering methods over the years and they have seen many success, most notably the Apollo program. However, the industry has also been riddled with many program failures and this could indicate the need for novel solutions today to resolve this problem. Therefore, better systems engineering methods may be required for affordability. Therefore, this paves the way for advanced systems engineering methods to be used. Advanced Systems Engineering is “a branch of engineering that concentrates on design and application of the whole as distinct from the parts... looking at the problem in its entirety, taking into account all the facets and variables and relating the social to the technical aspects” (Booton and Ramo, 1984). Therefore, a more holistic approach can be taken to tackle the affordability problem, which has its roots in all facets of performance, cost and schedule considerations. Also, it is clear that defense and aerospace systems are not simply physical products. They are sociotechnical systems that warrant attention to its social, technical, political and economic aspects through their design and development. It is through this justification that methods from the Systems Engineering Advancement Research Initiative (SEAri) at the Massachusetts Institute of Technology (MIT) can be operationalized for affordability purposes. SEAri is a MIT research lab that is affiliated with both the Engineering Systems Division (ESD) and the Department of Aeronautics and Astronautics. It aims to “advance the theories, methods, and effective practice of systems engineering applied to complex socio-technical systems through collaborative research”. Therefore, SEAri is uniquely positioned for interdisciplinary research in advancing systems engineering to meet contemporary challenges of complex socio-technical systems (Ross and Rhodes, 2008a). Through integrating SEAri constructs and applying SEAri methods, advanced systems engineering techniques has the potential for being effective in ensuring affordability in defense and aerospace systems. The underlying goal for systems engineering may be reduced simply to maximizing experienced, and therefore perceived, system success by stakeholders. Part of the success of the system can be defined narrowly, in terms of minimizing costs, improving scheduling efficiencies, or meeting performance requirements, or it can be defined more broadly by maximizing the net benefit experienced by stakeholders through interactions with the system, while meeting or exceeding expectations (Ross and Rhodes, 2008a). Through these advanced systems engineering methods, design and leveraging decisions made early in development can then be improved in terms of efficiency and reliability. MIT SEAri has developed a research agenda that spans over various important aspects of systems engineering. First and foremost, it aims to develop methods for value robustness through concept exploration, architecting and design using a dynamic perspective for the purpose of realizing systems, products, and services that deliver sustained value to stakeholders in a changing world (Ross and Rhodes, 2008a; Ross, Rhodes and Hastings, 2008; Ross, Rhodes and Hastings, 2009). Next, it seeks to enhance sociotechnical decision making through developing multi-disciplinary representations and analysis techniques, and adopts an economics-based view of systems engineering to achieve measurable and predictable outcomes while delivering value to stakeholders (Ross and Hastings, 2005; Richards, Viscito, Ross and Hastings, 2008). In addition, it also aims to achieve more effective systems engineering practice in the context of the system and the characteristics of the associated enterprise (Rhodes, Ross and Nightingale, 2009; Mikaelian et al., 2011). Finally, it focuses on developing prescriptive strategic guidance to inform the development of policies and procedures for systems engineering practice (Broniatowski and Weigel, 2008; Szajnfarber and Weigel, 2009). This makes MIT SEAri well positioned to approach affordability through a systems engineering perspective. Therefore, it is in this motivation that the author of this thesis attempts to address the affordability problem encountered in the defense and aerospace industry through constructs and methods developed by MIT SEAri. 1.6 Research Questions The aim of this thesis is to analyze current concepts and practices associated with the design of affordable systems, and attempt to address the affordability problem in engineering design through advanced systems engineering methods. Research for this thesis was guided by four principle questions outlined below: 1. What is affordability in the context of defense and aerospace systems? 2. How can affordability be incorporated in early-phase design? 3. What are the issues associated with considering affordability in early-phase design? 4. How can affordability concepts be propagated through the defense and aerospace industry? 1.7 Research Methodology and Thesis Outline In this research, a literature review was first conducted to collate and analyze different takes on affordability by various academic, industrial and government institutions. This was done to determine what has already been done in affordability research and what the knowledge gaps that may exist are. A new definition of affordability was then proposed for the purposes of this research. Methods and constructs developed in MIT SEAri were then modified and introduced as potential solutions to address the knowledge gaps and limitations in considering affordability during early-phase design. These methods were applied to a Space Tug system and program case study to demonstrate the feasibility of these methods. The author of this thesis was also a member of the Strategic Innovation Research Group (SIRG), a space systems engineering research consortium comprising student and faculty members from MIT and the Skoltech Institute of Science and Technology. As part of SIRG’s research agenda, additional research was conducted in the design and development of a new space systems concept: Federated Satellite Systems (FSS). The FSS was used as the primary case study in this thesis, where conceptual and computerized models were formulated to facilitate its early-phase design process. The methods introduced were applied to derive affordable solutions for a single satellite, a satellite constellation, and a portfolio of satellite constellations. After the application of the methods to these case studies, the methods were then analyzed in terms of their benefits, risks and cost. This allows potential users of these methods to remain aware of these issues and apply them appropriately to obtain best results. In order to determine how considerations of affordability can be propagated throughout industry, prominent acquisition frameworks were analyzed to assess areas in which the systems engineering methods introduced can be applied for maximal effect. Finally, other strategies for implementing affordability were also discussed so that they can potentially be used concurrently with systems engineering methods to design affordable systems in future. 2 LITERATURE REVIEW OF AFFORDABILITY 2.1 Motivation This chapter aims to provide a comprehensive review of current affordability research and practices in both academia and industry. A repertoire of journal articles, conference papers, industry reports, government documents, theses and books were reviewed in order to obtain a holistic understanding of the approaches taken to integrate the emerging concept of affordability into existing system engineering frameworks and acquisition practices. 2.2 The Meaning of Affordability 2.2.1 Lexicographic Analysis Affordability, a portmanteau of the words ‘afford’ and ‘ability’, has been colloquially defined as ‘the ability to afford’ by all who engage in the universal transactions of time and/or monetary resources in return for desired products and/or services. According to Merriam-Webster (2014), the word ‘afford’ can take on two generalized meanings: to manage to bear without serious detriment and to be able to bear the cost of; or to make available, give forth, or to provide naturally or inevitably. The word ‘ability’ also has two broad definitions: the quality or state or being able to perform or execute; or the natural aptitude or acquired proficiency. Central to these early conceptions of affordability is the notion of 'ability', which can be interpreted as an enabling characteristic or feature inherent to a product or a service that appeals to either the buyer or seller side of the transaction. 2.2.2 Common Usage and Application Affordability is a widespread concept that has been found in virtually every industry and facet of life. Bankole (2011) conducted an in-depth review of affordability-related articles appearing in academia, government and industry. In construction, affordability is “a measure of whether housing can be afforded by certain groups of households” (Semple, 2007) while in the public utility sector, it can be “the ability of customers to pay for utility service billed to them” (Smyth, 2005). In defense and aerospace, it is defined as “the ability to procure a system as the need arises, within a budget, operate at a required performance level; maintain and support it within an allocated life-cycle budget” (Kroshl and Pandolfini, 2000) or “the degree to which the life cycle cost of an acquisition program is in consonance with the long-range investment and force structure plans of national defense administrations” (North Atlantic Treaty Organization, 2007). 2.2.3 Significance in the Defense and Aerospace Industry 2.2.3.1 System vs. Program vs. Portfolio A system is defined to be a combination of interacting elements organized to achieve one or more stated purposes (INCOSE, 2012) while a program can be defined as a group of related and interdependent projects managed together to obtain specific benefits and controls that would likely not occur if these projects were managed individually (KLR, 2008). A portfolio can be defined as a collection of projects or programs grouped together to facilitate the effective management of efforts to meet strategic business objectives (KLR, 2008). Both programs and portfolios can be regarded as System of Systems (SoS). 2.2.3.2 The Better Buying Power Initiative The BBP initiative requires the DoD to “do more without more” by reducing low-priority overheads during periods of budgetary decline and use those funds for modernizing warfighting capabilities (DAU, 2010; DoD, 2012). “Mandate affordability as a requirement” is the first specific initiative. 2.2.3.4 Acquisition vs. Procurement Acquisition includes design, engineering, test and evaluation, production, and operations and support of defense systems (DAU, 2010). Procurement is the act of buying goods and services for the government (DAG, 2010). 2.3 Affordability in the Context of Engineering Design Ross (2006) describes the goal of design is to “create a system that fulfills some need while efficiently utilizing resources within some context”. 2.4 Current Ways of Understanding Affordability 2.4.1 Working Definitions Table 2-1 aggregates definitions across literature from Borky et al. (1998), Mavris & DeLaurentis (1998), Redman & Stratton (2001), Bever & Collofello (2002), Emmons (2010), Mallory (2011), Carter (2010), DAG (2013), Herald (2011), INCOSE Affordability Working Group (2011), NDIA Affordability Working Group (2011), and MITRE (2012). 2.4.2 The Affordability Triangle Tuttle and Bobinis (2012) conceptualized the relationship among Capabilities, Performance, Schedule and Budget. LCC is broken down into RDT&E, Procurement, MILCON, O&M, and MILPERS. 2.4.3 Should-Cost Review and Statistical Methods Should-Cost / Will-Cost management and interval cost estimation methods (Kroshl and Pandolfini, 2000). 2.4.4 Sand Chart Tool Developed by Emmons (2010) and Aerospace Corporation to plot program funding, expenditures, and schedule over time. 2.4.5 Cost and Time as Independent Variables Cost As an Independent Variable (CAIV) (Higgins, 1997) and Time As an Independent Variable (TAIV) (Patterson, 2013). 2.4.6 Effective Portfolio Management Wave model and robust portfolio optimization (Davendralingam and DeLaurentis, 2013; Dahmann et al., 2011). 2.5 Limitations of Existing Affordability Methods 1. Lack of Time Centricity 2. Failure to Recognize Complexities of Scale 3. Lack of Cost Breakdown Structures 4. Treating Affordability as a Constraint 5. Lack of Value-Centric Perspective 3 TRADESPACE-BASED METHODS FOR AFFORDABILITY ANALYSIS 3.1 Motivation Conceptual design formulation and system development require evaluating dynamic tradeoffs among performance, cost, and schedule across alternative futures. 3.2 The Tradespace Exploration Paradigm Tradespace exploration is the model-based investigation of many design alternatives in order to find better design solutions, while avoiding premature fixation on point designs and narrow requirements (Ross and Hastings, 2005). Tracing non-dominated solutions along the frontier produces the Pareto Front. 3.3 Capturing Value through Tradespace Exploration MAUT (Multi-Attribute Utility Theory) aggregates single attribute utilities (SAU) into multi-attribute utility (MAU): KU(X) + 1 = \prod_{i=1}^n [K k_i U_i(X_i) + 1], where K = -1 + \prod_{i=1}^n [K k_i + 1] (Equation 1) 3.4 Multi-Attribute Expense to replace ‘Cost’ To aggregate multiple cost and schedule attributes into a single metric, Multi-Attribute Expense (MAE) is defined using MAUT: KE(X) + 1 = \prod_{i=1}^N [K k_i E_i(X_i) + 1], where K = -1 + \prod_{i=1}^N [K k_i + 1] (Equation 2) MAE is a dimensionless metric on a scale from 0 (minimal dissatisfaction) to 1 (complete dissatisfaction). 3.5 Multi-Attribute Tradespace Exploration for Affordability Studies MATE is parameterized with MAU vs MAE, evaluating design points across epochs. 3.6 Affordability as an Ility Affordability is defined as the property of becoming or remaining feasible relative to resource needs and resource constraints over time. 3.7 Affordability in Systems of Systems Scalability from system to program (constellation) to portfolio (SoS). 3.8 From System to Program to Portfolio and Back A bottom-up approach is utilized to successively evaluate systems, aggregate into programs, and combine into portfolios. 3.9 Constraint Levels for Utility and Expense Defining minimum utility, maximum expense, and derived minimum expected expense constraint levels to form the affordable solution region. 3.10 Epoch-Era Analysis for Affordability EEA discretizes the lifecycle into epochs (fixed context and needs) and eras (ordered sequence of epochs) to track system state transitions of remaining affordable and becoming affordable. 3.11 Demonstration of Tradespace-based Methods in the Affordability Analysis for a Space Tug system and program Case study demonstrating system-level and program-level tradespace exploration for a Space Tug parameterized by manipulator capability, propulsion type, propellant mass, etc. 4 FEDERATED SATELLITE SYSTEMS SYSTEM ANALYSIS 4.1 The Federated Satellite Systems Paradigm FSS is an advanced distributed satellite system (DSS) where heterogeneous satellites share in-space data storage and processing assets opportunistically via inter-satellite links (ISL) (Golkar, 2013). 4.2 FSS Architectural Assumptions 14 fundamental assumptions are established, including static large satellites in LEO, Walker constellations, ISL communication interfaces, laser communications maturation, and reusable launch vehicles (RLV). 4.3 System Definition System Performance Attributes (SYS-PA): - SYS-PA-1: ISL Capability - SYS-PA-2: Annual Free Data Capacity - SYS-PA-3: Annual Science Value - SYS-PA-4: Delta-V System Resource Expense Attributes (SYS-EA): - SYS-EA-1: Development Cost ($ millions) - SYS-EA-2: Launch Cost ($ millions) - SYS-EA-3: Development Time (years) 4.4 System Level Design Variables and Epoch Variables 9 Design Variables (yielding 34,560 potential designs): - SYS-DV-1: Spacecraft Lifetime (10, 15 yrs) - SYS-DV-2: Initial Data Packet Capacity (10k, 15k, 20k, 25k, 30k packets) - SYS-DV-3: Annual Data Packet Usage Rate (5, 15, 25 packets/kg) - SYS-DV-4: FSS Interface Comms Technology (Small RF, Large RF, Laser) - SYS-DV-5: Terrestrial Capacity Real Option (Do not use, Use) - SYS-DV-6: Propulsion Type (Storable Bipropellant, Cryogenic, Electric, Nuclear) - SYS-DV-7: Payload Capacity (30, 100, 500, 1000 kg) - SYS-DV-8: Propellant Mass (500, 1000, 2000 kg) - SYS-DV-9: Launch Vehicle (Falcon 9, Falcon 9 Heavy, Ariane V, Atlas V) Context Variables (4 contexts across 2 variables with Low/High TRL): Laser Comms TRL and Reusable Launch Vehicle TRL. 4.5 Hypothetical FSS Development Missions Missions 0 through 5: Baseline, Space Science, FSS Pilot Mission, Initialization and Development, Age of Discovery, Orbit Reconfiguration. 4.6 Epoch Construction and Sequencing 7 development phases mapped to 7 distinct epochs over 2016-2055 (40 years). 4.7 Design-to-Value Mapping (DVM) DVM matrix (Table 4-7) defines interaction strengths (0, 1, 3, 9). 4.8-4.10 System Tradespaces, Lognormal Uncertainty, and Design Selection Lognormal distributions establish 99% confidence intervals on MAE. High, Medium, and Low risk Pareto designs are selected across the 7 epochs (predominantly in the 17000 series: 10 yr lifetime, 30k packet capacity, 25 usage rate, terrestrial option used, Falcon 9 Heavy). 5 FEDERATED SATELLITE SYSTEMS PROGRAM ANALYSIS 5.1 Program Definition A program is defined as a satellite constellation consisting of multiple homogeneous satellites. Program Performance Attributes (PRG-PA): - PRG-PA-1: Constellation Annual Free Data Capacity - PRG-PA-2: Constellation Annual Science Value - PRG-PA-3: Constellation Maneuverability Program Expense Attributes (PRG-EA): - PRG-EA-1: Constellation Development Cost - PRG-EA-2: Constellation Launch Cost - PRG-EA-3: Constellation Labor Cost - PRG-EA-4: Constellation Operations Cost - PRG-EA-5: Constellation Retirement Cost - PRG-EA-6: Waiting Time to Launch 5.2 Program Level Design Variables 6 Design Variables (1,080 designs per epoch): - PRG-DV-1: Satellite System Design Choice (High, Med, Low Risk from System Level) - PRG-DV-2: Number of Satellites (4, 6, 8, 10, 12) - PRG-DV-3: Number of Satellites in Concurrent Development (1, 2, 3, 4) - PRG-DV-4: Constellation Type (Aggregated, Disaggregated, Mixed) - PRG-DV-5: Legacy Operation (Not for legacy, For legacy) - PRG-DV-6: Program Contract Length (4, 7, 10 years) 5.4-5.6 Program Tradespaces, Trajectory Analysis, and Results Tradespaces are generated for all 7 epochs and constrained by utility and expense bounds. Trajectories across the era reveal fully and partially affordable constellation designs. Down-selected preferred designs (e.g., Design 349, 331, 1069, 691, 277, 1051) combine to produce 72 potential portfolio options. 6 FEDERATED SATELLITE SYSTEMS PORTFOLIO ANALYSIS 6.1 Portfolio Definition The portfolio consists of the full sequence of 7 satellite constellations developed over the 40-year lifecycle. Portfolio Attributes: - POR-PA-1: Overall Annual Free Data Capacity - POR-PA-2: Overall Annual Science Value - POR-PA-3: Overall Maneuverability - POR-EA-1 to POR-EA-6: Overall Portfolio Development Cost, Launch Cost, Labor Cost, Operations Cost, Retirement Cost, and Waiting Time to Final Constellation Launch. 6.2 Weights and Preferences Weights assigned: Free Data Capacity (0.4), Maneuverability (0.4), Science Value (0.2); Development Cost (0.2), Launch Cost (0.2), Labor Cost (0.2), Waiting Time (0.2), Operations Cost (0.1), Retirement Cost (0.1). 6.3-6.5 Analysis, Evaluation, and Selected Development Strategy From 72 portfolio designs, Portfolio 43 [1,2,2,2,1,1,1], Portfolio 19 [1,1,2,2,1,1,1], and Portfolio 1 [1,1,1,1,1,1,1] are closely evaluated. - Portfolio 43 exhibits performance troughs during satellite retirements. - Portfolio 1 is high performance but incurs budget breaches in Years 28 and 31. - Portfolio 19 provides progressive capability growth, maintains continuous high data capacity and science value, and stays strictly within annual portfolio budget constraints throughout the 40-year timeline, making it the most preferred and affordable solution for the FSS. 7 INTEGRATING TRADESPACE-BASED METHODS WITH INDUSTRY PRACTICES & POLICIES FOR AFFORDABILITY 7.2 Benefits and Concerns of Tradespace-Based Methods Benefits: 1. Systematic, Disciplined and Convenient Approach to Design 2. Layered and Scalable Methodology for Error and Complexity Reduction 3. Overcoming Limitations of Existing Methods (time-centricity, scale, cost breakdown structures, affordability as requirement, value-centricity) Drawbacks & Implementation Issues: 1. Inherent Subjectivity of Stakeholder Preferences and MAUT 2. Variations in Model Fidelity and Tractability 3. Qualitative vs. Quantitative Measures 4. Problems with Model Validation and Verification 5. Availability of Domain and Process Expertise 6. Availability of Decision Makers 7. Computational and Schedule Constraints 7.3 Applying Tradespace-based Methods in Defense and Aerospace Acquisition - DoD Acquisition Framework: Implementing MATE during Materiel Solution Analysis (AoA) and Milestones A & B. - NASA Project Life Cycle: Applying MATE to Pre-Phase A / Trade Study processes. - MITRE Affordability Engineering Framework (AEF): Integrating tradespace exploration with AEF's 4-step process and program trigger points. 7.4 Policy and Management Enablers Implementing mandatory affordability SE policies, multi-stakeholder negotiation, improved workflow communication, industry-academia synergy, and workforce education. 8 CONCLUSION & FUTURE WORK 8.1 General Conclusion & 8.2 Research Contributions Answers the four core research questions: 1. Affordability Defined: Property of becoming or remaining feasible relative to resource needs and resource constraints over time; treated as an ility. 2. Incorporation in Design: Utilizes MATE, MAE, and EEA in a layered bottom-up/top-down approach across system, program, and portfolio levels. 3. Issues & Liabilities: Identifies model fidelity tradeoffs, validation limits, stakeholder subjectivity, and computational demands. 4. Industry Propagation: Integrates evolutionary methods with DoD DAG, NASA systems engineering, and MITRE AEF frameworks alongside policy and workforce reforms. 8.3 Future Work - Enhancement of FSS communications and network modeling fidelity (closing link budgets, RF/optical channel modeling). - Applying top-down portfolio optimization. - Formulating candidate functions for interdependent expense aggregation. - Trading affordability with other ilities (survivability, changeability, evolvability). - Conducting multi-era analysis. - Exploring Product-Service Systems (PSS), manufacturer profitability, supplier sustainability, and broader legal/economic factors. 8.4 Epilogue "Affordability is ultimately in the eyes of the beholder. To define what is affordable, design what is affordable and sustain what is affordable, it is our responsibility to make it happen." REFERENCES (Pages 263-270): Comprehensive literature references cited throughout the thesis including DoD directives, GAO reports, NASA handbooks, and seminal systems engineering papers (Ross, Hastings, Rhodes, de Weck, Keeney & Raiffa, etc.). APPENDIX 1: SYSTEM TRADESPACE MODEL (Pages 271-286): MATLAB source code for Generate_Satellite_Design_For_Epoch, attribute and expense calculations, lognormal confidence intervals, and preference weighting for 24 epochs. APPENDIX 2: PROGRAM TRADESPACE MODEL (Pages 287-304): MATLAB source code for Generate_Constellation_Design_For_Epoch, constellation cost modeling (development, launch, labor, operations, retirement, schedule penalty), MAE/MAU computation, and scatter plotting. APPENDIX 3: PORTFOLIO LEVEL ANALYSIS (Pages 305-310): MATLAB source code for portfolio enumeration (72 portfolio designs), lifecycle cost summation, Pareto front evaluation, and portfolio selection.