The Art and Science of Systems Engineering [NASA]

wenyj · 发表于 2019-5-12 11:52:21

原文链接：https://www.nasa.gov/pdf/311199main_Art_and_Sci_of_SE_SHORT_1_20_09.pdf

The Art and Science of SystemsEngineering

The Scope of Systems EngineeringThe Personal Characteristics of Good Systems Engineer Summary

This work culminates years of experience in systems engineering and focused discussions among NASA leadership, systems engineers, and systems engineering trainers across the Agency. One consistent theme in these experiences and discussions is that NASA uses many definitions and descriptions of systems engineering. We use the terms and job titles of chief engineer, mission systems engineer, systems engineering and integration manager, system architect, vehicle integration, and so on for various pieces of the complete systems engineering function. We need to agree on a common understandingof systems engineering. In addition, no matterhow we divide the roles and responsibilities among people, we must ensure that those roles and responsibilitiesare clear and executed as a functional whole. Our objectives are to provide a clear definition of systems engineering, describe the highly"effective behavioral characteristics of our best systems engineers and make explicit the expectations ofsystems engineers at NASA.

Systems engineering is both an art and a science. We can compare systems engineering to an orchestra and its ability to perform a symphony. Most people understand what music is, but not everyone can play an instrument. Each instrument requires a different level of expertiseand skill. Some musicians spend their entirecareers mastering a single instrument, which is good because each one needs to be played well. But sophisticated music involves many different instruments played in unison. Depending on howwell they come together, they may producebeautiful music or a terrible cacophony.

We can think of a symphony as asystem. The musicians apply the science of music: they follow the process of translating notes on a page to play their instruments. But an orchestra conductor, a maestro,must lead them to connect the process ofplaying to the art of creating great music. Maestros do a lot more than just keep time They:

* Systems engineering is a critical core competency for successful NASA missions. This paper summarizes the collective wisdom of someof NASA’s best technical minds on the subject. So here the word “we” represents all contributorsto this effort: Michael Bay, Bill Gerstenmaier, Mike Griffin, Jack Knight, Wiley Larson, Ken Ledbetter, Gentry Lee, Michael Menzel, Brian Muirhead, John Muratore,Bob Ryan, Mike Ryschkewitsch, Dawn Schaible, Chris Scolese, and Chris Williams. Among them, they have more than 390 years—almost four centuries—of experience in aerospaceand systems engineering.

1/18/09 1

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image001.pngfile:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image002.png

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

Know and understand music—such matters as pitch, rhythm, dynamics, and sonic qualities—as well as the capabilities of various instruments and musicians
Are necessary once the orchestra reaches a certain size and complexity
Have typically mastered one or more musical instruments
May be composers
Select and shape the music that an orchestra plays
Interpret a composer’s music in light of the audience
Strive to maintain the integrity of the composer"s intentions
Organize and lead the musicians
Are responsible for the success of the performance

The systemsengineer is like the maestro, who knows what the music should sound like (the look and function of a design) and hasthe skills to lead a team in achieving the desired sound (meeting the system requirements).Systems engineers:

Understand the fundamentals of mathematics, physics, and other pertinent sciences, as well as the capabilities of various people and disciplines
Have mastered a technical discipline and learned multiple disciplines
Must understand the end game and overall objectives of the endeavor
Create a vision and approach for attaining the objectives
May be architects or designers
Select and shape the technical issues to be addressed by multidisciplinary teams
Must often interpret and communicate objectives, requirements, system

architecture, and design

Are responsible for the design’s technical integrity
Organize and lead multidisciplinary teams
Are responsible for the successful delivery of a complex product or service

The similarities between maestros and systems engineers are useful in describing the latters’ desired behavioral characteristics and capabilities.

A great systems
completely understands and applies the art of leadership and has theexperience and scar tissue from trying to earn the badge of leader from his or herteam.

Harold Bell NASA Headquarters

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image003.png

Systems engineering is the artand science of developing an operable system that meets requirements within imposed constraints. This definition is independent of scale, but our discussion here focuses on developing complex systems, such as aircraft, spacecraft, power plants,and computer networks.

Systems engineering is holistic and integrative. It incorporates and balances the contributions of structural, mechanical, electrical,software, systems safety, and power engineers, plus many other, to produce a coherent whole. Systems engineering is about tradeoffs and compromises, about generalists rather than specialists.

1/18/09 2

engineer

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

Systems engineering is not onlyabout the details of requirements and interfaces among subsystems. Such details are important, of course, in the same way that accurate accounting is important to an organization’s chief financial officer. But accurate accounting does not distinguish betweena good financial plan and a bad one, norhelp to make a bad one better. Similarly, accurate control of interfaces and requirements is necessary to good systems engineering, but no amount of care in such matters can make a poor design concept better. Systems engineering is first and foremost about getting the right design—and then about maintaining and enhancing its technical integrity, as well as managing complexity with good processes to get the design right. We define interfacesin a system design to minimize unintended interactions and simplify development and operations—and then we document and control the design. Neither the world’s greatest design, poorly implemented—nor a poor design, brilliantly implemented—isworth having.

The principles of systems engineering apply at all levels. For example, engineers who are developing an avionics system must practice creative design and interface definition to achieve their goals.Similar activities are essential to the architecture, design, and development of elements and subsystems across the broad spectrum of NASA developments. But for theremainder of this discussion, we use the term “systems engineering” in the context of complex, multidisciplinary system definition, development,and operation.

In his 2007 presentation, “Systems Engineering and the ‘Two Cultures’ of Engineering,” Mike Griffin describes how the complexities of today’s aerospace systems and the ways they fail have led to branchingwithin the industry. For our purpose, we divide systems engineering into technical leadership and its ally, systems management.

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image004.png

Technical leadership focuses on a system’s technical design and technical integrity throughout its lifecycle
Systems management focuses on managing the complexity associated with having many technical disciplines, multiple organizations, and hundreds or thousands of people engaged in a highly technical activity

Once a credibledesign and architecture are established, the systems engineer’s job is tomaintain technical integrity throughout the complex system’s very rigorous andchallenging lifecycle phases.

Robert Ryan, Marshall Space FlightCenter

Technical leadership, the artof systems engineering, balances broad technical domain knowledge, engineering instinct, problemsolving, creativity, leadership, and communication to develop new missions and systems. It is the system’s complexity, and severity of its constraints—notjust its size—that drives the need for systemsengineering.

NASA systems are often large andcomplex, so they require systems engineers to work in teams and with technical and other professionalexperts to maintain and enhance the system’stechnical integrity. The creativity and knowledge of all of the people involved must be brought to bear to achievesuccess. Thus leadership and communications skills are often as important astechnical acumen and creativity. This partof systems engineering is about doing the job right.

1/18/09 3

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

For large complex systems, there are literally millions of ways to fail to meet objectives, even after we have defined the “rightsystem.” It is crucial to work all the details completely and consistently and ensure that the designs and technical activities of all the people and organizationsremain coordinated—art is not enough.

Systems management is the science of systems engineering. Its focus is on rigorously and efficiently managing the development and operation of complex systems. Effective systems management requires applying a systematic, disciplined engineering approach that is quantifiable, recursive, repeatable, and demonstrable. Here the emphasisis
on organizational skills, processes, and persistence. Process definition and control are essential to effective, efficient, and consistent implementation. They demand a clear understanding and communication of the objectives, and vigilance in making sure that all tasks directly support theobjectives.

Systems management applies to developing, operating, and maintaining integrated systems throughout a project or program’slifecycle, which may extend for decades.Since the lifecycle may exceed the memory of the individuals involved in the development, it is critical to documentthe essential information.

To succeed, we must blend technicalleadership and systems management into complete systems engineering. Anything less resultsin systems not worth having or that failto function or perform.

The Scope of Systems Engineering

Since the late 1980’s, many aerospace#related government and industry organizations have moved from a hard#core, technicalleadership culture (the art) to one of systemsmanagement (the science). History has shown that many projects dominated by only one of these cultures suffer significant ill consequences. Organizations that focus mainly on systems management often create products that fail to meet stakeholder objectives or arenot cost effective. The process often becomes an end unto itself, and we experience “processparalysis.” Organizations that focus solely on technical issues often create products or services that are inoperable, or suffer from lack of coordination and become too expensive or belated to be useful.

To achieve mission success, wemust identify and develop systems engineers that are highly competent in both technical leadershipand systems management. That is why we focuson the complete systems engineer, who embodies the art and science of systems engineering across all phases of aerospace missions—a type reflected in Figure 1. In any project, it is critical that systems engineering be performed well during all lifecycle phases. Thescope of systems engineering and the associatedroles and responsibilities of a systems engineer on a project are often negotiated by the project manager and the systemsengineer. The scope of systems

1/18/09 4

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image005.png

Systemsmanagement provides a framework for problem solving...creative problem solvingfor complex systems.

Dinesh Verma, Stevens Institute ofTechnology

One of thebiggest challenges for a systems engineer of a large complex project is to“bring order from chaos.”

Chris Hardcastle, Systems Engineeringand Integration Manager, NASA’s Constellation Program, Johnson Space Center

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

engineering and the activitiesfor which the systems engineer is both responsible and accountable should be understood and documentedearly in the project.

Figure 1. The Scope of Systems Engineering. Systems engineers often focuson one lifecycle phase like architecture and design versus development oroperations, but good systems engineers have knowledge of and experience in allphases.

Here we describe the characteristics, some innate and others that we can develop, that enable select people to “systems engineer” complex aerospace missions and systems—to design, develop, and operatethem. Then, we focus on how to further developNASA’s systems engineers to help them deal better with the complexities of sophisticated missions andsystems.

The Personal Characteristics ofGood Systems Engineers

Figure 2 depicts the personal behavioralcharacteristics of effective systems engineers.

Intellectual  curiosity.  Perhaps  the  most  important
personal  characteristic  of  successful  systems  engineers  is
intellectual  curiosity.  People  who  prefer  boundaries  around  their work to be comfortable, know what they know,and enjoy
a  focused  domain  may  want  to consider  another  occupation.
Systems  engineers  continually  try  to  understand  the  what,
why,  and  how  of  their  jobs,  as  well as  other  disciplines  and  situations that other people face. They are alwaysencountering new technologies,  ideas, andchallenges, so they must feel comfortable with perpetual learning.

1/18/09 5

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image006.pngfile:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image007.png

People who have“systems engineer” in their title, regardless of the modifiers —project,program, flight system, and so on—are responsible for everything.

Gentry Lee, Jet Propulsion Laboratory

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image008.png

Figure 2. Characteristics of a Good Systems Engineer. The characteristics are shownin decreasing priority from top to bottom. Some of them are innate, whereasothers can be learned and honed.

Ability to see the big picture. Good systemsengineers maintain a big"picture perspective. They understand that their role, though always significant, changes throughout a project’s lifecycle. At any pointin the lifecycle the systems engineer must be fully cognizant of what has been done, what is necessary, and what remains to be done. Each phase has a differentemphasis:

Concept—mission and systems architecture, design, concept of operations, and trade studies
Development—maintaining technical integrity throughout all lifecycle phases: preliminary design review, critical design review, verification, validation, and launch
Operations—making sure that the project meets mission requirements and maintains technical integrity

Systems engineers pay particular attention to verification and validation. Verification answers the question: “Did we build our system right?” If we are successful, it proves our product meets the requirements.We emphasize the hard# earned lesson, “Test like you fly, fly like you test.” Validation,on the other hand, answers the question: “Did we build the right system?” If we are successful, the

1/18/096

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

system does what it is supposedto do, which often goes well beyond just meeting requirements

Good systems engineers are able to “translate” for scientists, developers, operators, and other stakeholders. For example, “Discover and understand the relationship between newborn stars and cores of molecular clouds,” is meaningful to a scientist. But developers and operatorswould better understand and use this version:“Observe 1,000 stars over two years, with a repeat cycle of once every five months, using each of the fourpayload instruments.” The systems engineer that knows the project’s objectives, helpsdetermine how to meet them, and maintainsthe system’s technical integrity throughout its lifecycle has a good chance of succeeding. A corollary is to check everyone"s understanding of each other to make sure the team truly IS on thesame page.

Ability to make system"wide connections. First#rate systems engineers understand the connections among all elementsof a mission or system. They must often help individuals on the team see how their systems and related decisions connect to the bigger picture and affect mission success. The Chandra X#ray Observatory offers a practical example of these connections. The star tracker’s designer must understand that the star tracker is part of an attitude control system—specifically, of an attitude estimator used to take precisely pointed observations—and that the star tracker’s output determines whether or not the properimages are obtained. If the designer does not understand this, the project is in trouble. Good systems engineers can anticipatethe impact of any change injected into the system or project, and describe the nature and magnitude of the impact throughout their system.

Exceptional two"way communicator. Communications skills are the great enabler. Systems engineers need to be able to get out of their offices and communicate well—listen, talk, and write. GeorgeBernard Shaw once stated that England andAmerica are “two countries separated by a common language,” but engineers are separated by their separate languages—even more so since the advent of electronic communications. Systems engineering helps bridge the communication gaps among engineers and managers with consistent terms, processes, and procedures. A key to success isthe ability to see, understand, and communicate the big picture, and be effective in helping others develop a big#picture view.

1/18/09 7

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image009.png

Strong team member and leader. Here we distinguish between management and leadership, realizing that a systems engineer must be skilled in both.

So far, we have described the characteristics that good systems engineers share. Ideally, as they gain experience, they are able to deal with more complex systems through

Breadth of technical knowledge and expertise, combined with execution excellence
Passion for the mission and challenges, combined with force of personality and leadership ability
Creativity and engineering instinct —ability to sense the right way to attack a problem while appreciating inherent risks and implications
Ability to teach and influence others

Comfortable with change. Systems engineers should be comfortable with change.They understand that change is inevitable. They anticipate change, are able to understand how it affects their systems, and deal with thoseeffects properly, usually without losingsleep at night.

While managementand leadership are related and often treated as the same, their centralfunctions are different. Managers clearly provide some leadership, and leadersobviously perform some management. However, there are unique functionsperformed by leaders that are not performed by managers. My observation overthe past forty years...is that we develop a lot of good managers, but very fewleaders. Let me explain the difference in functions they perform.

• A managertakes care of where you are; a leader takes you to a new place

• A manager isconcerned with doing things right; a leader is concerned with doing the rightthings

• A managerdeals with complexity; a leader deals with uncertainty

• A managercreates policies; a leader establishes principles

• A manager seesand hears what is going on; a leader hears when there is no sound and sees whenthere is no light

• A managerfinds answers and solutions; a leader formulates the questions and identifiesthe problems

James E. Colvard

The number of changesmust decrease with time. If projects continue to change, they will never get tothe launch pad. This is particularly true with requirements. While it isundesirable to freeze them too early, it is much more likely that requirementswill continue to change way too long. ...At some point, the design must beimplemented, at which time “change” is the enemy.

Ken Ledbetter, NASA Headquarters

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image010.png

Comfortable with uncertainty. A companion characteristic is being comfortablewith uncertainty—indeed, embracing uncertainty. We usually do not know when we willfinish a task, or even a mission. We knowrequirements are not complete, so we have to interpret them. This is the simple side of uncertainty. But uncertainty hasa more complex side, so a strong background in probability and statistics is important. A good systems engineer understands and encourages quantification of uncertainty. For example, if the mission objective is to land a probe on a comet,the location and severity of jets or debris may be unknown or the comet’s albedo may be uncertain. The systems engineer must be able to work with a team to designa system that accommodates the uncertainties.

Proper paranoia. Anotherimportant characteristic is proper paranoia: expecting the best, but thinking about and planning for the worst. This suggests that the systems engineer is constantly checking and crosscheckingselected details across the system to besure that technical integrity is intact.

Diverse technical skills. A systems engineer must be able to apply sound technical principles across diverse technical disciplines. Good systems engineers

1/18/09 8

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

know the theory and practice of many technical disciplines, respect expert input, and can credibly interact with most discipline experts. They also have enough demonstrated engineering maturity to delve intoand learn new technical areas that should be integrated into the system. They must be strong technical leaders, in addition to having broad technical competence.Systems engineers must meet the special challengeof commanding diverse technical knowledge, plus managing, and leading effectively

file:////Users/wenyuejie/Library/Group%20Containers/UBF8T346G9.Office/TemporaryItems/msohtmlclip/clip_image011.png

Self confidence and decisiveness.

Systems engineers must have well#earned self"confidence. They know what they know and are aware of what they do not know, and are not afraid to own both. It does not mean systems engineers never make mistakes. We haveall made mistakes...at least occasionally.

Commission, notomission. This should be written on the door of every systems engineer. Thereis no excuse for omission. A systems engineer does not need authority fromanyone to investigate anything. The systems engineer’s job is the whole space.You go out, you make decisions. If someone tells you to stop, you use yourcommunication skills and listen.

Gentry Lee, Jet Propulsion Laboratory

Appreciate the value of process. Good
systems engineers appreciate process. That does not mean systems engineering is just one process, plus another, plus another—like recipes in a cookbook. Let uslook back at our metaphor. To create the music of a symphony, musicians use their instruments, musical scores, and notes. Thesetools provide them with a common frame ofreference, help them keep proper time, and allow the orchestra to work together to create beautiful music. Processes serve the same purpose for the systems engineer. But just providing sheetsof music to a group of musicians does not guarantee a great orchestra. While each orchestra uses the same tools and many have very skilled musicians, they do not all sound like the New York Philharmonic.

Herein lies the art—how well does the maestro lead the people and use the tools provided? Maestros know how to bring outthe best in their musicians; they know howto vary the tempo and the right moment to cue the horn section to draw in the listeners. The same is true for systemsengineers. We must all use processes to getthe job done, but it is what we DO with the processes and talents of the team that matters.

Summary

Systems engineering is a crucial core competency within NASA. Systems engineering has two key components: technical leadership, the art, and systems management, the science, that are necessary for mission success. Technical leadership balances broad technical domain knowledge, engineering instinct, problem solving, creativity, leadership, and communication to develop and maintain new missions and systems at NASA.

Systems management"s focus is on rigorously and efficiently managing the development and operation of complex systems. Effective systems management requires applying a systematic, disciplined engineering approach that is quantifiable, recursive, repeatable, and demonstrable. Here the emphasis is on organizational skills, processes, and persistence.

1/18/09 9

THE ART AND SCIENCE OF SYSTEMS ENGINEERING

Systems engineering at NASA is most successful when there is a healthy balance of technical leadership and systems managementengaged in a project.

Systems engineers are a criticalresource for the Agency, and as such, we are dedicated to develop highly capable systems engineers that are able to lead and manage our missions and systems.

1/18/09 10

wenyj · 发表于 2019-5-12 12:02:32

14 DEC 2010Michael D. Griffin Explains How to Fix System Engineering[size=0.875]SHARE THIS STORY

[size=1.25]This week, as part of the ongoing School of Systems & Enterprises (SSE) Dean’s Seminar Series, Stevens Institute of Technology was honored to host Michael D. Griffin, eminent scholar and current professor of mechanical and aerospace engineering at the University of Alabama in Huntsville.

More than 120 guests filled the standing-room only classroom, with another 100 people dialing in to join the event online. SSE Dean Dinesh Verma welcomed everyone, describing the Dean’s Seminar Series as a venue for “folks who need no introduction to come to campus and speak to us about topics in engineering.” Beginning a year and a half ago, the Series is a bi-annual event and features distinguished guests from industry, government and academia.

Stevens’ Interim President George P. Korfiatis then introduced Jeff Wilcox, VP of corporate engineering at Lockheed Martin and member of the SSE Board of Advisors. Dr. Korfiatis said: “We are privileged to have both Jeff Wilcox and Michael Griffin here today.”

“I really wanted to be a part of today’s event,” Wilcox stated. “Michael Griffin is a rare engineering leader as he has combined experience and expertise in government, industry and academia. Stevens is at the forefront of engineering education today, making this an ideal venue for Mike’s seminar.”

Wilcox went on to give a brief bio of Griffin, encompassing his tenure at NASA, five Master’s degrees and inclusion in Time Magazine’s 100 Annual List of the Most Influential People in the World in 2008. Additionally he noted that in November of this year, Griffin was ranked seventh in an informal survey conducted by the Space Foundation of inspirational space heroes.

Currently, Griffin is the King-McDonald Eminent Scholar and Professor of Mechanical and Aerospace Engineering, and the Director of the Center for System Studies at the University of Alabama in Huntsville. From 2005-09 he was the Administrator of NASA. Prior to re-joining NASA he was Space Department Head at the Johns Hopkins University Applied Physics Laboratory. He has also held numerous executive positions within industry, including President and Chief Operating Officer of In-Q-Tel, Chief Executive Officer of Magellan Systems, General Manager of Orbital Science Corporation’s Space Systems Group, and Executive Vice President and Chief Technical Officer at Orbital.

Griffin took the podium and led the attendees through his seminar entitled How do we fix System Engineering? After offering a look at the history of system engineering (and engineering in general), Griffin transitioned seamlessly into his suggestions for the future with well-spoken ease and clearly apparent knowledge. At the end, he opened the floor for some questions, all of which were met with thoughtful clarity.

“In academia and advanced research, I believe we must first ask interesting questions. From there we can set up experiments and studies to find the answers,” Griffin asserted. “My observation is that System Engineering has not followed this process yet.”

In his belief, the true goal should be design of products that are “elegant.” As an engineer, one can easily see when something fits this definition, though Griffin said it’s harder to explain. He said: “We can identify good system engineers, but we can’t always identify good system engineering.”

[size=1.25]To ascertain a product’s “elegance,” Griffin asks four questions: Does this product work? Is it robust? Is it efficient? What does it do that you didn’t want it to do? His conclusion is that there aren’t measures in place yet to determine how well things are done. “We don’t have a theory of system engineering,” Griffin stated.

Griffin closed his seminar with his suggestion for “fixing” system engineering: a proactive, collaborative research effort. “I see the future being driven by inter-disciplinary teams,” he concluded. “We need to combine government, industry and academia and come up with a theory – then we can advance the frontiers of system engineering.”

		自动登录	找回密码
密码			立即注册