Factors Affecting Systems Engineering Complexity during Developmental Phase: Systems Practitioners, Developers, and Researchers’ Perspectives-Systematic Review

The Systems Engineering design process is challenged to deliver successful complex systems in multidisciplinary and heterogeneous components. Growing human needs and evolving society bring ever greater challenges in the formation of a complex and large engineered system. System complexity is related to lots of parts and large size of the system when there is difficulty in understanding how the system works or in predicting the consequences of any change that may affect the process and systems develop itself. The leadership is performing an important role to manage a complex system. Leaders should be able to set back from immediate focus and look at the desired big picture. In practice, many factors contribute to Systems Engineering complexity in this review. This study intends to explore and analyze the complexities and the factors that contribute to the complexity of the Systems Engineering design approach. The data in this study were collected systematically from several electronic scholarly databases, including the ISI Web of Science, Scopus, Wiley Online Library. This study quantified the challenges and causes of the Systems Engineer complexity. Then, the challenges were categorized into two groups, managerial and technical causes. Ultimately, seven Systems Engineering complexity factors were identified, and their impact on the Systems Engineering processes was ranked using the Pareto principle. Among the factors, rapidly emerging technology was the most significant factor contributing to Systems Engineering complexity.

Structure diagram terms variation of searching Table 1 represents the initial search script and results of the literature review studies in the field of Systems Engineering Complexities in each database.

SCREENING -INCLUSION AND EXCLUSION CRITERIA
The population includes published articles on the Systems Engineering approach, which could be technology-based or management-based, including related journal articles, technical publications, standards, and special papers.
The Endnote references management tool has been used to manage all compiled lists of references. Fig. 2 demonstrates the flow diagram of identification, screening, eligibility, and included studies. 362 documents from the complied list at identification level were filtered to exclude theses, conference papers, book chapters, and magazines. 25 references were excluded due to duplications. Finally, one reference was removed because it was not in English. About 31% of results (178 titles and abstracts) reached the eligibility level. A systematic review quality was made to exclude non-relevant topics.   Website : www.ijirmps.org Email : editor@ijirmps.org 46

III. COMPLEXITY AND SYSTEMS ENGINEERING
This section highlights the definition of the term complexity and determines the factors that influence the system engineering process based on the data extracted from the systematic review. Fig. 3 shows the flowcharts used to achieve the research objectives.  Figure 3 Flowchart of the study steps to achieve the research objectives

COMPLEXITY
Wildly divergent definitions of "complexity" have emerged; it has no singular meaning [8,9]. Applying such a broad topic as complexity to the equally common topic of Systems Engineering is extremely difficult [10]. However, this study introduces an operational definition of complexity1,2 regarding Systems Engineering and its life-cycle. Hall [11] stated that Systems Engineering originated to deal with complexity. Nevertheless, new and emerging complications have arisen due to fast change in requirement.
Snowden and Boone [12] showed that the system could be Simple, Complication, Complex, and Chaotic. The complex systems are always taking the characteristic of Unknown Unknows in advance, such as in Systems Engineering process in Systems Engineering process a major change in the system requirement, unpredictable emerging technology used in the main system, a shift in management. At the complex system, the Cause-and-Effect relationships are so intertwined they are only evident in the late stages [12]. This feedback brings no right answer, with many competing ideas. The domain of complex systems needs creative, innovative methods [12]. Besides, leadership is performing an essential role in managing a complex system by creating environments that allow patterns to emerge and increasing levels of interaction and communication. Also, leaders should be able to set back from immediate focus and look at the desired big picture and applying holism principle [13].
Sheard and Mostashari [8] stated "Complexity is associated with difficulty of understanding, difficulty of teasing apart the problem (or system) without destroying the emergent functionality, and difficulty of prediction and control. Complexity is also associated with large size, lots of parts, things that are densely interconnected, things that have many different types of parts." 2. Cloutier [2] defined complexity is " a measure of how difficult it is to understand how a system will behave or to predict the consequences of changing it. It occurs when there is no simple relationship between what an individual element does and what the system as a whole will do, and when the system includes some element of adaptation or problem solving to achieve its goals in different situations." ISSN: 2349-7300

IJIRMPS2006006
Website : www.ijirmps.org Email : editor@ijirmps.org 47 White [14] introduced well processes of leadership on complex adaptive systems, while Snowden and Boone [12] said the leaders have to act differently from the previous solution during managing complex systems due to deferent variables and want to apply complexity science. Also, he introduced the leadership's framework for decision making, which could be easier to follow in the complex adaptive systems.
However, Systems Engineering principles and theories could help to deal with big picture view which is required to cater with complex systems, such as, Principle of Holism "A system has holistic properties not manifested by any of its parts. The parts have properties not manifested by the system as a whole" [15], and also Hitchins [16] stated, "systems engineering addresses the whole problem, and creates the whole solution."

DEVELOPERS.
Based on the reviewed literature, the terms 'complexities,' 'challenges,' and other synonyms have the same sense. The researcher has developed descriptions and typologies to discuss the complexities and challenges of Systems Engineering. Systems Engineering challenges were appropriately categorized as external and internal challenges ( Table 2). External challenges include changing global policies, regulations, or even technical patterns. Internal challenges are difficulties within the Systems Engineering processes such as bad requirement definition, lack of training, or holistic view deficiency in managing the SE process. These external and internal challenges were further classified into technical-based and management-based challenges for more specificity. The technical-based typology was used by Young, Farr [17] in Systems Engineering integration. Similarly, Sheard [18] used the same typology to further study the classification of Systems Engineering challenges. Management-based challenges were cited by Young Young, Farr [17] and INCOSE [19] as social-political complexity challenges.

IJIRMPS2006006
Website : www.ijirmps.org Email : editor@ijirmps.org 49 Fig. 4 illustrates the seven most-cited complexity factors that have direct effects on the system's life-cycle. The primary features of the information in Fig. 4 provide the context of problem background. It further shows how the problem evolved and how it reached the current status of complexities in the Systems Engineering approach.  The seven complexities that have a stronger correlation with the Systems Engineering phases include:

SE Complexities
1. Resource globalization. It is affected by human needs and last-minute change requests by stakeholders. Human always seek comfort with emerging new designs if they think that makes their lives better; nevertheless, that puts pressure on system engineers to meet these requirements while designing a system. The last-minute change is usually the consequence of a customer request or technical feasibility. Furthermore, it has facilitated the use of faster and more efficient transportation /communication means for expediting procurement and related processes [23,31,32]. 2. Increasing the utilization of Commercial Off The-Shelf (COTS). It may be derived from the competition of the items' suppliers and the growing needs of humans and society. In addition, the international competition of parts suppliers creates a variety of options from products. If COTS is built in the system, the complications arising from integration besides alignment of specifications [23,33]. 3. Rapidly Emerging Technology. This is influenced by COTS and progressive human needs. Correspondingly, it causes difficulties in the prediction of systems performance and in integration while developing a system [3, 34-37] 4. The change in requirements. This factor influences more than four motivates, as shown in Fig. 4, all are related to change requests. In the industrial sector, inconsistent requirements and different performance objectives usually make the decision-making process more complex [38]. New requirements are a challenge to delivering efficient system [39]. 5. Lack of holistic view and requirement traceability. The main concern stems from a poor coupling between the technical and management sides of the systems builders, due to a large number of subsystems and change [2,40,41]. 6. Lack of consistency tool and integration. The inconsistency originates from the variation of a multidiscipline system of design and a large number of subsystems within traditional Systems Engineering (document-centric). Complex systems are called interdependence systems if there is no shared management between their components/subsystems, and if they are still developing. This often results in emerging behaviour [42]. However, Systems Engineering is challenged to deliver the lowest possible interdependence in the system [2]. 7. The tightness of budget for the overall system life-cycle. Basically, considerable attention has been paid to reducing the costs associated with the acquisition and procurement of systems, and little attention has been paid to the costs of system operation and support. When designing systems, it is important to observe all decisions in the context of total costs [23]. The literature listed in Table 3 provides relevant observational evidence for Systems Engineering complexity. It also shows the seven complexity factors most related to Systems Engineering design process, corresponding to the individual author's opinion and findings on each factor. More than half of the cited authors stated that the fast pace of technology growth creates complexities in the integration process of Systems Engineering. Furthermore, three factors-lack of consistency tools and integration, resources globalization, and lack of holistic view and traceability of the requirements-were cited more than ten times by different authors. Two factors-bad requirements definition and overall life-cycle-have shown the lowest impact on the Systems Engineering process as they were stated less than ten times. Nevertheless, several works of literature have dealt with the concept requirements definition criteria as an essential question on Systems Engineering in other areas out of this study's scope. The Pareto principle, or more accurately, the rule of "80/20," which explains cause and effect, was used in this study. It is a statistical analysis tool used to select a limited number (20%) of overall variables for decision-making to achieve a meaningful overall effect (80%) [61,62].
The Pareto principle was used to determine the factor that has the highest impact on Systems Engineering life-cycle during the developmental phase. The frequency and accumulative percentage of the factors that have an impact on the SE complexity are shown in Fig. 5.

IJIRMPS2006006
Website : www.ijirmps.org Email : editor@ijirmps.org 52 The results showed that technology growth (Rapidly Emerging Technology Items) was the most significant factor (Fig. 5). Moreover, the Rapidly Emerging Technology Items lay above the line of the percentage of citation accumulations, which indicates that if a researcher reduces the impact of this factor, they are likely to solve 80 percent of Systems Engineering complexities problems. Table 4 shows the Systems Engineering phases beside short descriptions of their tasks and levels of the system. It also reflects the associated complexity factors for the design phases of Systems engineering that were obtained from the preliminary analysis of Fig. 5. Website : www.ijirmps.org Email : editor@ijirmps.org 54 All the factors contributing to the SE complexities extend to more than one stage. Three factors-"the overall cost of the life-cycle," "lack of a holistic view," and "poor definition of requirements-" influence all phases of the System Engineering design approach. Rapidly emerging technologies and wrong definition of requirements weighted highest among all other factors.
Finally, any changes to the requirements, items concepts, or technologies upgrade will lead to a redesign of some subsystems, which in turn, will increase the cost of iterations periods. The factors contributing to the SE complexity, agreed on by the authors, practitioners, and designers, will be discussed in the following sections.

Rapidly Emerging Technology
There is no single definition of the central concept of Emerging Technology [65]. In this study, Rapidly Emerging Technology refers to the advancement of the items or components used during the development phase of the systems. The rapid technological advancement discussed by Blanchard and Blyler [23] shows that technology is growing faster than our ability to control and manage it through a traditional Systems Engineering approach. Systems designers should consider existing and potential future improvements in technologies to the system while designing it. Otherwise, there is a significant possibility that the main system will be out of trend. The International Council On Systems Engineering INCOSE [19] ) has named the following technologies as ones that will create challenges to Systems Engineering: -sensor technologies: -material science -miniaturization -human-computer interaction technologies -computational power Additionally, it is very useful to consider electrification and hybridization technologies as a fast-growing technologies. This adds a level of challenges to current systems under development and the Systems Engineering design approach as these technologies are under development and making very real progress (Brelje & Martins, 2019; Schäfer et al., 2019).
One of the problems of ongoing technologies is technology insertion in the man system. The insertion scenario of new technologies while designing using the Systems Engineering approach has not been mentioned [66]. Moreover, some of the technology insertion processes are still struggling and may cause significant reworking: for example, the lithium-ion batteries in the Boeing 787 Dreamliner model [67].

Lack of Consistent Tools
Madni and Sievers [5] indicated that traditional Systems Engineering models such as Waterfall, V-model, and Incremental were facing inconsistency in addressing the heterogeneity of subsystems. As the systems continued to scale and increase in application complexity, they were unable to maintain consistency and assure traceability during systems development at the same level.
Based on the data that was extracted from the literature listed in Table 3, the causes of inconsistency in systems integration could be summarized as follow:  a large number of systems levels,  multidiscipline in systems,  sequential of traditional Systems Engineering. Even though concurrent Systems Engineering has overcome such deficiencies, it still lacks full functionality, and A study by Madni and Sievers [5] presented Model-Based Systems Engineering (MBSE) as a methodology equipped with suitable tools like repository (single source of truth) for data and information exchange between systems team builders. This model may overcome most of the traditional Systems Engineering deficiencies.

Globalization and marketplace competitions
Globalization means there are interdependencies in the world [23]. In depth, many systems are currently being developed by several departments at different places, by multiple suppliers, and through multiple organizations [68]. It is facilitated by rapid and improved communications practices [18]. Globalization affects the Systems Engineering phases that involve multisource components and items. Due to the increasing global competition caused by globalization, systems developers have better access to global resources to fulfil systems requirements. A strong correlation has been observed between globalization and the utilization of Commercial Off The-Shelf (COTS), which further improves the accessibility of the required resources in Systems Engineering [31]. However, multiple studies identified globalization as the complexity factor to multi-resource management in the Systems Engineering life cycle [27]. In the development phase of the Systems Engineering life cycle, the designer needs to perform preliminary prototyping and contracting according to major Systems Engineering standard frameworks from the Department of Defence (DoD) or the National Aeronautics and Space Administration (NASA) [24]. Otherwise, the traditional Systems Engineering approach might experience downstream failure on verification and validation, or even in the production phase.

IJIRMPS2006006
Website : www.ijirmps.org Email : editor@ijirmps.org 55 4. Lack of holistic view Systems Engineering is defined as a holistic and integrative discipline (Hirshorn, Voss [24]. Nevertheless, Cloutier [2] from the INCOSE global community noted that one of the factors that made Systems Engineering more challenging was the lack of a holistic view by practitioners and system developers. The development phase is one of the important phases of the systems life-cycle, where most of the engineering developments happen. A holistic view is an essential personal skill for practitioners and systems developers in planning, organizing, and realizing the system-ofinterest. This will remain a challenge if enterprises wish to be competitive in the current technology trend. A holistic approach helps to reduce risks and difficulties in managing tasks in the development phase and improve the communication between the systems levels. Furthermore, Cloutier [2] stated that use of Holism principle is essential in Systems Engineering to reduce individualization risk. Currently, the Systems Praxis Framework, which was developed by INCOSE and International Society for the Systems Sciences (ISSS), is the potential solution for Lack of holistic to today's complex systems [2].
Possible explanations of the consequences of the "lack of holistic view" in Systems Engineering is unable to explain the behavior of the overall systems by individual parts [50].

Commercial Off The Shelf (COTS) utilization
Increasing numbers of systems have adopted COTS to lower the initial procurement costs and shorten the acquisition cycle (Blanchard and Blyler, 2016;Eisner, 2008). This is because COTS usually makes use of the latest commercial technology, which will be replaced by new technology innovation in a short period of time, and are taken as consumables [69].
To its advantage, COTS could be selected and implemented for technical (shortening the developmental phase), organizational (reducing the overall cost of the developmental phase), or strategic reasons (access to technology not available internally) [33]. Recently, professional COTS often comes with supporting documentation such as proof of verification, validation conformance, and manufacture specifications or fact sheet. It is always recommended to review the specifications to ensure the COTS fit the requester's requirements [24].
On the other hand, the advantages of COTS challenged by integration concerns such as performance (what is supposed to do), compatibility (no standards), product assessment (uncertainty of meeting the required needs), or supplier behavior (agreements promises false) [33]. Additionally, MacKenzie, Bryden [44] observed that many companies struggle with COTS integration into the Systems Engineering processes, the possible reason the employee is unfamiliar with that COTS.
Moreover, stakeholders occasionally order COTS, which indicates that they have decided to apply a readily available solution to their system without first validating that solution [54].
Furthermore, risks associated with the use of COTS during system life-cycle include the obsolescence of models and of improvements in system interfaces. The competition of COTS providers makes individual components obsolete within two years [34]. The incompatibility between two or more systems requirements in COTS specifications may cause further complications to systems engineers. It is a nontrivial task to tackle this type of complexity, as it depends on the experience and design expertise of hardware and software engineers [46].
According to the literature listed in Table 3, COTS usage was an unavoidable situation in many cases. Multisource of COTS generates complexity to integration in Systems Engineering processes during the development phase due to deferent standards used. It will be difficult to integrate new COTS during the development process, as this may lead to the re-integration and re-assessment and re-validation of the system under construction.

Bad Definition of the Stakeholder Requirement
Mordecai, Dori [54] defined the Stakeholder requirements as ideas, expectations, requests, a set of needs, goals, assumptions, guidelines, preferences, objectives, constraints, intentions, or desires. Requirements are the needs or demands of the stakeholder collected as statements to constrain and identify a system or process. The ideal requirements are clear, unambiguous, consistent, unique, traceable, verifiable, and "SMART" (Specific, Measurable, Attainable, Realizable, Time-bounded) [70]. However, the inadequate definition of the requirement for stakeholders indicates that the requirement is improperly collected or elicited, which leads to developmental phase difficulties, extra cost for a delayed change request, functional analysis, or issues in integration and systems evaluation.
Practically, requirements are the input to the design process at the beginning of systems formulation, while specifications are the output of the development phase. In other words, the development phase is begun with well-defined requirements [23,24]. Yet, some inexperienced developers are unable to differentiate a requirement from a specification.
Continuous monitoring of the requirements during the Systems Engineering life-cycle results in internal complexity as it requires controlled traceability, and hence adds more tasks to systems developers. de Weck [71] and Hirshorn, Voss [24] proposed some tools that could assist the requirement monitoring process, including Excel spreadsheets, professional commercial tools like DOORS for large complex systems, and metadata.
While traditional sequential Systems Engineering is unable to address the complexity from continuous requirements monitoring that involves rapid and dynamic changes of input, a holistic view and MBSE can reasonably overcome this issue [5].
Standard Briefly, a Systems Engineering life-cycle cost is the expenses from all the systems phases [2]. Usually, a tight project budget has a positive impact on the project/system owners, the sponsor, and other stakeholders. In this case, however, systems engineers and developers are challenged to deliver the system within that limited cost range. As a result, the cost tightness combined with a poor definition of the requirements and a dynamically changing environment makes task management complicated.
When calculating the overall life-cycle cost, the systems engineer needs to understand that any of the abovementioned processes might need to be repeated until the desired specification is achieved. Moreover, rapid and frequent change in technology and requirements will affect the life-cycle cost analysis of the systems of interest as well, even though it may lead to better performance.
Concisely, overall life-cycle cost is causing complexity in Systems Engineering in the following ways:  requirement of minimum life-cycle cost,  dependency of the lifecycle cost in decision making and design reviews output, and  dependency of life-cycle cost in making the decision that relates to the change of technology in the systems-of-interest.

IV. CONCLUSION
A Systems Engineering complexity is defined by different sources as a measure of the difficulty in understanding the behavior of a system or in predicting the consequences of a change. This study identified and analyzed factors that lead to the complexity of Systems Engineering throughout the phases of the life-cycle of Systems Engineering.
Complexity factors were found to be linked to each other and intertwined with the Systems Engineering stages or processes. This creative synthesis is essential to produce an integrated system that can meet the end-user requirements.
Although other factors are crucial, rapid emergent technology has the highest impact on Systems Engineering complexity, particularly during the developmental phase in complex systems like Aviation industry, because any changes accrue during design process lead to verification and revalidation and reintegration as well. MBSE can be used to resolve some of the SE complexity issues. Still, some of the factors remain as challenges to the traditional Systems Engineering.
In order to integrate cutting edge technology into the Systems Engineering processes, a system engineer or developer needs to follow the technology development closely. The rapid change in emergent technologies happens during systems development will lead to repetition in requirements verification, specifications validation, subsystems or components design, trade-off studies, alternatives evaluation, engineering and prototype models development, production planning, and tests and evaluation development. The available Systems Engineering frameworks or guidelines lack means to accommodate and address the issue of rapidly emerging technologies.
Some questions remain unanswered, especially in interconnectivity improvement. Improving system interconnectivity eliminates the erratic behavior of complex multidiscipline components and helps to produce a better systems design, especially a smart and dynamic systems solution. However, in order to improve system interconnectivity, external technical-based challenges such as globalization and COTS need to be tackled first. MBSE has the advantage of a holistic view and traceability over traditional document-centric Systems Engineering. However, the majority of the companies do not use MBSE due to a lack of training and experience in their employees.