[PDF] a brief review of systems engineering programs and a vision - NASA

29076_335_mesmer_a_brief_review_of_systems_engineering_programs_and_a_vision_for_the_future_of_systems_engineering_education_0.pdf Proceedings of the American Society for Engineering Management 2015 International Annual Conference

S. Long, E-H. Ng, and A. Squires eds.

Copyright, American Society for Engineering Management, 2015 A BRIEF REVIEW OF SYSTEMS ENGINEERING PROGRAMS AND A VISION FOR THE FUTURE OF SYSTEMS ENGINEERING EDUCATION

Bryan L. Mesmer, Ph.D.*

Phillip A. Farrington, Ph.D.

University of Alabama in Huntsville

*Bryan.Mesmer@UAH.edu ____________________________________________________________________________________________

Abstract

A basic premise of academia and science is that knowledge is developed using information generated in the past. As

a result, academics is inherently flexible, able to adapt to new knowledge and theories. In the last decade, Systems

Engineering (SE) has seen a significant increase in the amount of knowledge and theories. This recent wellspring of

knowledge has been driven by government organizations, industry, academia, and inter-organizational groups such

as the NASA SE Consortium involving multiple universities and NASA. These groups, and recent workshops and

conferences, have recognized the shortcomings of the traditional SE process and have begun examining a

fundamental rethinking of the theoretical basis of SE. Essentially the development of a science of SE is underway.

SE programs must remain flexible during this time to best serve the students, offering courses that teach the

practices of today (recognizing their benefits and drawbacks) as well as the theories and tools that will drive the

future. The paper examines SE programs across the country in an effort to understand their similarities as well as the

aspects that differentiate them. The paper will offer a vision for the future of SE education that will incorporate

discussions from the NASA SE Consortium which is led by faculty at the University of Alabama in Huntsville. The

courses offered in the UAH SE program will be examined and multiple paths forward will be discussed.

Keywords

Systems Engineering Education, Systems Programs, Systems Courses

Introduction

In recent years the understanding of what a Systems Engineer is and what knowledge they should possess has been

challenged at multiple venues. These venues include workshops such as the NSF sponsored Science of Systems

Engineering Workshop (Collopy and Mesmer 2015),

meetings such as the Council of Engineering Systems Universities (CESUN) Annual Meeting (CESUN 2015). At

the Science of Systems Engineering workshop the theories and fundamental understandings of what defines the

discipline of systems engineering (SE) were explored (Collopy and Mesmer 2015). At the IEEE SysCon 2015

Conference the keynote speech, given by Robert Lyons Jr., discussed various military programs that had technical

issues due mainly to systems engineering (Lyons 2015). A panel of industry, NASA, NSF and academia

representatives at SysCon discussed the Theory of Systems Engineering and the need to understand the fundamentals

beneath the systems engineering practice (McGowan et al. 2015). At 2015 annual meeting a university

roundtable addressed the question of how to educate systems engineering leaders (CESUN 2015). These

conferences, workshops, and meetings are just a handful of the many gatherings of academics, industry, and

government personnel on the future of systems engineering. These presentations and discussions at various

platforms are a product of the pressures being applied by industry and government on academics to address the

current state of systems engineering.

The growing consensus among these groups is that the current practice of systems engineering is flawed.

The Department of Defense experiences daily losses of over $200,000,000 due to cost and time overruns and project

cancellations (Maddox et al. 2013). Traditional tools such as the systems engineering V-model (Shishko 1995)

restrict the ability of the designer and forces the flowing and decomposing of requirements down a hierarchical

Mesmer & Farrington

2

organization structure. In order for the design process of large-scale engineered systems to drastically improve, and

reduce such waste as is seen by the Department of Defense, the fundamental theories of systems engineering must be

identified, understood, and taught. The simple process of identifying the theories is a difficult task that has been the

sole topic of workshops (Collopy and Mesmer 2015). Previous work has examined topics that are important to

systems engineering and has been published in the form of a wiki called the Guide to the Systems Engineering Body

of Knowledge (SEBoK) (BKCASE 2015). The SEBoK does not encapsulate all of the topics that modern systems

engineering is headed towards. The SEBoK is composed of tools and methods that are useful in the current,

traditional practice of engineering. In order to move the field forward a different perspective must be taken that

includes non-traditional engineering fields such as psychology, game theory, organization theory, and philosophy.

Non-traditional engineering concepts from these fields are likely to form some of the fundamental theories behind

systems engineering, and hence must be part of a curriculum to educate a modern systems engineer. The Graduate Reference Curriculum for Systems Engineering (GRCSE) (Pyster et al. 2012) proposes a

framework for graduate studies on systems engineering. The GRCSE focuses on masters programs and does not

concern itself with those interested in research or education that would be in doctoral programs. The GRCSE is

closely tied with SEBoK, focusing more on traditional systems engineering topics such as requirements and

limitation tradeoffs on time, cost, and risk. The reference proposes a common knowledge (foundation knowledge)

that all SE master degree graduates should possess, emphasizing the traditional topics. The GRCSE lays out a

foundation concerning the general types of courses to be offered at each level of depth, but does not set out a course

listing, instead referring to SEBoK for the course topics. The GRCSE provides a start to normalizing systems

engineering education and suggests different programs will have different specializations. This paper envisions the future of the systems engineering program at the University of Alabama in

Huntsville (UAH). We first will examine the characteristics that define a systems engineer both traditionally and in

the future. The needs and definitions from employers (especially NASA through conversations with the NASA SE

consortium being led by UAH) will be examined as well as those from an academic perspective. A short survey on a

select list of SE programs will be performed to examine how each program distinguishes themselves. Future path

forwards for SE program are then described.

Characteristics of a Systems Engineer

The definition of what a systems engineer is depends greatly on the environment and position/background of the

definer. In particular, the definition of what a systems engineer does, how they should do it, and what their

background should be may differ between industry and academia. This section explores these differences based on

conversations with government and industry personnel at multiple conference and workshops.

Traditional Systems Engineer Roles and Training

There are many definitions that exist within industry and government on what a systems engineer is and does. A

common term that is found in many of the definitions is multidisciplinary. Another common term used is integrator.

These terms project the notion that systems engineers understand the physical and organizational ties that bind the

subsystems together. At a high-level there would likely be very little disagreement from the academic community on

this definition. The differences arise when the tasks that define the multidisciplinary integrator are explored.

At present, Industry and government systems engineers are heavily document-centric. Much of the work

executed by them involves the allocation of resources, the communication of requirements, and the traceability of the

work being performed. At a recent meeting of the NASA SE consortium, current role was referred to as that of a (NASA Systems Engineering Consortium: SE Practitioner's Guide

Discussion 2015). The system engineers are the traffic cops of the information that flows to and between the

subsystems/disciplines.

Due to the traditional roles of the systems engineers, the training of those individuals have been molded to

meet the role expectations. Knowledge of traditional approaches such as the systems engineering V-model (Shishko

1995) and of more modern approaches such as Six Sigma (Harry and Schroeder 2006; Neuman and Cavanagh 2000;

Pyzdek and Keller 2003), are necessary information for traditional systems engineers to function in many companies.

These skills, if not taught in a higher education setting, will likely be taught through on the job training or seminars.

Companies and government organizations strive for consistent approaches that can be implemented in many if not all

programs to show efforts in improving their profits or more efficiently or effectively reaching their goals. The

traditional and modern approaches offer a means to show these efforts and have been implemented in many

organizations.

Mesmer & Farrington

3

A consensus by industry and government representatives at a recent workshop focusing on SE in practice

(Research Needs in Systems Engineering 2015) was that systems engineers must have years (i.e., 5 or so) of

experience to truly be capable at their jobs. Furthermore, it is the current view of a group in the NASA SE

Consortium that engineers should not apply for a SE graduate degree until they have spent time working in industry

or government as a systems engineer. In this view the belief is that only through experience and mentoring can one

truly understand the role of a systems engineer and implement the tasks of a systems engineer correctly. This

conflicts with many of the current observations of undergraduate students across the country being hired in positions

Envisioned Systems Engineer Roles and Training

As previously mentioned, there is little argument from the academics in the NASA SE Consortium on the descriptors

of multidisciplinary and integrator as they relate to systems engineers. The topic in general of how to proceed with

the definition of the modern role and training of systems engineers is still being debated amongst the general

academic population and within groups such as the NASA SE Consortium. Due to this on-going debate, the authors

will be expressing their opinion in the following section, understanding many of their views are supported by

members of the consortium and some are not.

The authors believe that the future of systems engineering lies less in a document-centric role and more in a

preference communication role. These preferences being communicated are those of the stakeholder and will

describe what is truly desired. In a commercial setting this preference may be to maximize net present value of

profit. In a government setting this preference may be to maximize the probability of mission success. The

preference is a quantifiable statement that captures the complete life cycle of the system (i.e., a value function

(Collopy 2001; Collopy and Hollingsworth 2011)). Furthermore, this value statement should be an input in a Von

Neumann Morgenstern utility function (Von Neumann and Morgenstern 1944) that enables decisions to be made,

conduit to transfer this information the systems engineer will have a prominent role in extracting the true desires of

the stakeholder and understanding each of the impacts of the system attributes on the system value.

some extent, the subsystems and disciplines that play a role in the system. The NASA SE Consortium is in

agreement that systems engineers should have a broad education where they are exposed to many different fields.

Where some difference in opinion lies is in what fields compose that set.

should not only include the traditional technical fields (aerospace engineering, mechanical engineering, electrical

engineering, industrial engineering, etc.) but also the social, political, and business fields (psychology, economics,

policy and law, decision theory, organization theory, etc.). These fields are important to a systems engineer who

between subsystems and

disciplines. The general agreement that systems engineers should have broad knowledge with a deep understanding

of a specialty is related to the definition of a t-shaped person (Berger 2010). It is the fields that comprise this

breadth and depth that are still in debate.

The question of needing experience to be a good engineer is also still in debate. At first glance it is quite

obvious that experience leads to better performance within the framework of a particular organization. Also with

iplines increases and they can better facilitate the

interactions between the subsystems. As the engineer interacts with more stakeholders they will be better able to

design team more effectively and efficiently. However, with

experience comes a mindset shaped by the environment. If the environment is of the traditional systems engineering

nature, with a focus on traceability and requirement documentation, then that is what the engineer will grow

accustomed to and be an expert in. When going back for further education the engineer may be biased towards their

with a holistic systems view present throughout their education, then they will be entering the workforce with the

mindset of system value, and not of requirements. This is strengthened further in a graduate program that allows for

the acceptance into a program of inexperienced engineers and those with non-engineering undergraduate degrees.

The systems engineering community is recognizing that the discipline must adapt to the modern large-scale

engineered systems, as is evident from the many workshops in recent years on the topic (Bloebaum et al. 2012;

Collopy and Mesmer 2015; Collopy 2011; DARPA/NSF 2009; Simpson and Martins 2010) role to prepare engineers for this change through evolving SE programs.

Mesmer & Farrington

4

Brief Survey of Systems Engineering Programs

A handful of universities recognized for their systems engineering programs are now explored in order to study the

current role of universities in preparing systems engineers. The Worldwide Directory of Systems Engineering and

Industrial Engineering Academic Programs (Worldwide Directory of Systems Engineering and Industrial

Engineering Academic Programs 2015) recognizes 133 United States programs and 107 International programs in

Industrial and Systems Engineering. According to the directory there are 97 universities that have a graduate degree

with the word systems . Due to these large numbers the following survey is not meant to be a

comprehensive examination of SE programs across the nation but instead is a sampling of what is available. This

survey focuses on the graduate systems engineering programs.

Stevens Institute of Technology

Stevens Institute of Technology (Stevens) offers a Masters of Engineering in SE, a Doctor of Philosophy in SE, and

multiple systems related Graduate Certificates. Stevens focuses heavily on the practice of SE, with a statement that

practitioners for practitioners(School of Systems & Enterprise 2015).

level SE courses include traditional probability/statistics, intro to SE, design of experiments, and optimization. The

description of the introduction courses emphasizes the holistic view of SE

focus on traditional aspects of SE roles, such as validation and verification, modeling and simulation, requirements

definitions, quality function deployment, and risk analysis (i.e. Analytic Hierarchy Process). Robust engineering

design is explored using such approaches as Taguchi Methods and Response Surfaces Methods. Multiple courses

are offered concerning specific systems and the tailoring of SE that is performed to handle each. Many of these

courses investigate the systems through a process oriented approach. A number of economic-based courses on

forecasting, demand modeling, and dynamic pricing are also offered. Certification courses through the Defense

Acquisition University are also available. The program also offers SYS 660 Decision and Risk Analysis, which

explores such topics as uncertainty, value of information, risk attitudes, and sensitivity analysis. The 700 level

advance courses are on topics such as data mining, decision analysis, and system and software architecture modeling

(Stevens Institute of Technology: School of Systems & Enterprises: Courses 2015). Stevens captures its goal of

being practitioner oriented but also offers many courses that are being emphasized by modern approaches (such as

uncertainty, risk attitudes, and value of information).

Delft University of Technology

Delft University of Technology (Delft) offers a Masters of Science in SE, Policy Analysis and Management

(Worldwide Directory of Systems Engineering and Industrial Engineering Academic Programs 2015).

masters program is broken into two years (Master Programmes: Systems Engineering, Policy Analysis and

Management 2015). The first year of the program offers the students a broad, foundational understanding of SE

through courses from the perspectives of both the SE and . The students also attend courses on ethics

and legal aspects, institutional design, international designs, and the behaviors of actors that need to be considered in

SE. Also in year 1 the student chooses one domain out of a set of four (Transport & Logistics, Energy & Industry,

Built Environment & Spatial Development, and Information & Communication) (Master Programmes: Systems

Engineering, Policy Analysis and Management: Programme: Domains 2015). Through their chosen domain the

student develops a specialization. Year 2 of the program focuses on the interdisciplinary nature of SE, emphasized

through the use of lecturers from various fields. Progress on projects and challenges are a driver in Year 2. Also in

year 2 the students choose one of six specializations (Emerging Technology-Based Innovation & Entrepreneurship,

ICT Management and Design, Infrastructure and Environment Governance, Economics and Finance, Modelling,

Simulation and Gaming, and Supply Chain Management) (Master Programmes: Systems Engineering, Policy

Analysis and Management: Programme: Specialisations 2015). The Delft masters program offers a broad

understanding of SE, incorporating many non-traditional engineering disciplines, with the ability for the student to

gain the depth in a topic of their interest.

University of Michigan

The University of Michigan (UM) offers a Masters of Engineering in SE & Design (Worldwide Directory of Systems

Engineering and Industrial Engineering Academic Programs 2015). The masters program is composed of three

course segments: the Program Core Courses, Engineering Specialties, and Fundamentals (University of Michigan:

Master of Engineering in Systems Engineering + Design 2015). The program core courses provide the foundation

Mesmer & Farrington

5 SE education. The student takes an introduction course and selects three other courses out of ten

including courses on probability, optimization, control, linear programming, cognitive ergonomics, microeconomics,

and human factors in systems engineering. These courses represent both traditional and modern systems engineering

approaches. The engineering specialty courses allows the student to choose three courses out of a wide range of non-

SE disciplines. These disciplines include: Aerospace, Civil and Environmental, Energy/Electrical,

Mechanical/Automotive, Naval Architecture and Marine Engineering, and Nuclear Engineering. These courses give

the student depth in a non-SE field that is still engineering. The fundamentals course segment is one course chosen

from a large set spanning many disciplines (mathematics, physics, economics, mechanical, manufacturing, industrial,

economics, nuclear engineering). This segment provides the student breadth in a basic course in a field of their

choosing.

George Washington University

George Washington University (GWU) offers a Masters of Science in SE and a Doctor of Philosophy in SE

(Worldwide Directory of Systems Engineering and Industrial Engineering Academic Programs 2015). GWU

describes the SE program as providing broad knowledge on the systems approach and examines systems specific to

NASA, the Department of Defense and U.S. corporations (George Washington University: Systems Engineering

2015). Students in the program must choose one of three specialties: operations research and management science,

SE and integration, or enterprise information assurance. The masters program consists of four core courses

involving management, economics, decision making under uncertainty, and foundational SE. The student then takes

courses related to their specialty to give the student depth. The doctoral program partially includes courses that are