Stages of Systems Engineering: An Analysis and Characterization of Systems Engineering Approaches

Abstract

In the engineering of complex technical systems, Systems Engineering (SE) is a key approach that is becoming increasingly relevant in more and more industries due to the ever-increasing complexity of systems. Over the decades of practical application and research, various specializations and forms of the Systems Engineering approach have developed, but there has so far been a lack of an overarching context and positioning in meaningful stages for the introduction of Systems Engineering in companies. For this reason, this research will systematize common Systems Engineering approaches and bring them together in a stage model for Systems Engineering. Based on a systematic literature review, use cases are identified for each approach and stage, which support companies in selecting an approach suitable for their own organization.

1. Introduction

In recent decades, Systems Engineering (SE) has emerged as a foundational approach for the engineering of technical systems [1]. In view of the increasing complexity of products and the need to organize engineering processes in an efficient and targeted manner, numerous SE approaches have emerged that, for example, build bridges to other engineering approaches [1]. These include approaches, such as agile Systems Engineering [2], lean Systems Engineering [3], model-based Systems Engineering (MBSE) [4,5], data-driven Systems Engineering [6], among others. Each of these approaches offers individual, specialized tools and perspectives for overcoming specific challenges in the engineering of systems. However, this also leads to an increasing confusion about different approaches and their individual benefits. Organizations are therefore faced with the challenge of identifying a suitable approach for overcoming existing engineering challenges.

This research project aims at providing a comprehensive and systematic investigation, comparison, and categorization of the numerous Systems Engineering approaches that have emerged in recent decades. In light of the aforementioned considerations, a model is presented that defines four stages of Systems Engineering, thereby providing companies with a reference point for the introduction and optimization of existing SE implementation. Each of these stages represents an increasing degree of maturity and integration of SE approaches considering relevant processes, methods, tools and environment (PMTE [7]) within a company. The various SE approaches are not considered in isolation; rather, they are analyzed in a vertical strand across the levels in order to illustrate their degree of maturity and breadth of application.

Within this research, the following research questions (RQs) will be answered:

• RQ1: How can the various approaches of Systems Engineering, established in science and practice, be systematically categorized?
• RQ2: Which dependencies and further developments can be derived from the analysis of the approaches?
• RQ3: How can a model support companies in selecting a suitable Systems Engineering approach and in setting up a transformation program?

2. Scientific Approaches

This research is conducted based on the scientific, application-oriented research approach by Ulrich [8] in five steps, as shown in Figure 1. In the first step, a scoping review is conducted to identify related work. Based on this, the literature is reviewed to identify the broad variety of established Systems Engineering approaches. In the next step, meta-data analysis and bibliometrics as well as content analysis are proceeded to categorize the SE approaches. Based on these results, the model of four stages of Systems Engineering is developed. Ongoing practical use cases are described at each Systems Engineering stage. Finally, the model is discussed to identify strengths and weaknesses as well as further research fields.

3. Foundations of Systems Engineering

As previously stated in the introduction to this article, the objective of this research is to systematize and classify established Systems Engineering approaches in order to provide companies with a foundation for decision making on Systems Engineering Transformation. In order to establish a foundation for the classification of the various SE approaches, the following section on the state of the art initially presents a classification and understanding of Systems Engineering in general. This serves as the basis for the subsequent identification of approaches, which is conducted in the latter portion of the research. The second part of the state of the art presents the fundamentals of Systems Engineering Transformation and its relationship to organizational development. Finally, it also introduces the concept of maturity models of Systems Engineering, which serves to delineate the boundaries of the model developed in this research.

3.1. Systems Engineering

Systems Engineering is a transdisciplinary engineering approach that encompasses the technical and managerial facets of translating customer needs, including expectations and constraints, into a solution [9]. Furthermore, Systems Engineering encompasses not only the engineering process itself but also the engineered solution over its lifetime [9]. According to further definitions, Systems Engineering focuses on the engineering within a given time and cost frame [1,10].

Systems science and systems theory form the basis for the fundamental conceptual model of Systems Engineering [1]. Systems science examines systems in terms of their structures, functions, and dynamics [1,11,12,13,14]. Systems theory provides the necessary theoretical foundations to support this examination [15,16,17]. General principles that apply independently of domains, such as feedback loops, emergence, and hierarchies, are described [15]. Further systems theory can be described by a model of linked axioms [17,18]. This in-depth consideration and investigation of the system throughout its entire lifetime is established in Systems Engineering as systems thinking [19,20,21]. An essential point is the understanding of dynamics in complex environments [21]. Systems thinking encompasses both the system within the system boundary, along the system architecture across all levels and system elements, as well as across its own system boundary within the system-of-systems [10]. In order to gain an in-depth understanding of the system and its internal interactions, an overall system is decomposed into smaller subsystems, with the smallest relevant system element being the fundamental unit of analysis [1,22]. The requirements, functions, logical systems, and physical systems introduced by Baughey (2011) [23] have been extended to include validation and verification elements (RFLPV2). This extension is based on systems thinking and describes the relationships between the definition of requirements, the development of a comprehensive system architecture comprising functional, logical, and physical layers, and the processes of verification and validation [24]. Moreover, systems thinking necessitates the integration of operational considerations, including maintenance and subsequent decommissioning, at every stage of the engineering process [1].

In addition to systems thinking, engineering methodology plays a central role in Systems Engineering [10]. From one perspective, this includes the delineation of systems’ life cycle processes, as defined in ISO/IEC/IEEE 15288 [5], which provide orientation throughout the engineering process [1,10]. The V-model, initially comprised by Forsberg and Mooz (1991) [25], also provides a conceptual framework that structures the engineering process both visually and logically in terms of content [26,27,28]. Also, further specified models can be applied as conceptual frameworks, such as the Boeing Diamond model, offering a life cycle-oriented approach [29]. The content of the defined SE processes is supported by established methods and proven approaches for systematically solving engineering challenges [10]. The operationalization of these processes and methods is facilitated by the use of appropriate tools, including those for modeling, simulation, and project management [1,30,31].

Specific competencies and responsibilities of SE roles are delineated by SE role models, which also define areas of responsibility along SE processes [32]. Competency models, such as those developed by NASA [30], delineate the requisite technical and interpersonal skills. In addition to a generic SE role set (compare to Sheard [32]), further developments and specifications of Systems Engineering necessitate adjustments and additions to the role model [33].

In practice, there is a variety of SE approaches and handbooks that delineate the theoretical foundations and objectives of SE. The internationally accepted standard is defined in the standards of systems and software engineering, such as the ISO/IEC/IEEE 15288 [5]. Further practical examples include the handbooks by INCOSE [1], NASA [30], and the US Department of Defense [34]. In addition to these foundational handbooks, a variety of specializations and further developments of the SE approach have been published in research and practice. One of the most frequently discussed further developments is model-based Systems Engineering (MBSE). In model-based Systems Engineering, the document-centered approach is replaced by modeling and networking of models [1]. Individual partial models are integrated into an overarching system model, which is constructed with varying, subject-specific perspectives in mind [1]. Model-based Systems Engineering enables the mapping and analysis of continuous effect chains for the purposes of traceability [35].

Other approaches have been developed on the basis of interfaces to other engineering approaches. For example, agile Systems Engineering represents the bridge to agile engineering and is therefore used in particular at the interface to software engineering [1,2,36]. Similarly, lean Systems Engineering builds on the principles of lean engineering/lean production [1]. Further developments in Systems Engineering are seen by INCOSE and various research groups in performance support through artificial intelligence, for example [1,37]. These are referred to as AI4SE (Artificial Intelligence for Systems Engineering) in the INCOSE Vision 2035 [37].

3.2. Systems Engineering Transformation

The introduction of Systems Engineering into an organization means a fundamental change in the way it works, especially in the processes, methods, tools, and structures it uses. In the sense of Systems Engineering Transformation, established PTME must be critically examined and transformed in order to be aligned with the principles of Systems Engineering [38]. In the scientific literature, different understandings of Systems Engineering Transformation are proposed. While one understanding proposes the transformation of the Systems Engineering approach itself, the second understanding, which is the basis for this research, means the transformation of an organization’s engineering approach towards Systems Engineering [1], which is consistent with the understanding of the research field of organizational transformation and development.

In organizational development, human factors are of particular importance and a central component of established procedures and models, as organizations and change processes are significantly influenced by the people involved, their skills, attitudes, and behaviors [39]. For this reason, in addition to technological and structural adjustments, targeted change management is used in change processes to deal with resistance, promote acceptance, and enable a culture conducive to learning [40].

As in organizational development research, Systems Engineering Transformation authors point out the importance of structured change and transformation management based on a structured transformation approach [10,41,42]. Essential components of this approach are a clear definition of the transformation targets [39] and the tailoring of the identified Systems Engineering approach based on these targets [10,38]. The target definition requires a clear understanding of the existing engineering challenges and the technical expertise to select and tailor an appropriate SE approach. Based on this, the transformation is carried out with continuous monitoring of progress and target achievement [43]. Maturity models can be used to measure the implementation and transformation of Systems Engineering [44].

3.3. Maturity Models in Systems Engineering

Maturity models are used in various application areas to evaluate processes and progress and have been transferred to various approaches for use in Systems Engineering. A distinction can be made between maturity models for assessing process maturity and overarching models that claim to provide a holistic assessment of Systems Engineering implementation.

Several authors propose to use the Capability Maturity Model Integration (CMMI), originally developed for software engineering, for the assessment of process maturity [45]. In a comparison of three central maturity models, Sheard et al. [46] show the individual focal points of SE-CMM, SECM, and SECAM and clarify the focus on process maturity and effectiveness. Steffen et al. combine the process maturity models CMMI and SE-CM with success factors from SE standards and present a maturity model that focuses on action fields along technical and management processes [47].

Wilke et al. propose an overview of existing models and build a model for Systems Engineering maturity that considers not only process maturity but also structures and mindset aspects [44]. A similar approach is taken by Bretz, who concretizes organizational fields of action [41]. While the previously outlined approaches focus on the maturity of a specified SE approach in an organization, INCOSE [1] points out a further development of the applied SE approach in three steps: First, coming from no SE to the implementation of “full” SE; second, from the use of traditional SE to agile SE; and third, from document-based to model-based SE. Nevertheless, a categorization of the broad range of existing SE approaches within the further development of an applied SE approach is missing.

4. Identification of Systems Engineering Approaches

In the following, a scoping review is conducted to identify related work. Further, the broad range of literature is examined for discussed focal points and emerging trends. In the next step, the literature is reviewed to identify the variety of existing Systems Engineering approaches. The identified approaches are shortly characterized and categorized. The derived categories lay the foundation for the model of four stages of Systems Engineering.

In the first step, a scoping review is carried out according to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) [48] to identify relevant research work. Based on the consideration of fundamental Systems Engineering literature (especially INCOSE Handbook [1], Systems Engineering by Haberfellner et al. [22], NASA Systems Engineering Handbook [30], and Systems Engineering by Graessler and Oleff [10]), two search strings are defined and supplemented by ‘review’, as proposed by the PRISMA working group [48]. The search is carried out in the online databases Scopus, Web of Science, IEEE explore, Science Direct, and Google Scholar. Filters are set to restrict the search to the domain ‘Engineering’ and the type of article as ‘Review Article’. In Google Scholar, the top 300 results are included. A total of 1038 articles are included. The scoping review reveals that no research has yet been published on the systematization and categorization of existing Systems Engineering focal points in the context of Systems Engineering Transformation or introduction. However, isolated approaches provide an overview in specific domains or focal points. Yang et al. present a state-of-the-art review focusing on ontology-based Systems Engineering [49]. In addition, several approaches focus on specific industries, such as the public sector.

For that reason, in a first step, the broad field is researched to identify existing intellectual structures and trends emerging in the field of Systems Engineering implementation. According to Donthu et al., a broad search string is defined to proceed a bibliometric analysis [50]. The broadly defined search string is applied/used in different scientific online databases (see

Table 1. Search strings applied for bibliometric analysis.

Search String	Web of Science	Scopus
Applied filters:		Subject area: Engineering
“systems engineering” AND “Implementation” OR “Integration” OR “approach” OR “Introduction”	14,782 reports unconsidered due to limited number of exports	20,006 reports first 20,000 considered due to export limitations

). The results are exported and analyzed using the software VOSviewer (version 1.6.20) [51].

Due to the export limitations of Web of Science, the research results are limited to the SCOPUS database. During the analysis of the results of the conducted search, all keywords are extracted and grouped with duplicates and synonyms. To ensure the clarity of the clusters, a random examination of publications is conducted for applied keywords. This qualitative analysis yields insights that guide the subsequent reduction in the keywords to the 40 most frequently used (excluding “systems engineering”), which are further considered and clustered. Based on the resulting graph of conjunctions between keywords used in related research work, content-related clusters can be identified and interpreted (see Figure 2).

In Figure 2, five clusters become visible: The red cluster focuses on the fundamentals of Systems Engineering, such as Systems Engineering theory, architecture, integration, requirements, and systems thinking. The green cluster enhances the research work on model-based Systems Engineering, focusing on sysML, for example. In addition, the yellow cluster complements the green cluster about the keyword modeling, simulation, and optimization. The blue cluster contains keywords related to artificial intelligence, such as prognostics, deep learning, and machine learning. In a fifth cluster, aspects of human factors in Systems Engineering are contained. Although human factors, such as social, emotional, cultural, or political factors, are discussed in many ways in the context of organizational development, according to this research, they are isolated and rarely considered in the Systems Engineering community.

In the next step, the results are analyzed on a time-related view (see Figure 3). A time shift in terms of importance of keywords becomes visible. While fundamental Systems Engineering aspects have been the focus of research since the 1960s, between 2015 and 2018, model-based Systems Engineering including sysML became an important research field. Since 2018, aspects of artificial intelligence, machine learning, and internet of things have become the focus of research. Based on the analysis of meta-data and bibliometrics, the following clusters of Systems Engineering research fields, highlighted in Figure 3 and explained in

Table 2. Description of the identified research fields based on meta-data analysis.

Research Field	Description
fundamental Systems Engineering	Fundamental principles and values of Systems Engineering are discussed, including key enablers, such as systems thinking, requirements engineering, architecture design, integration, and V&V practices.
model-based Systems Engineering	The fundamental Systems Engineering principles and activities are enhanced by modeling aspects. Modeling enables new opportunities in terms of effect chains for security and safety reasons.
AI-based Systems Engineering	The internet of things becomes consciously important as system-of-systems. Potentials of artificial intelligence including machine and deep learning are discussed for performance support in the engineering of highly complex systems.
human-centered Systems Engineering	Human factors become important in the design and engineering of systems.

, are derived.

In the subsequent stage of the analysis, a particular emphasis is placed on the examination of industrial application and Systems Engineering Transformation within organizational contexts. To this end, two distinct search strings are defined based on findings from prior analyses and seminal studies in the field. The research documented in

Table 3. Search strings explored to identify research of Systems Engineering implementation.

ID	Search String		WoS	Wiley	IEEE Explore	Scopus	G. Scholar
	Applied filters:		-
#1	“systems engineering transformation”	total	2	0	2	5	145
		incl.	0	0	1	1	6
#2	(“systems engineering implementation” OR “implementation of systems engineering” OR “systems engineering adoption” OR “adoption of systems engineering” OR “systems engineering introduction” OR “introduction of systems engineering” OR “systems engineering application” OR “application of systems engineering”) AND (“Culture” OR “Embedding” OR “Cultural” OR “Anchoring”)	total	34	258	13	14	1.830
		incl.	0	3	1	0	18

is further supported by the PRISMA guidelines. In the initial phase, a total of 2303 records are identified from five scientific online databases. This initial set is subjected to a process of elimination, whereby duplicates are removed, and resources that are either inaccessible or in a foreign language are excluded from further consideration. Further, the records are screened based on title, authors, keywords, and abstract. Finally, 30 records are included in the review.

The analysis shows the importance of internalization in Systems Engineering Transformation. In general, a separation between two wordings becomes visible: applied and implemented (e.g., through implementation of processed and methods [52]) vs. established (e.g., approach is anchored in the organizations culture [52,53]). Terry et al. point out that technical skills are not enough and a deep behavioral understanding resulting in the further development from a communicated approach towards an organizational mindset is important [52]. Whitcomb et al. address the importance of a cultural change to achieve success in transformation [54]. Based on this separation identified in the literature, a separation in two stages of Systems Engineering implementation is derived (see

Table 4. Derived stages of SE implementation.

Stage of SE Implementation	Description
applied	Systems Engineering processes and methods are implemented in the organization. First projects are conducted using Systems Engineering as engineering approach [52,55,56,57].
established	Systems Engineering is internalized by the organization, including a shift in corporate culture focusing on the principles of Systems Engineering (sometimes called “holistic introduction”) [42,52,53,54,55,56,57].

).

In the next step, research is conducted based on the fundamental literature to identify the variety of established Systems Engineering approaches. On this basis, the forward-directed literature research is conducted to identify additional approaches. As INCOSE points out, there is a broad range of existing approaches with different understandings and levels of abstractions [1]. Therefore, the following lists (see

Table 5. Frequently discussed specialized Systems Engineering approaches in the community.

Systems Engineering Approach	Key Value	Reference
(fundamental/traditional) Systems Engineering	- handling systems complexity due to a systematic approach of interdisciplinary engineering - achieving an interdisciplinary engineering optimum within a predefined time and cost frame	[1,5,22,30,34]
Agile Systems Engineering	principle-based approach for dynamic environment and uncertain knowledge	[1,2,22]
Artificial Intelligence for Systems Engineering (AI4SE)/AI-based Systems Engineering	consideration of technical advantages of AI and machine learning to enhance engineering results and processes	[1,37]
Digital (Systems) Engineering	using key technologies and achievements as enabler in Systems Engineering, such as industry 4.0, digitalization, data science, AI, IoT, etc.	[34,58]
Human-centered Systems Engineering (HCSE)/Human Factors Engineering (HFE)	- focuses on human factors in engineering and product usage - supports communication and empathy in engineering teams	[1,37,59,60]
Lean Systems Engineering	- applying lean thinking and principles in Systems Engineering - reducing waste in engineering	[1,61]
Loss-driven Systems Engineering (LDSE)	- consideration of losses through the whole life cycle - focus on concerns in fields like reliability, availability, maintainability, safety, etc.	[1,62]
Model-based Systems Engineering (MBSE)	- modeling in a formalized manner - formalization of SE	[1,22,34,37]
Model-based Systems and Software Engineering (MBSSE)	coupled consideration of systems and software in a digital environment	[5]
Product Line Engineering/Systems Family Engineering/Product Systems Engineering (PSE)	- consideration of product lines containing multiple variants and different markets - handling mismatch of models, tools, etc., for product varieties	[1,5]
Software Systems Engineering (SwSE)	design and implementation of complex software systems	[5,34,63]
System-of-Systems Engineering (SoSE)	engineering of a systems environment for independent systems to achieve an overarching value	[63,64]

and

Table 6. Further in the literature review, specialized Systems Engineering approaches are identified.

Systems Engineering Approach	Key Value	Reference
Data-Driven Systems Engineering	engineering based on the foundation of engineering data with associated structure and inter-relations	[6]
Domain-specific Systems Engineering	domain-specific approaches, e.g., for health sector, open sector, enterprise level (Enterprise SE), etc.	[65]
Evidence-based Systems Engineering	evidence-based methods supporting decision making in SE	[66]
Function-based Systems Engineering	detailed focus on functional modeling in extension of MBSE	[67]
Object-oriented Systems Engineering	systems structure and modeling on basis of objects and entities	[68]
Pattern-based Systems Engineering (PBSE)	using system patterns in combination with a systems meta-model	[69]
Sustainable (Systems) Engineering	supporting considerations of circular economy in the engineering of systems	[1]
System Safety Engineering	focus of reducing likelihood of harming people or environment	[1]
Systems Security Engineering	consideration of anomalous events during systems usage, e.g., due to cyber attacks	[1,34]
(Product-)Service Systems Engineering (PSSE/SSE)	multidisciplinary approach to address service systems from life cycle and customer perspective	[70,71]
Resilient (Systems) Engineering	design of robust systems against adversities and uncertainties	[1]
Requirement-driven Systems Engineering (RDSE)	constraints of, e.g., production or resources are considered as main criteria in the design of systems	[72]

) are a comprehensive collection of existing approaches identified by the fundamental literature of Systems Engineering but does not claim to be exhaustive. The basic hypothesis of the identified Systems Engineering approaches is that they extend the fundamental principles of (traditional) Systems Engineering (according to INCOSE [1] and ISO/IEC/IEEE 15288 [5]) by additional scopes of consideration. For this reason, the following is not an evaluation of the content of the approaches but merely an overview with the core ideas of the content extension.

5. Model of Five Stages of Systems Engineering

Based on the analysis described in the previous section and an allocation of the identified SE approaches, four plus one stages of Systems Engineering are derived. In the following (see

Table 7. Derived stages of Systems Engineering including short descriptions and allocated approaches.

Stage	Description	Implementation Depth	Examples
initial	Initial established engineering approaches may contain aspects that are aligned with Systems Engineering, e.g., partial requirements engineering.	-	component-based engineering function-based engineering simultaneous engineering value-based engineering
applied	Systems Engineering processes and methods are implemented in the organization. First (pilot) projects are successfully conducted using Systems Engineering as engineering approach.	successful in single (pilot) projects	Systems Engineering(“traditional”) agile Systems Engineering lean Systems Engineering function-based SE
formalized	The implementation and application of the fundamental Systems Engineering approach is supported by a standardized described and evaluable form. This can be implemented, for example, in the form of formalized models or data sets.	successful in single (pilot) projects	model-based Systems Engineering data-driven Systems Engineering pattern-based Systems Engineering
established	Systems Engineering is formalized and internalized organization wide, including a shift in corporate culture focusing on the principles of Systems Engineering.	organization-wide implementation
performance supported	Systems Engineering is enhanced by performance support due to means of artificial intelligence, e.g., measures and tools are used to increase the effectiveness of engineering in two ways: better engineering and better results of engineering.	in all established SE units	digital Systems Engineering AI4SE

and Figure 4), the four plus one stages are first described with their characteristics, and second, the allocation of SE approaches is presented.

5.1. Stage One—Applied Systems Engineering

The first stage describes the initial implementation of Systems Engineering in an organization to solve engineering challenges in specific areas. Within this stage, organizations start defining an engineering target and tailoring a fitting Systems Engineering approach. Ongoing necessary processes, methods, tools, and structures are implemented. In this stage, an organization-wide consistent PMTE integration does not have to be ensured. It is also suitable that partial previously applied procedures are used in limited range, for example, due to time and budget restrictions. Through the application of Systems Engineering in a limited range, a short-term achievement of engineering success shall be reached to achieve a foundation of further organization-wide establishing of Systems Engineering.

5.2. Stage Two—Formalized Systems Engineering

The second stage describes the formalization of Systems Engineering through, for example, modeling or data structuring. This requires complete standardization and documentation of applied procedures, processes, tools, and methods. Engineering artifacts are created and inter-related in a formalized manner to enable uniformity and ongoing analytical processes. The formalized application of Systems Engineering enables engineering process improvement as well as engineering result improvement through, e.g., achieved traceability.

5.3. Stage Three—Established Systems Engineering

The third stage describes the establishing and internalization of Systems Engineering within an organization. In this stage, Systems Engineering is the fundamental engineering approach on which new engineering projects and orders are built. Systems Engineering processes and methods are standardized, documented, and internalized by the employees. The organization’s tool landscape and organizational structures are aligned according to the needs of Systems Engineering. This includes that Systems Engineering is a fundamental part of the organization employee’s development and training concept. All in all, the organization’s culture is aligned with the Systems Engineering principles—systems thinking is internalized in the organization.

5.4. Stage Four—Performance-Supported Systems Engineering

In the fourth stage, advances in technology (e.g., artificial intelligence, augmented and virtual reality, communication techniques, etc.) are adapted to reach both better engineering and better results of engineering [73]. For example, artificial intelligence and data science can be used to improve Systems Engineering practices and to achieve process improvement in the engineering of very complicated systems. On the other hand, these techniques can be used to achieve better engineering output due to, for example, automatically generated and optimized building structures or AI-based requirement definition and optimization [74]. As can be seen from the INCOSE Vision 2035 [37] and the literature research conducted in this paper, performance support in Systems Engineering is a prominent field of research but still lacks broad industrial application.

5.5. Enhanching the Systems Engineering Scope Through Specified Approaches

Within the conducted literature research, a variety of different specializations and further developments of the Systems Engineering approach have been studied. In accordance with the previously derived stages, the identified Systems Engineering approaches are categorized based on the following principles:

(1)

Does the approach extend the fundamental Systems Engineering approach in terms of additional views and specifications?

(2)

Does the approach enable the formalization of Systems Engineering?

(3)

Does the approach focus on performance support due to technologies?

Considering the analysis conducted, the subsequent evaluation and categorization (based on the principles presented above), and the resulting findings, the following assumption is taken: The identified and presented approaches are based on the principles of Systems Engineering. The implementation of each approach thus necessitates a certain degree of the fundamental Systems Engineering principles (according to Systems Engineering fundamental handbooks [1,22,30,34]). Furthermore, additional scope for specialization, such as agile or lean Systems Engineering, is included. Figure 5 shows the approaches identified based on principle 1. These approaches give an additional scope in terms of Systems Engineering implementation, resulting in additional implementation effort (e.g., due to additional PMTE) throughout all Systems Engineering stages. From an applicational point of view, organizations must define which contributions of particular SE specializations are required for solving the engineering tasks and have to check its appropriateness. This becomes especially relevant when considering additional implementation effort of these specifications. As visualized in the following, these approaches are not allocated to a specific stage; rather, special features of the approaches crystallize particularly in later phases of application and accompany all stages.

When considering stage model as a pyramid, the implementation effort becomes visually apparent. This model is predicated on the assumption that the initial effort is the greatest and that the additional effort decreases as the transformation progresses (vertical progress in the model). For instance, fundamental principles must be developed and implemented at the initiation of the implementation of Systems Engineering. This necessitates a substantial effort in the definition, piloting, and operationalization of PMTE, for example. As the transformation progresses, smaller transformation scopes are added, which build on the previous ones. However, special attention must be paid to the additional effort required for the specialization of the SE approach. Figure 6 shows the basic model visually. When further dimensions are considered, the transformation effort visually increases in volume.

In addition to the different kinds of characterization of Systems Engineering, the formalization of SE opens a third dimension in Systems Engineering implementation (see Figure 7). While different characterizations of Systems Engineering set specific focal points, the third dimension describes to what extent the formalization is applied.

The formalization of Systems Engineering, for example, through modeling, is based on the applied characterization of Systems Engineering and introduces specific formalization aspects to be specifically developed. An example is the formalization of agile Systems Engineering due to model-based Systems Engineering: MBSE fundamentals have to be implemented and specified to the fundamental Systems Engineering approach and in addition to specialties emerging due to agile aspects of the approach (see Figure 8). As became visible through the literature study, the identified approaches give special focuses considering MBSE as a basis. The following approaches have been identified as formalized approaches:

Model-based Systems Engineering (MBSE);

◦

Data-driven Systems Engineering (DDSE);

◦

Object-oriented Systems Engineering (OOSE);

◦

Pattern-based Systems Engineering (PBSE);

◦

Function-based Systems Engineering (FBSE).

Performance support in the engineering processes generally requires a formalized and thus evaluable basis. In the current literature, on the one hand, “digital (Systems) Engineering” is the transfer of achievements through digitalization (e.g., industry 4.0, data science, etc.); on the other hand, “AI4SE” [75] includes established approaches. While AI4SE can be understood as extension of the (traditional) Systems Engineering approach in the field of performance support [73,74,75], digital Systems Engineering can be interpreted as emerging influences. Therefore, established Systems Engineering implementations are enhanced and supported by different technologies, such as data science, artificial intelligence, machine learning, or deep learning (see Figure 9).

6. Illustration of Use Cases in Industrial Practice

For further foundation, the identified stages of Systems Engineering are illustrated by use cases derived from research and industrial practice.

6.1. Introduction of Systems Engineering (Stage 1)

During a project at a German car manufacturer, which has been accompanied by the authors for 2.5 years, Systems Engineering was piloted to tackle emerging challenges in the engineering of future vehicle systems. Based on a formerly established function-based engineering approach, the Systems Engineering principles had been tailored for its application in the organization. Fundamental Systems Engineering works of INCOSE laid the foundation for a company-specific tailoring. In addition, agile Systems Engineering had been piloted for better synchronization in interfaces to software engineering units. Therefore, fundamental principles of agile Systems Engineering have to be defined and implemented in the organization [36].

Within this project, the fundamental PMTE of agile Systems Engineering was developed in a centralized department and further applied in a pilot project based on an ongoing engineering order. For example, based on a Systems Engineering role model, the structural organization had been built on and roles were staffed by content-specific experts from different units. Within the tailoring of the role model, a responsibility assignment took place to identify relevant roles [33]. Also, processes were defined and implemented: According to the RFLP scheme [24], the requirements engineering and processes were applied, as an example. Overall, the V-model was applied as a fundamental engineering model, which offers several potentials as in the engineering, such as its wide range of target groups and the consideration of the whole product life cycle, which is consistent with the fundamental Systems Engineering principles [76].

6.2. Effect Chain Modeling in Automotive Industry (Stage 2)

In the automotive industry, standards and regulations, such as the ISO 26262, UN ECE (such as R 156 and R 21), or A-Spice, require a high degree of formalization to enable traceability [35]. Therefore, car manufacturers are forced to model effect chains to ensure compliance with regulations and standards. At a German car manufacturer, different construction series of premium passenger vehicles were modeled in sysML-based effect chains to ensure these [10,77]. As an example, a methodology for model-based effect chain analysis was developed and applied in the industrial field [35,77]. As becomes visible in the V-model of VDI 2206, modeling and analysis are important processes that go along the whole engineering process [78].

6.3. Established Systems Engineering at NASA (Stage 3)

Systems Engineering has been established in aerospace for decades. A prominent example of successful Systems Engineering implementation is the Apollo space program [79]. As a study from 2004 shows, Systems Engineering is highly encouraged at all levels at NASA and repeatedly leads to project success [80]. The 2008 NASA Systems Engineering Behavior Study [81] points to the success of Systems Engineering through hands-on learning and the establishment of a common understanding and SE mindset. Furthermore, the study shows that there are different approaches and cultures within the large organizations, but they are all based on a common foundation and behaviors [81]. All in all, several studies and reviews show that NASA has established a deep fundamental Systems Engineering culture that lays the foundation for project success and further process improvements in terms of achieving formalization of Systems Engineering [82]. As shown in research work, Systems Engineering may require cultural changes to enable a fundamental establishment of Systems Engineering [83].

6.4. Performance Support in Product Creation (Stage 4)

As INCOSE points out in the Systems Engineering Vision 2035, artificial intelligence (AI) will have several impacts on Systems Engineering and its application [37]. On the one hand, AI will have impacts on the engineering practices in the engineering of systems [73]. On the other hand, AI enables Systems Engineering practices and tools: Engineering in the automotive industry requires high compliance with standards and regulations, resulting in a high effort in requirements engineering. Based on AI-based requirements extraction, resulting requirements from regulations, for instance, can be derived and formalized in a standardized pattern [74]. As different research points out, especially routinized work in a large scale can be efficiently solved by AI [73]. Therefore, different AI approaches offer different potentials (compare to [73]).

Various research programs focus on tapping the potential of artificial intelligence in product creation, such as the DFG SPP 2443 focusing on the potential of hybrid decision support in product creation. The aim of the priority program is to supplement established engineering methods with data-driven approaches in order to be able to make faster and better engineering decisions.

7. Discussion

The stage model of Systems Engineering consolidates the wide variety of Systems Engineering approaches and specializations that are included and established in research and practice. While common Systems Engineering approaches focus on a specific characterization, a broad structure of these approaches and their relationships was missing, which is given with the comprised model. Therefore, a bibliometric analysis was performed to identify representative clusters of Systems Engineering approaches based on common research areas. Finally, the model comprises four stages in three dimensions, considering a separation between approaches of specific characterizations of Systems Engineering and approaches focused on the formalization of Systems Engineering (see Figure 6). In addition, use cases are derived from practice and the literature according to the stages.

In the first step, a bibliometric analysis is carried out using a very broad search string resulting in more than 20,000 results. Despite the set restrictions, included results can only be randomly checked for their actual relevance, and it can, therefore, be assumed that the result contains a certain degree of imprecision. Nevertheless, the subsequent research on Systems Engineering approaches and the analysis of the relevant literature show the plausibility of the identified clusters. In further research, this research could be sharpened using the PRISMA scheme, and deviations from the clusters could be investigated.

In this research, first, specializations and approaches of Systems Engineering are identified by broad literature research. As a foundation of this research the Systems Engineering domain has been reviewed. Further, other established approaches from different domains could be reviewed and analyzed in order to discuss their relationship to Systems Engineering. As an example, approaches prominent in the software domain could be considered, like DevOps, CI/CD (continuous integration/continuous development), or SDV (software-driven vehicle). Therefore, further analysis of the demands of different sectors and established approaches may be required.

The model proposed provides transformation leaders an orientation regarding the content-related localization of different approaches and focal points. The model is based on the hypothesis that fundamental Systems Engineering constitutes a foundation, which is expanded by specialized views and approaches as required. This contributes an extended scope of content to the transformation, as additional processes must be defined and implemented or additional tools must be used, for instance. The pyramid model also makes the effort required to reach the next stage visually visible. The responsibility of the transformation leader is to delineate the specific scope and required stage to overcome existing challenges in the affected organization. This determination needs to consider a cost–benefit analysis of the expenses for additional scopes of transformation and the added value achieved through additional approaches and stages. Within this research, the conceptional basis for these determinations are proposed, but more detailed guidelines, including checklists, for instance, are missing for operational application.

Within the model, there are four stages (applied, formalized, established, and performance supported). On the one hand, the order of these stages can be discussed, e.g., if the “applied” stage is a relevant prerequisite for the implementation of a formalized approach. Therefore, the necessity of a complete fulfillment of a stage and an understanding of a sequential model must be distinguished. Further discussions could be laid on the sequence of ‘formalized’ and ‘established’: Due to the formalization of Systems Engineering, lots of effort is invested in structuring and documenting PTME before these are operationalized and internalized in the organization. Early formalization without prior broad establishment can lead to PMTE not being sufficiently adapted to the organization’s needs, resulting in less acceptance of the organization. On the other hand, kinds of formalization, such as modeling, offer useful first steps into Systems Engineering.

The fourth stage of the model addresses performance support in Systems Engineering through technological advancements. By incorporating initiatives of the Systems Engineering community and the relevant literature, the model examines topics related to data science and artificial intelligence, aligning with contemporary buzzwords prevalent in various contexts. The literature review indicates that procedures of data science and artificial intelligence can facilitate performance support; however, technologies that extend beyond these procedures can also lead to performance support and improvement. In this model, however, the fact of performance support is presented independently of the actual implementation. Consequently, further research endeavors must entail a more profound problem analysis and systematic evaluation of technologies to facilitate opportunities of performance support in Systems Engineering.

According to the argument of a sequential logic of the model, the model may apply that formalization is required for full establishment of Systems Engineering in the organization. Also, even if formalization is certainly an important basis for the application of AI- and DS-supported methods, it is debatable whether a complete establishment must have taken place in advance or whether these methods can also be used early on to enable Systems Engineering.

The applicability of the stages of this model is illustrated by use cases from industrial practice. The first use case reflects the first stage based on Systems Engineering application in the automotive industry. The increasing interdisciplinarity in the engineering of vehicles is tackled by implanting Systems Engineering as an interdisciplinary engineering approach. The second use case reflects the need of formalization because of increasing regulatory demands for traceability. Even though the window lifter discussed in the referenced project seems to be a simple use case, interconnections and the consideration in the overall systems require formalization in the engineering process. The third use case is reflected by existing Systems Engineering cultures in industry, which verifiably lead to higher project success. The fourth use case is mainly illustrated on a state-of-the-art use case, referencing research projects. In the literature research, it became visible that there is a gap between state-of-the-art and industrial applications. For that reason, reliable insights from industrial practice are missing, even though AI tools are available on the market. Further validation of this model requires expert interviews, for instance.

Even though the model presented feels like a maturity model in parts and shows parallels to maturity models like hI, further differentiation is required. This model does not determine the capabilities or maturity levels of an organization. Similarly, no assessment specifications are proposed for locating an organization in the proposed stages. This stage model brings established approaches and specializations of Systems Engineering into an order and helps organizations identify an approach with the necessary specialization to solve engineering challenges. In contrast to CMMI models, for example, there is no assessment of an organization’s process maturity or learning ability. Further, fundamentals of developing maturity models should be considered, such as De Bruin et al. [84], to develop a comprehensive maturity model on the results of this research work. Therefore, organizations’ ability and progress of the process adaption, considering international standards, such as ISO/IEC/IEEE 15288 and 12207 [5,9], could be assessed.

8. Conclusions

The stage model of Systems Engineering comprises a solution for the identified research gap. As discussed above, the derived model offers a comprehensive, systematic view on the variety of Systems Engineering approaches and comprises four stages of Systems Engineering implementation in an organization. These stages are derived based on categories identified through biometrical analysis (RQ1). As the research shows, the identified approaches extend the fundamentals of Systems Engineering for specific characterizations (RQ2). It became visible that the approaches are not mutually exclusive, but can be combined with each other, which leads to additional implementation effort. Formalizing approaches, in particular, extend the fundamentally implemented approach with measures for formalization. Finally, the proposed model offers companies an orientation of existing approaches and their relation to each other (RQ3). Based on the stages defined, a stage level, necessary to reach the engineering target, can be identified, and transformation initiatives can be aligned according to this. By defining the scope of transformation, necessary interim steps can be set.

Even though the order of the stages proposed opens room for scientific discussion, this model offers a first order and systematization of established approaches laying the foundation for further refinement. In further research, the model should be delimited against maturity models of Systems Engineering implementation. Further, the stages can be more precisely specified or completed by checkpoints. Based on criteria of fulfillment strategies for specific transformation, initiatives can be derived or an assessment scheme can be defined.