A Historical Perspective on Development of Systems Engineering Discipline
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
ERDC/ITL MP-21-3
A Historical Perspective on Development of
Systems Engineering Discipline
A Review and Analysis
Niamat Ullah Ibne Hossain, Raed M. Jaradat, Michael A. Hamilton, April 2021
Charles B. Keating, and Simon R. Goerger
Information Technology Laboratory
Approved for public release; distribution is unlimited.
The U.S. Army Engineer Research and Development Center (ERDC) solves the nation’s toughest engineering and environmental challenges. ERDC develops innovative solutions in civil and military engineering, geospatial sciences, water resources, and environmental sciences for the Army, the Department of Defense, civilian agencies, and our nation’s public good. Find out more at www.erdc.usace.army.mil. To search for other technical reports published by ERDC, visit the ERDC online library at https://erdclibrary.on.worldcat.org/discovery.
ERDC/ITL MP-21-3
April 2021
A Historical Perspective on Development of
Systems Engineering Discipline
A Review and Analysis
Simon R. Goerger
Information Technology Laboratory
U.S. Army Engineer Research and Development Center
3909 Halls Ferry Road
Vicksburg, MS 39180
Niamat Ullah Ibne Hossain and Raed Jaradat
Department of Industrial and Systems Engineering
Mississippi State University
75 B. S. Hood Road
Starkville, MS 39762
Michael A. Hamilton Charles B. Keating
Institute for Systems Engineering Department of Engineering Management
Mississippi State University, ISER and Systems Engineering
3090 Halls Ferry Road Old Dominion University
Vicksburg, MS 39180 Norfolk, VA 23529
Final report
Approved for public release; distribution is unlimited.
Prepared for U.S. Army Corps of Engineers
Washington, DC 20314-1000,
Under Program Element 0603461A, Project Number DW5, Task Number 01
ERDC/ITL MP-21-3 ii
Preface
This study was conducted for the U.S. Army Corps of Engineers and
funded under Army Direct Program Element Number 0603461A, Project
Number DW5, Task Number 01. The technical monitors were Dr. Simon
R. Goerger and Dr. Randy K. Buchanan.
The work was performed by the Institute for Systems Engineering
Research (ISER) (CEERD-IER) of the Computational Science and
Engineering Division, U.S. Army Engineer Research and Development
Center, Information Technology Laboratory (ERDC-ITL). At the time of
publication, Dr. Simon R. Goerger was Director, IER; and Dr. Jerrell R.
Ballard Jr. was Chief, IE. The Technical Director was Dr. Robert Wallace.
The Deputy Director of ERDC-ITL was Ms. Patti S. Duett, and the Director
was Dr. David A. Horner.
This paper was originally published in the Journal of Systems Science and
Systems Engineering on 1 February 2020.
The Commander of ERDC was COL Teresa A. Schlosser and the Director
was Dr. David W. Pittman.
DISCLAIMER: The contents of this report are not to be used for advertising, publication, or promotional purposes.
Citation of trade names does not constitute an official endorsement or approval of the use of such commercial products.
All product names and trademarks cited are the property of their respective owners. The findings of this report are not to
be construed as an official Department of the Army position unless so designated by other authorized documents.
DESTROY THIS REPORT WHEN NO LONGER NEEDED. DO NOT RETURN IT TO THE ORIGINATOR.
A Historical Perspective on Development of
Systems Engineering Discipline:
A Review and Analysis
Abstract. Since its inception, Systems Engineering (SE) has developed as a distinctive discipline, and there
has been significant progress in this field in the past two decades. Compared to other engineering dis-
ciplines, SE is not affirmed by a set of underlying fundamental propositions, instead it has emerged as a
set of best practices to deal with intricacies stemming from the stochastic nature of engineering complex
systems and addressing their problems. Since the existing methodologies and paradigms (dominant pat-
terns of thought and concepts) of SE are very diverse and somewhat fragmented. This appears to create
some confusion regarding the design, deployment, operation, and application of SE. The purpose of this
paper is 1) to delineate the development of SE from 1926-2017 based on insights derived from a histogram
analysis, 2) to discuss the different paradigms and school of thoughts related to SE, 3) to derive a set of
fundamental attributes of SE using advanced coding techniques and analysis, and 4) to present a newly
developed instrument that could assess the performance of systems engineers. More than Two hundred
and fifty different sources have been reviewed in this research in order to demonstrate the development
trajectory of the SE discipline based on the frequency of publication.
1. Introduction distinctive specialized discipline since its in-
After World War II, there was a fundamental ception. There have been rapid and continuing
operational transformation in industrial and advances in this area in the last two decades,
construction sectors around the world. During ultimately targeted to address the intricacies
the war, a new engineering discipline known stemming from increasingly sophisticated and
as "Systems Engineering (SE)" evolved as a ma- diversified complex systems permeating every
jor new paradigm to countervail the complexi- aspect of society. Unlike traditional engineer-
ties associated with newly emerging processes ing, systems engineering is not grounded by a
and systems (Gorod et al. 2008). Systems en- set of rigidly defined basic theorems anchored
gineering has continued and developed as a in science related to physical properties. In-
1
2
stead, SE has evolved as a set of best techniques achieve this purpose, the paper will explore
for managing the ill-structured complex prob- the evolution of the SE field by segmenting the
lems based on circumstances (Hallam 2001, discipline development timeline into three dif-
Hossain and Jaradat 2018). ferent intervals and examining the significant
At the most basic level, SE is implementa- developments within those intervals. It is an-
tion of systematized methodologies to guide ticipated that this view will offer the reader a
the design, analysis, execution, and develop- comprehensive map of the development of SE
ment of systems that addresses needs and re- and highlight the involvement of past contrib-
solve problems (Hossain and Jaradat 2018). utors to the progression of SE. The objectives
Systems engineering addresses the life-cycle of this paper are as follows:
of product systems from conception to dis- - To trace the history development of SE
posal, and it operates to trace and satisfy from 1926-2017 based on insights derived
customer requirements within constraints of from a histogram analysis. This would
the system. Traditional Systems Engineering provide a comprehensive overview of SE
(TSE) deals with single complex system prob- domain.
lems in order to optimize the performance of - To derive a set of common characteristics
the system. Currently, the representation of of SE using advanced coding techniques
SE consists of different interpretations in- and analysis. This would serve as a base-
cluding life-cycle based approaches, manage- line snapshot to invoke a dialogue that
ment technology paradigms, process-problem possibly contribute to fruitful to future
archetypes, discipline-oriented paradigms, advancement of SE field.
and systems thinking and non-systems think- - To lessen the confusion pertaining to SE
ing approaches (Kasser and Hitchins 2011). and its derivative terms. This would al-
While this suggests a somewhat fragmented low the practitioners to understand the
discipline, a more rigorous development of the applicability of SE terminology and how
historical roots and evolution of development these nomenclature are embedded in SE
might serve to better understand two central definition.
issues. First, how this discipline arrived at its - To present a newly developed instrument
present state. Second, what this historical basis that could capture the performance level
portends for future development of the disci- of individual system engineers. It will
pline. help individuals/ group of systems engi-
Although SE has been introduced in the neers to identify their weak zone and de-
defense and space industries, efforts are being velop themselves to encounter the intri-
made to extend the application of the disci- cacies emanating from complex systems
pline to different fields as well (Shenhar and in where they are anticipated to be en-
Bonen 1997). However, regardless of having gaged.
diversified applications of SE, many scholars To achieve the objectives of the paper, more
and practitioners continue to publish their re- than two hundred and fifty different resources
search under the domain of the SE discipline. have been coded and analyzed. The spectrum
The state of art of SE literature is a somewhat of sources includes scholarly journal articles,
fragmented compilation of apparently modi- conference proceedings, letters, technical pa-
fied perceptions of related domains. The main pers, special features, books and book chap-
purpose of this paper is to trace the chrono- ters. Since it is difficult to trace all works per-
logical development of SE from 1926-2017. To taining to SE, related works that contributed
3
Figure 1 Classification of SE Interval for Histogram Analysis.
most significantly to the field of SE (based on 2. Histogram Analysis
the frequency of citations) are used as a pri-
In this section, the design and execution of
mary criterion for the selection of publications
the histogram analysis are developed (results
for inclusion in the analysis. To trace the pro-
summarized in Figure 6). The following top-
gression history of SE, we considered Ferris
ical areas were selected to guide examination
(2007a), Ferris (2007bc), Gorod et al. (2008),
of the literature to comprehend the histogram
Brill (1998) as a grounded references where
analysis: (1) definitions of SE, (2) characteris-
Ferris (2007a), Ferris (2007bc) explored the
tics for SE, (3) principles and axioms for SE
early history of SE during pre and post-world
and (4) different perspectives and methodolo-
war era. Gorod et al. (2008), and Brill (1998)
gies supporting SE. The histogram analysis
traced the history of SE from 1950-1995. This
provides a comprehensive discussion of differ-
research provides the comprehensive review
ent aspects of SE on a chronological develop-
of SE history from 1926-2017 and traces the de-
ment scale, rather than other potential organiz-
velopment of SE discipline over the years.
ing constructs (e.g. sector, geography, theme,
Although not all SE works are included, etc.). Chronological tracing of the SE disci-
the underlying overview originating from this pline development is offered as a path to poten-
synthesis will provide a good understanding tially different insights and future implications
of the field as a whole. Even though there is based on the time-based development of the
not a detailed discussion of all the references, SE discipline. To create a histogram analysis, a
all 250+ sources are incorporated in the analy- time range of 91 years was used, the difference
sis. Grounded Theory Coding (Charmaz and between the highest value (2017) and the low-
Belgrave 2012) techniques were employed with est value (1926). This range would cover the
the use of Nvivo 12 (QRS International 2017) historical context of SE from its inception to
software that helped in structuring the large 2017 through three intervals namely, (SE intro-
dataset. ductory, SE development, and SE revolution-
The construction of the histogram analysis, ary periods). Figure 1 provides the interval
consisting of three main intervals, is presented classifications for the histogram.
below. The examination of the intervals is fol- The purpose of the histogram plot is (1)
lowed by the progression history of SE pertain- to provide quantitative information about the
ing to those three intervals. From the results underlying frequency distribution of literature
of the analysis, the paper concludes with a dis- spanning the SE discipline history from 1926-
cussion of the implications of the analysis for 2017 and (2) to discuss the main themes and
the SE discipline along with the avenue of fu- challenges for the SE discipline that are derived
ture research. from each interval. The horizontal axis in the
histogram signifies the timeline of the study
(classes) whereas the vertical axis embodies the
4
relative frequency of contribution activity for ity of equipment was not as central of a con-
each class (see Figure 6). This organization cern. However, during World War II, electronic
offers one of many possible ways in which the equipment became so sophisticated that relia-
literature might be organized and examined. bility became a serious concern. For instance,
However, although not absolute, the inclusion due to poor radar reliability, numerous battle
of both frequency and content themes provides ships were sunk at the beginning of the war in
a clearer picture of the discipline development the Pacific. Along the same line, during the Ko-
from the perspective sought in this paper. rean War, bombing missions were halted due
to inability to effectively operate the complex
2.1 Intervals
electronic weapon systems (Brown 1953). The
Based on the histogram analysis and the
complexity of the equipment exceeded capabil-
grounded theory coding, three main intervals
ities of service operators to maneuver the ap-
were derived. Each interval reflects the de-
paratus properly during operation, resulting
velopment of SE history during that period of
in reliability becoming a prime concern of mil-
time. The first interval, labeled as the ’SE Intro-
itary applications (Romig 1956). In order to ad-
ductory’ interval, is from 1926-1960, the second
dress this issue, the American military sought
from 1961-1989 labeled as ’SE Exploratory’ in-
help from large numbers of engineers and sci-
terval and the third labeled as ’SE Revolution-
entists to develop a technique to deal with
ary’ interval is from 1990-2017. For each of the
these increasingly complex problems. This
intervals, an interpretation of the major con-
joint military-civilian endeavor was named Op-
tributions to the body of SE are identified and
eration Research. The accumulated knowledge
discussed.
and experience that resulted from World War
2.1.1 Interval I (1926-1960): Introduction of SE
II stimulated the application of the systems ap-
SE is entrenched in older management
proach in different domains. A noteworthy ex-
archetypes that were used during the construc-
ample of invention during World War II were
tion of numerous ancient projects. Among
"black boxes" used on aircraft. Demand for
these projects were the pyramids in Egypt, multiple types of electronic gear essential for
the water distribution and irrigations systems airborne operations triggered the development
in Mesopotamia, and the infrastructure ex- of a widespread types of elemental devices,
pansion in Greece and Rome, as well as the commonly known as "black boxes" (Engstrom
more modern 19th-century canals and rail- 1957). These inventive avionic architectures
roads (Kasser 2002). The construction of John included multiple systems that were synchro-
Ericsson’s iron-clad battleship from the Civil nized with the aircraft system to perform indi-
War era presented another example of histori- vidual functions (Tolk et al. 2011).
cal evidence of the use of SE (Engstrom 1957).
The earliest foundations of SE can be traced to During the 1930’s and 1940’s a rapid ad-
Smuts (1926) who first coined the term "holon" vancement took place in the field of technol-
to describe the "wholeness or the integration ogy, especially in space and control engineer-
of the elements of a system." The concept of ing, power distribution, and communication
holism which developed from this term is still systems. Reflections of these technological ad-
considered to be one of the fundamental at- vances led to thinking about building struc-
tributes of SE. tures that could be made even more robust by
Prior to World War II, military weapons and combining different interdisciplinary en-
equipment were not as complex as those in gineering approaches. This interdisciplinary
use and development today, thus the reliabil- systematic approach was actively incorporated
5
in radio, telephone, and television industries prised of three different groups; systems en-
during the late 1930s and ushered in the gineering, design and development, and pure
evolution of modern telecommunications net- research (Kelly 1950). Bell Telephone Labora-
work. For instance, the Radio Corporation tories was perhaps the first organization to coin
of America (RCA) and Bell Telephone Com- the phrase "systems engineering" (Schlager
pany aimed to expand the television trans- 1956).
mission domain and long-distance telephone
The first operational intercontinental bal-
network, respectively, using new broadband
listic missile (ICBM) program, known as the
technologies. However, these experimental
Atlas ICBM program, also bears significance
projects failed to progress due to the inter-
to the inception of SE. Before the Atlas ICBM
ruption caused by World War II. As a conse-
program, the prime airframe manufacturers
quence, in lieu of the telecommunications in-
were only contractors accountable for design-
dustry leading the SE discipline development,
ing military aircraft and supervising all the
the Department of Defense (DoD) was placed
subcontractors under the authority of the U.S.
"front and center" in leading SE development.
Air Force. As a result, there was a scarcity of
World War II was arguably the first time resources to produce the military weapons for
practitioners realized the importance of man- the U.S. Air Force. In the early 1950’s, when
aging and synchronizing various complex sys- further development of an ICBM capability be-
tems to achieve long-term objectives. As came necessary, the Air Force again looked to
an outcome, "quantitative management" tech- enlist the services of the airframe manufactur-
niques were developed out of World War II. ers. Subsequently, the Strategic Missile Eval-
In the post-war era, many perceived that the uation Committee (codenamed Teapot Com-
techniques developed during the war could be mittee) was formed to assess various missile
extrapolated and applied to other fields as well. development projects all over the U.S (Hallam
For instance, after World War II, the scientists 2001). The primary charge of this committee
and researchers from RAND (research and de- was to track the duplication of implementa-
velopment) corporation, Bell Telephone Labo- tion strategy and to appraise the competence
ratories and RCA capitalized on the war-time of airframe prime contractors in order to de-
experiences in advancement and expanded the velop a system requiring substantial electronic
technology of modern telecom and electrical and computational capabilities. Several thou-
power systems (Tolk et al. 2011). The RAND sand skilled engineers, scientists, contractors,
Corporation, originated in 1946 by the United subcontractors and specialists were involved
States Air Force, developed a "systems anal- in the Atlas program. The Teapot commit-
ysis" methodology which is still considered to tee (lead by Simon Ramo) contributed to the
be one of the fundamental concepts of SE. RCA establishment of SE as a discipline by devel-
also deployed the "systems approach" for the oping an administration responsible for mon-
advancement of electronically scanned, black itoring and coordinating all the necessary ac-
and white television (Engstrom 1957). In 1943, tivities for subcontractor design,development
to further advance the Aircraft Warning Com- test, integration, verification and validations
munication Service, the National Defense Re- (Hallam 2001). Following the success of the
search Committee (NDRC) formed a systems Atlas program, scholars from different disci-
committee in conjunction with Bell Laborato- plines extrapolated the technique followed in
ries to conduct a project named C-79 (Buede the military program to management science,
and Miller 2016).Bell Laboratories was com- and SE evolved as a budding discipline at that
6
time. - To encourage the harmony of science
After World War II, MIT Radiation Labo- through enhancing the communication
ratory, known as Rad Lab, published a series among the practitioners (Checkland,
of books, which discussed the application and 1993:93)
evolution of radar systems during the war. Al- There are some other theories, such as
though the series did not cite the term "sys- Game Theory and Information Theory (Shan-
tems engineering" they did highlight how a non and Weaver 1949) that somewhat re-
holistic approach could be applied to an en- semble or are related to the themes of gen-
gineering system (Ferris 2007c). In 1950, the eral system theory, and these theories were
first formal endeavour to teach SE was made widely adopted during this period of time.
by G. W. Gilman who was the Director of sys- During and after World War II, a number of
tems engineering at Bell Laboratories at Mas- projects were undertaken in the U.S. to defend
sachusetts Institute of Technology (MIT) (Hall its people and protect its borders such as the
1962). In 1955, the biologist Ludwig Von Berta- Analyze air defense system (1937) and Nike-
lanffy along with economist K.E. Boulding, line-of-sight -anti-aircraft missile system (1945-
physiologist R.W. Gerard, and the mathemati- 1953). The complexity and stochastic nature of
cian A. Rappoport developed the idea of gener- the projects necessitated a systemic holistic ap-
alizing ’Systems Thinking’ or ’Holistic Think- proach to successfully accomplish the project
ing’ to any kind of system; their ideas became goals.
known as "General Systems Theory (GST)"
Schlager (1956) was the first person to for-
(Bertalanffy 1968). This theory emerged due to
malize a brief outline of the SE process encom-
the inadequacies of science alone to offset the
passing planning, analysis, optimization, in-
challenges of complexity and confronted the
tegration and testing. He also suggested the
effectiveness of reductionist based approaches
adaptation of different types of systems analy-
for increasingly complex systems. They pre-
sis methods such as game theory, decision the-
sented the applicability of general system the-
ory, linear and dynamic programming, prob-
ory for any kind of systems and suggested a
ability and statistics, information theory, sym-
universal language and laws that could be used
bolic logic in system analysis and optimiza-
in different areas with the objective of global
tion process. Ramo, Engstrom, and Schlager
acceptance. GST also engender the concept of
portrayed SE as a significant method to deal
systems thinking (ST) that facilitated higher
with challenges in identifying and satisfying
levels of cognitive skills to better understand
customer needs. The principle behind their
of the context of complex problems. Some of
proposition was that the integration of satis-
the GST objectives included:
factory components does not always produce a
- To formulate a theory that represents un- satisfactory system to achieve the desired goal.
derlying principles for all systems, irre- Engstrom (1957) provided a basic definition
spective of the context of the system. of SE writing that "This method is best de-
- To explore the identical principles, laws scribed by stating the two major requirements
and models in many disparate fields, and for its success: first, a determination of the
to aid successful transformation of these objective that is to be reached; and second,
axioms from one field to another, and as- a thorough consideration of all factors that
similate these understandings to avoid bear upon the possibility of reaching the ob-
unnecessary duplication and ambigui- jective, and the relationships among these fac-
ties between fields. tors(p.1)." Although Engstrom first introduced7
the concept of "interdisciplinary approach" in the need for a systemic holistic approach to
the SE process, he did not explicitly use the elicit the design requirement and necessary so-
phrase "interdisciplinary approach" but rather lutions for Boeing commercial aircraft.
coined the term "collaborative work." He am- The first book on SE was written by Goode
plified the idea of "interdisciplinary approach" and Machol in 1957 (Goode and Machol 1959)
by mentioning that a system project needs a and was titled Systems Engineering – An Intro-
wide range of expertise from disparate fields duction to the Design of Large-Scale Systems. This
so that the system can be adequately assessed book follows a theme that shows how systems
from different perspectives. thinking and approaches facilitate the design
of equipment. The overlap between manage-
Olthuis (1954) probably was one of the
ment and engineering was also acknowledged
early advocates, who introduced the idea of
by Goode and Machol in early 1959 when they
holistic perspective of top down approach to
wrote: "Management has a design and op-
design, emphasizing the need to draft the con-
eration function, as does engineering Goode
ceptual design of the entire system prior to ex-
and Machol (1959, p.514)." The commonal-
plicit details or knowledge of the constituent
ity and dissimilarity between the roles of SE
elements. For instance, most of the commu-
and project management have also been dis-
nications missile subsystems of military sys-
cussed in various publications, which will be
tems were designed from a holistic perspec-
discussed in the third interval (SE Revolution).
tive (Spanke 1954). Likewise, in the area of
A survey of the literature from (1926-1960)
acoustics, the necessity for a holistic approach
shows that: (1) World War II and several pre-
was recognized for the proper dissemination
war government projects had a significant im-
of acoustic energy in the audible space to have
pact on the inception of SE, (2) late in the 1950s,
a better performance of audio reproduction.
the focus toward holistic approaches to deal
By the same token, a holistic view of acous-
tic communication was also identified in the with increasingly complex systems and their
development of voice communication devices fundamental problems became apparent and
(3) several pervasive concepts pertaining to
for incorporating in aircraft system, where the
SE such as "system analysis" techniques, "sys-
all the necessary components and comminuca-
tems engineering process" and "system think-
tion channels were integrated together (Haw-
ing" were introduced. Figure 2 highlights the
ley 1956). In another case, the invention of jet
main themes in interval I.
aircraft challenging air traffic control systems
emerged in response to the need for complex 2.1.2 Interval II (1961-1989): Exploration of SE
system versatility (Kirshner 1956). This versa- In the 1960s-1990s, SE had significant growth
tility created a need for a holistic approach to along with widespread application. During
integrate ground to ground, ground to air, and this interval, the diversified characteristics of
air to air communication systems to enable a this discipline encountered some successes as
trouble-free air traffic channel. In this SE devel- well as failures and gave rise to debates based
opment interval, a number of articles (Speaks on subjective application of the discipline. Var-
1956, Okress et al. 1957) were published that ious aspects of SE and its process can be better
illustrated the necessity of considering the en- understood from the literature of Arthur Hall.
gineering work in a holistic technical manner In 1962, Hall introduced a concept of "sys-
(i.e., consider the technical environment of the tems engineering methodology" or "process
operating system as a whole instead of focus- of systems engineering" through three funda-
ing on particulars). Steiner (1959) described mental principles. First, SE definition is com-8
Figure 2 Main Themes for the 1926-1960 SE Development Timeline.
posed of diverse paradigms such as manage- nut emphasized that to explicate the problem,
ment technology, process-oriented approach systems requirements must be derived from
and problem-solving methodology. Second, the user specified need. While Shinners offered
to have a better understanding of complex sys- a set of seven general strategies in conjunction
tem problems, a systems engineer has to ap- with the concept of a feedback loop to explore
praise a system from three different perspec- a large complex system, Chestnut proposed an
tives: the physical or technical, the business optional feedback process to compare results
or economic, and the social (Gorod et al. 2008, being attained to meeting the customer’s re-
Hall 1962). Third, SE is designed specifically quirements.
to fulfill customer requirements in the most Jenkins (1969) provided a basic definition
effective way based on available information. of SE that somewhat refers to the system inte-
Hall’s SE methodology consists of five phases: gration or holistic perspective of a system. He
1) system studies or program planning; 2) ex- defined SE as "the science of designing com-
ploratory planning, which embodies problem plex systems in their totality to ensure that the
definition, determining the objectives, synthe- component subsystems making up the system
sizing and analyzing the system followed by are designed, fitted together, checked and op-
selecting the best system and communicating erated in the most efficient way." Jenkins ex-
the output; 3) development planning, which is plained that the SE approach deals with local
replications of phase 2 in a more comprehen- authorities, organizational norms, whole orga-
sive way; 4) studying the development, inte- nizations and hardware systems to weave to-
gration, and testing of the system; 5) current gether. His definition served as a grounded
engineering which refers to the operational ac- reference for further advancement regarding
tivities while the system is functioning and be- all aspects of SE.
ing refined (Buede and Miller 2016, p.7). In the 1970s, several SE theories and mod-
Shinners (1967) recommended that to solve els were introduced in the SE literature. Fol-
a system-oriented problem, a systems engi- lowing Von Bertalanffy’s work on GST, Ackoff
neer must grasp the fundamentals of the sys- (1971) opposed the idea of analyzing systems
tem problem, elicit the overall requirements by segregating the systems into sub-elements.
and objectives of the system, and understand Rather, he proposed that the entire system
the comprehensive knowledge concerning the should be treated as a whole. He asserted that
constraints inherent in the system. Shinner’s the interdependencies among the elements
problem formulation and solving methodol- within systems shoulde be considered aggre-
ogy are somewhat aligned with the earlier ad- gately. Thus, he concluded that reductionist-
vice recommended by Chestnut (1965). Chest- based approaches are not adequate for under-9
standing these overall interactions and inter- A year later, Chase (1974) emphasized the
dependencies. In addition, Ackoff addressed importance of the development of proper se-
several caveats and limitations in reductionist mantics and lexicology for the systems con-
approaches whenever they are applied to real cept. He asserted that language difficulties
life complex situations. Similarly, Beer (1972) might cause barriers to effectively communi-
introduced the term "meta-system" to desig- cate on topics pertaining to the system con-
nate the integration of systems by means of a cept and that work was needed in this area.
cybernetic perspective. He developed the A remarkable contribution came from Blan-
viable system model (VSM) which consisted chard and Fabrycky (1981) who introduced
of five main functions including the produc- the concept of "System Development Life Cy-
tive function, coordination function, operation cle (SLDC)." The concept is based on Hall’s
function, development function and identity (1962) methodology (problem identification;
function. Beer felt these functions were indis- problem definition; planning and designing of
pensable when ascertaining the viability (con- a system; construction and disposal).They de-
tinued existence) of a complex system, and that scribed the steps of system life-cycle as "start-
together they deliver a broad understanding of ing with the initial identification of a need
the mutual interdependencies among the ele- and encompassing the phases (or functions)
ments of the systems. The insights drawn from of: planning; research; design; production or
Beer’s concept provided a noteworthy contri- construction; evaluation; consumer use; field
bution to realize the structure of a complex support; and ultimate product phase out (Blan-
system. chard and Fabrycky 1981, p.19)." This concept
is still upheld as one of the underlying princi-
At the beginning of 1971, a series of ten
ples of SE.
lectures titled "Systems Concepts for the Pri-
vate and Public Sectors" was presented at the In 1974, The Defense Standard of the United
California Institute of Technology by several States (Military Standard), introduced the con-
scholars, with a primary purpose to criticize cept of "Systems Engineering Management
the many perspectives of the reductionist ap- Plan (SEMP)." They described SE as a practi-
proach(Ramo 1971). Ramo articulated that the cal use of scientific effort that incorporates all
systems approach focuses on analyzing and the "ilities" to meet the technical objectives of
designing a system from a holistic perspective the system. This observation can be mapped
while considering all possible parameters from into the management oriented paradigm. Ac-
both societal and technological standpoints cording to MILSTD499A (1974), SE is defined
rather than dealing with different individual as "engineering efforts to:(1) transform an op-
elements or parts. Miles (1971) stated that sys- erational need into a description of system per-
tem approaches work well when the objectives formance parameters and a system configu-
of the system are clearly defined and the nec- ration through the use of an iterative process
essary technologies are adequately developed. of definition, synthesis, analysis, design, test,
The lectures were later edited and published and evaluation; (2) integrate related technical
by Miles (1973). Miles identified the follow- parameters and ensure compatibility of all re-
ing steps needed for the systems approach: (1) lated, functional, and program interfaces in a
goal definition or problem statement (2) objec- manner that optimizes the total system defini-
tives and criteria development (3)systems syn- tion and design; (3) integrate reliability, main-
thesis (4) systems analysis (5)systems selection tainability, safety, survivability, human, and
(6)systems implementation(Brown 1953). other such factors into the total technical en-10
Figure 3 Main Themes for the 1961-1989 Timeline.
gineering effort to meet cost, schedule, and system methodologies and operation research
technical performance objectives (Buede and techniques can be applied. However, pluralis-
Miller 2016, p.9)." tic problems are more dynamic, uncertain and
complex in nature, and thus new techniques
Wymore (1976) indicated that an inter-
are needed. Clemson (1991) writings in the
disciplinary approach is an essential com-
same year underscored the importance of ex-
ponent of the SE discipline which is gov-
ploring complex system problems from differ-
erned by three fundamental attributes "mod-
ent standpoints that are mutually supportive
elling human behaviour, dealing with com-
to the axioms dervied from the cybernatics. In
plexity and largeness-of-scale, and managing
1986, Perrow (1984) made a contribution to
dynamic technology"(Wymore 1976, p.78).
the SE field by exploring the stochastic nature
Wymore also extended the application of SE by
of failure in large complex systems.
adding the education, health, and legislative
systems to the paradigm along with the exist- A survey of the literature within this in-
ing systems of communication and construc- terval (1961-1989) indicates that: (1) there was
tion(Checkland 1981). In 1984, M’Pherson a clearly recognized need and corresponding
(1986) brought another dimension to the shift in paradigms to holistic-based thinking
SE definition by proposing the term "hybrid and approaches to address complex system
methodology." He stated that SE is "a hybrid problems, (2) several definitions were pro-
methodology that combines policy analysis, posed that embodied numerous characteristics
design, and management. It aims to ensure of SE, (3) some fundamental models were de-
that a complex man-made system, selected veloped recognizing SE life-cycle and manage-
from the range of options on offer, is the one ment oriented concepts, and (4) several prob-
most likely to satisfy the owner’s objectives lem solving methodologies were developed to
in the context of long-term future operational address the SE problem domain. The timeline
or market environments" (IEEE P1220 1994, in Figure 3 shows the main themes in interval
p.130-133). II.
In 1984, Jackson and Keys (1984) made 2.1.3 Interval III (1990-2017): Revolution of SE
a notable contribution by classifying the This interval witnessed the widespread ad-
problem- solving methodologies of SE based vancement of SE. Several perspectives and con-
on unitary (pursuit of a definite set of objec- cepts were articulated, and the field was in full
tives) and pluralist (pursuit of multiple, poten- progress during this period. Many studies and
tially diversified goals) approaches. Unitary investigations tempted to synthesize the defi-
approaches are applicable for simple systems nitions of SE from different standpoints and
where the context of the problem is static and tried to establish the objectives of SE. Another
can be solved by a predetermined set of tech- stream of research focused on developing a
niques. For unitary problems, SE tools, hard SE body of knowledge encompassing different11
SE methodologies, unifying the systems the- tion of the consumers’ operational need, man-
ories, developing various models/processes agement processes coordinate different design
and building standardized frameworks. A sig- and configuration control groups and encom-
nificant number of presentations, conferences, passes handling risk, schedule, and budget as-
articles, symposium and journals pertaining to sociated with the task. Similar to Sage’s defini-
SE were also made available. To disseminate tion of SE, the Department of Defense used the
the SE principles and practices and to provide term "management" in their SE definition, but
better solutions to complex societal and techni- they also incorporated the concepts of "inter-
cal challenges, a non-profit organization, The disciplinary approach" and "life cycle process."
International Council on Systems Engineering Theme II: Requirement driven process
(INCOSE), was established in 1990. In 1998, a and SE process (life-cycle)
dedicated SE journal titled "Systems Engineer-
Forsberg and Mooz (1992) described SE
ing" started its proceedings to cover the full
as "The application of the system analysis and
spectrum of research germane to SE and Sys-
design process and the integration and verifi-
tem of Systems (SoS). The following themes
cation process to the logical sequence of the
can be derived for SE during this period:
technical aspect of the project life-cycle." In
- Management grounded technology 1994, Shenhar (1994) introduced the ideas of
- Requirement driven process and SE "management" and "interdisciplinary" in the
process(life-cycle) definition of SE. He mentioned that SE is a
- Interdisciplinary approaches technology oriented management process that
- Problem solving encompasses a sequential order of activities in-
Theme I: Management grounded technol- cluding: 1) identifying the customer need and
ogy convert it into system performance parameters
Although many works have been published and ultimate system design, 2) tracing and allo-
that brought about a sense of management cating the functional requirements, 3) selecting
technology in SE processes, Sage (1995) was the appropriate system concept and design, 4)
the first who explicitly incorporated the term integrating and testing the system architecture
"management technology" in the definition of and finally 5) evaluating the system’s perfor-
SE. Based on his definition "SE is the manage- mance. Another process-oriented SE defini-
ment technology that controls a total life-cycle tion came from Shishko (1995), who wrote
process, which involves and which results in that SE is "iterative" in nature. The iterative
the definition, development, and deployment nature assists in compensating for undesirable
of a system that is of high quality, trustwor- consequences and ensuring higher level quali-
thy, and cost effective in meeting user needs" ties of the system (Shishko 1995, p.4).’Iterative
(Sage 1995, p.3). His definition was based process’ is used in many SE definition (ECSS-E-
on three fundamental levels: SE management, 10-01 1996). Martin (1996) called SE a system
SE methodology, and SE methods and tools. development process that works to achieve op-
The three fundamental levels involved three timal system balance among all sub-elements.
key points: structure, objective and function. Skyttner (1996) defined "SE as a method by
Sailor (1990) stated that SE comprises which the orderly evolution of man-made sys-
both technical and management processes that tems can be achieved." Gardy (2000) described
transform the customer’s need into the desired SE as a process-oriented approach that trans-
system design. In distinction, whereas techni- forms a set of intricate technical needs into fea-
cal processes involve the systemic transforma- sible solutions via detail design and manufac-12
turing processes. In his work, Arnold (2000) isfied throughout the life-cycles of the sys-
mentioned that every organization must fol- tems." To support his argument, Wymore illus-
low a standard SE process and SE is tradition- trated the definition of SE discipline provided
ally associated with a single process, standard- by Kline (1995, p.3): "a discipline possesses a
ized objectives and a course of development specific area of study, a literature, and a work-
actions. A simple definition of SE came from ing community of paid scholars and/or paid
Hitchins (2003, p.309) "the art and science of practitioners." Hazelrigg (1996) provided a
creating systems." NASA handbook described more specific definition of systems engineering
SE as a decomposition (design), recomposition and introduced the term "information-based
(creation/integration) and operation of a sys- approach." He emphasized that mathematical
tem. intensity in the systems engineering approach
A somewhat different SE definition came fostered better decisions pertaining to system
from Hallam (2001) who used the term "pull design and synthesis. The general threads run-
process" and mentioned that SE is a customer ning through these definitions are that SE is
requirement driven "pull process" where a cus- a top-down approach that encompasses both
tomer demands influence the flow of system technical and managerial efforts to integrate
development activities. In the updated ver- the diversified processes to optimize system
sion of military standard handbook MIL-STD- performance. Additionally, SE is a require-
499B, SE was defined in terms of standard pro- ments driven process where a customer’s need
cesses, system analysis and control. Accord- is transferred into a requirements statement in
ing to MIL 499B systems engineering is an in- order to develop the fundamental attributes of
terdisciplinary approach including the set of a functional physical design.
technical endeavor to develop and verify an Theme III: Interdisciplinary approach
integrated set of system people, product, and Several other SE definitions developed in
process solutions in order to meet customer this interval that echo the theme of "interdis-
need." Kossiakoff et al. (2011, p.3) used the ciplinary approach." IEEE P1220 (1994, p.12)
term "guide" in his definition: "The function defined SE as “an interdisciplinary collabora-
of SE is to guide the engineering of complex tive approach to derive, evolve, and verify a
systems," where "to guide" means direct and life-cycle balanced system solution that satis-
lead towards achieving the best solution. This fies customer expectations and meets public
definition stresses the aim of SE as a process acceptability.” The Capability Maturity Model
of selecting the optimal solution out of many Integration (CMMI, 2001) described SE as an
possible alternatives. interdisciplinary collaborative approach that
Wymore brought a new terminology in the encompasses technical and managerial efforts
definition of SE. He defined SE as a "disci- to transfer the customer requirement into prod-
pline" instead of process. Wymore (1994, uct solutions. Jerome Lake asserted that "sys-
p.5) argued that SE is not only a process tems engineering is an interdisciplinary, com-
but also a distinctive discipline, where exist- prehensive approach to solving complex prob-
ing recognized SE processes are only applica- lems and satisfying stakeholder requirements
tions of the SE discipline. His definition in- (Martin 1997, p.244)." Abdallah et al. (2014)
cluded "the intellectual, academic, and pro- provided a more contemporary definition of
fessional discipline, the principal concern of SE mentioning that SE integrates all the dis-
which is to ensure that all requirements for ciplines to pursue a well-structured technical
bioware/hardware/software systems are sat- effort and governs design, development and13
verification of a system to satisfy the customer tem life-cycle. Activities are grouped based
need. Grasler and Yang (2014) also pointed on the corresponding vertical layer and hor-
out the attribute of an interdisciplinary ap- izontal life-cycle to represent the role of the
proach in the SE process to satisfy the stake- systems engineer. The third dimension is still
holder need. Shenhar (1994) added another under development which describes the prob-
layer to the definition of SE by including the lem solving activity.
concept of interdisciplinary approach, holis- Literature shows that there is an over-lap
tic perspective, and management process. SE and correlation between systems engineering
deals with identifying operational needs of processes (SEP) and the generic problem-
customers, forecasting operational and tech- solving processes. However, the set of activ-
nological processes, developing new concepts ities of SEP and problem-solving process are
and design by considering the overall system fairly distinct in nature. For instance, the steps
life cycle. Rechtin and Maier (2000) empha- involved in the generic problem-solving pro-
sized that there is close link between SE and cesses (OVAE 2005, GDRC 2009) are differ-
decision making, suggesting that SE is a mul- ent in contrast to the general SEP approaches
tidisciplinary design-oriented process where such as EIA 632 (1994), generic V-model and
decisions are made based on their impact on SIMILAR (Bahill and Gissing 1998). This
the system as a whole. A comprehensive def- common misunderstanding between problem-
inition of SE came from INCOSE. "Systems solving and SEP can be resolved by under-
Engineering is an interdisciplinary approach standing the SE emphasis on the holistic per-
and means to enable the realization of suc- spective of generating a human-made system
cessful systems. It focuses on defining cus- as a solution to a defined problem. A com-
tomer needs and required functionality early in mon meta-SEP can be developed by uniting the
the development cycle, documenting require- Hitchins (2007) and Mar B (2009) approaches
ments, then proceeding with design synthesis into the following 10-step sequence. This se-
and system validation while considering the quence combines the problem-solving process
complete problem." The thrust of this move- and the solution recognition process together.
ment was recognition of the interdisciplinary This 10-step sequence is feasible if we consider
nature of the SE approach. the systems engineering activity as a project
Theme IV: Problem solving (see Figure 4).
Kasser (2007) developed a framework that In 2005, Hitchins (2005) pointed out
clarifies the reasoning behind overlapping SE an interesting analogy between "soft system
and management and offered a concept for methodology (Checkland 1981)" and the gen-
planning fundamental problem-solving to off- eral problem-solving paradigm. For a bet-
set the challenges associated with a complex ter solution to the ill structured problem,
system. This framework also paved the way to Hitchins, in his model, combined two differ-
having a broader understanding of the SE body ent paradigms: exploration of initial problem and
of knowledge. The framework consists of three development of technological solution. The model
dimensions. The vertical dimension encom- consists a set of activities that addresses the
passes five layers: socioeconomic, industrial background of the problem and develops the
systems engineering, business systems engi- technological solution by considering the sys-
neering, project or system level and project or tems from a holistic perspective.
system level, whereas the horizontal dimen- Another contribution came from Vencel
sion signifies the sequential phases of the sys- and Cook (2005, p.8). They explore the14
Figure 4 10-step Problem Solving Process.
typology of complex system problems, de- For more in depth exploration of the
fined the entire problem space and catego- problem-solving approach, interested readers
rized it based on seven-dimensional problem are referred to study the nine-system model by
attributes: problem of interest, the nature of Kasser et al. (2014) , seven principles for sys-
the problem, level of the problem, phase of tems engineering solution system developed
the problem, problem complexity, structured- by NASA and summarized by Hitchins (2007,
ness and dynamicity (Vencel and Cook 2005, p.85). Another stream of research during this
p.8). The importance of identification of the period, focused on investigating the similar-
appropriate problem space also discussed in ities and dissimilarities between systems en-
the literature by Stevens et al. (1998). Flood gineering and project management. In many
and Jackson (1991) also made a noteworthy cases, Systems engineering and Project Man-
contribution through development of a sys- agement are considered to be different disci-
temic meta-methodology named as Total Sys- plines. Mooz and Forsberg (1997) recognized
tems Intervention (TSI). TSI directs the stake- some significant reasons for this distinction:
holder through a systemic process to select the
appropriate problem-solving procedure based - INCOSE expertise is concerned with
on the context and situation of the problem, technical solutions whereas PMI consul-
following through phases of creativity, choice, tants are oriented towards schedule and
and implementation. To address the formula- cost management. As a consequence,
tion of the problem, Ford (2010) proposed a project managers are more concerned
framework that traces the difference between about managing cost and schedule with-
subjective and objective complexity and cate- out taking into account the technical as-
gorizes the problem by pect while system specialists, who al-
ways pursue the superior feasible solu-
- Level of difficulty of the problem. (Easy, tions, rarely address budget and sched-
medium, ugly, and hard) ule.
- Structure of the problem. (well struc- - The nomenclature and terminology of
tured, ill structured, wicked) INCOSE and PMI are different.
- Level of complexity of the problem. (De- - INCOSE and PMI work autonomously
pends on the number of variables and and rarely participate in each other’s con-
the types of interdependency among the ferences. PMI members are seldom affil-
variables associated with the problem) iated with INCOSE and vice versa.15
Figure 5 Main Contributions of 1990-2017 Timeline.
Further discussion of the above arguments The GTC application was comprised of three
is illustrated by Roe (1995). He indicated that levels of coding : open coding (free form coding
tech specialists observe the systems from the of ideas), axial coding (clustering of codes into a
inside, and they are not concerned about other hierarchy of relationships), and selective coding
systems elements unless they affect their own (reformulation of coding into higher level core
design task. The project managers, on the categories) to derive the central theme from
other hand, consider the system from outside large unstructured dataset. It is also impera-
with a broader viewpoint acting as the advo- tive to mention that we collected the frequency
cate for the system. Project managers deal with of the publication from "Scopus" database by
all systems elements that would impact overall inserting input as "systems engineering" in the
system performance/budget/schedule. They search field and filtered the number of publica-
are also concerned about how to offset the tions based on the timeline. Scopus database is
constraints of system elements to ensure that more comprehensive than any other databases
projects reach their goals in an economic way as others include only ISI indexed documents
within stipulated time limits. However, in re- (Yong-Hak 2013). The Scopus database covers
ality, project management and systems engi- almost twelve million of different types of re-
neering are not independent disciplines. search documents from variety of publication
We have identified this interval as a "revo- houses.
lutionary interval", acknowledging that there
was significant generation of new concepts, 3. Histogram Analysis
approaches, frameworks and formal organiza- Figure 6 shows the histogram analysis of SE.
tions established with a view to disseminat- The horizontal axis in the histogram signifies
ing the knowledge of SE. Several applied fields the time line of the study, and the vertical axis
such as system of systems (SoS), and MBSE, displays the frequency of publications pertain-
also evolved during the revolutionary inter- ing to SE for that time period.
val. These fields are especially pertinent to It is evident from the histogram analysis
most engineering-governed approaches. The that the final interval (1990-2017) possesses the
main contributions of the 1990-2017 timeline highest frequency and the highest cumulative
are shown in Figure 5. value signifying that this interval experienced
The next section presents the histogram the peak of SE development. A larger number
analysis of SE development through three of presentations, conferences, journals, sym-
main intervals, with each interval represent- posiums, and research work related to SE was
ing a particular stream in the development tra- published in this interval. One of the most
jectory of SE. Following the histogram analysis, significant events was the establishment of the
based on the Ground Theory Coding (GTC) ap- International Council on Systems Engineering
proach, main characteristics of SE are derived. (INCOSE). INCOSE was founded in an effort16
Figure 6 Histogram Analysis of SE.
to unite the research germane to diversified the traditional engineering discipline to solve
branches of SE under the same umbrella and complex problems and moved towards more
to disseminate knowledge from the field of SE. holistic and integrated approaches.
Many universities and schools introduced sys-
3.1 Co-Citation Analysis
tems engineering into their academic curricu-
Co-citation analysis visualizes the relation-
lum as well. There will be many future oppor-
ships between sources/documents based on
tunities where the knowledge and information
their citations Barnett (2004). This biblio-
gained during this interval will be used to ex-
graphic coupling is conducted based on the
plore and solve various complex system chal-
graph theory (Saukko 2014). A co-citation
lenges.
map comprises of a set of nodes representing
The (1961-1989) interval is identified as an different research sources/documents (e.g., ar-
exploratory interval. This interval is consid- ticles, conferences papers, letters, and techni-
ered to be a transition from a discussion of cal report) and a set of edges signifying the co-
fundamental theories to the development of occurrence of nodes listed in different sources
real world applications, tools, processes and of the corresponding map (Barnett 2011).
approaches. The advancement resulting from More precisely, co-cited sources/documents
this interval set the foundation to support fur- appear together in the reference lists of other
ther development of systems engineering. The documents (Fahimnia et al. 2015).
concept of SE became the focus of attention In order to perform co-citation analysis,
and achieved widespread acceptability across .NET file contained of 278 sources was devel-
the world. The histogram shows that the fre- oped and imported in Gephi for the visual rep-
quency of publications increased in this inter- resentation. The visual output didn’t show any
val compared to interval 1; even though there is discernible pattern due to the random charac-
some fluctuation, a strong growth trend is still teristics of the coordinate. To better represent
obvious. The first interval (1926-1960) is recog- the map, we further ran a Fort Atlas driven
nized as the introductory interval of SE. In this algorithm and adjust the values of repulsion
interval, practitioners began thinking beyond strength, node size, gravity, speed, and otherYou can also read
