New Governance Approaches to Prevent the Collapse of Complex Socioeconomic Systems

M odern socio-economic and technological systems are constantly becoming more complex, and as a consequence, the risks of their failures are increasing. Effective management requires tools appropriate to the new challenges. Complexity science offers a number of concepts that individually help to cope with increasing complexity and its effects to a greater or lesser extent. However, a more effective approach is their skill-ful synthesis, which allows to cover the system holistically, to identify the origin of potential crises and catastrophes that would otherwise remain «hidden», and to outline preventive corrective measures. The article presents a review and comparative characterization of paradigms of perception of complex systems extrapolated to the sphere of management. Using multilayer causal analysis, the case of two high-profile disasters that occurred with Boeing airplanes is considered. The concept of «orphan systems» is proposed, which allows to catch weak signals about the dangerous drift of the system, to react in time and take an appropriate managerial actions.


Introduction
Complex systems have been, and to a large extent remain the terra incognita of modern management science.Despite the enormous increase in knowledge about dynamic complexity achieved in natural sciences and engineering, and the growing interest in relevant tools, the understanding of this phenomenon remains fragmented and blurred, which does not allow to capture it fully.Given the variety of methods for describing complex processes and systems, in most cases the latter's behaviour can only be explained post factum.Although they are believed to be unpredictable, in some cases it is possible to identify the forces which determine their development vector.The paper attempts to classify the existing approaches to describing and managing complex socio-economic systems.Using the multilayer causal analysis method, the Boeing case was explored, which illustrates how the lack of a holistic approach to a complex production system, and failure to understand the nature of its hidden transformations did not allow to detect the weak signals -harbingers of disasters in time, and switch the system to a safe mode.The paper begins with an analysis of modern scientific paradigms of complex socio-economic systems' perception: their initial assumptions, features, and predictive potential.Then the cases of two crashes of Boeing 737 MAX aircraft (in 2018 and 2019) are analysed as examples of major failures of complex social systems.An attempt was made to identify the underlying causes of the system's collapse in the scope of one of the paradigms.The author's vision of complex socio-economic systems' development dynamics is proposed.The "orphan system" and "system drift" concepts are introduced, which help to better understand the processes taking place in the systems under consideration, and the logic of their changes.

Complex Systems' Management Paradigms
The professional community identifies four main complex system perception paradigms, each of which, with their respective strengths and limitations, can enrich management practices.Mechanistic paradigm.Complex socio-economic systems are compared with corresponding feedbackdriven technical systems based on interaction between their elements (Rosenblueth et al., 1943;Wiener, 1948;Boulding, 1956;Von Bertalanffy, 1950;Forrester, 1969Forrester, , 1971)).This modelling area is currently known as system dynamics (Richardson, 1991;Sterman, 2002).The origins of this approach are sometimes traced back to Isaac Newton's works.In economics it was first applied by Adam Smith, but it has reached its prime with the rise of Taylorism in the 20 th century.The economy is seen as an equilibrium machine brought to balance by an external force: the "invisible hand of the market".It can be represented as a model, albeit a simplified one (Raskov, 2005).From the standpoint of a mechanistic, engineering perception of the world, the more com-plex the system, the more unpredictable its behaviour, and the higher the probability of its elements' failing.Automation strengthens the links between system elements and subsystems, so the system becomes less and less controllable (Perrow, 1999).Various failure warning mechanisms only complicate the system, thus increasing the risk of accidents further.Many large, resonant catastrophes resulted from such processes, and thus can be classified as "natural".Detecting weak signals (harbingers of accidents) is a way to minimise risks, but this approach often does not work because by the time "sufficient" information is obtained, no time to respond is left (Ansoff, 1979).Natural science paradigm extrapolates the natural sciences' patterns (mainly from physics and chemistry) to socio-economic systems.Its main areas include econophysics and synergetics.Researchers with a background in physics who have devoted themselves to studying economic matters, primarily financial markets-related, work in this area.Concepts such as power laws of distribution, phase transitions, diffusion, correlation, turbulence theory, etc. are applied.This is justified by the fact that it is impossible to perform large-scale experiments in economic theory and finance, so one cannot do without statistical physics tools.Financial markets are seen as non-linear complex open systems.Econophysics gained momentum in the 1980s thanks to the work of the Santa Fe Institute researchers (Arthur, 2001;Mantegna, Stanley, 1999;Sornette, 2003;Helbing, 2012;etc.),and in its turn contributed to the development of an agent-based simulation method which describes market players' interaction using physics principles.Synergetics studies complex systems' self-organisation (Haken, 1981;Prigogine, Stengers, 1984;Ebeling, Feistel, 1986;Kurdyumov, 2006;etc.)Self-organisation is defined as non-equilibrium processes which under the influence of systemic driving forces lead to the emergence of more complex structures.One of these processes is thermodynamic equilibrium: a mechanism describing complex chemical reactions similar to phase transitions in physics (Prigogine, Stengers, 1984;etc.).In their development, complex systems periodically come to bifurcation points characterised by high uncertainty, so even minor events can radically change the course of the system's evolution.Proponents of this approach suggest that the system's behaviour can be predicted through identifying the order-defining parameters (attractors), which are few.They are determined by the behaviour of system elements and subsystems, but then suppress them and set the vector for the whole system, in dynamics.Knowing potential attractors, and understanding the laws of complex systems' evolution makes it possible to predict their path with a certain probability.By affecting complex systems near bifurcation points, one can turn their further development to a preferred direction, since "when passing through forks, the environment becomes sensitive to collective and individual actions which can lead to the emergence of new social, cultural, technological, and other patterns" (Knyazeva, 2020).Evolutionary biological paradigm uses the biological metaphor and the evolutionary mechanism concept to describe complex socio-economic systems (Schumpeter, 1912;Alchian, 1950;Moore, 1993;Nelson, Winter, 1985;etc.).This approach is primarily reflected in the "evolutionary economics" theory according to which markets, as complex systems, dynamically change over time due to competition and survival of the fittest (Williamson, 1996;Beinhocker, 2006;Dosi, 1982;etc.).Change is open, and determined by heredity and survival.Business processes can be changed by introducing new practices and technologies, and passed on to new generations of economic agents like genetic information.Change can be intentional or accidental.The survival is determined by the effects of the external environment (market); the fittest take root in it, which corresponds to the logic of innovation dissemination.According to the new look at adaptation, companies not only adapt to the external environment, but can themselves change it to suit their needs, creating market niches (in the economic context, territorial clusters, value creation ecosystems, industry-specific competition rules, etc.) (Nelson et al., 2018).Macroevolutionary leaps do not add up to a set of microevolutionary changes, but are also explained in terms of macrolevel phenomena such as behavioural patterns.As the animal world adapts to climatic and geological changes, so complex social systems have to adapt to changing external conditions.The ecosystem paradigm is one of the mainstreams of modern strategic management, based on the competitive cooperation and the development of business ecosystems concepts.A popular tool is multi-agent modelling, which reproduces agents' behaviour (individuals, organisations, and other autonomous subjects), the rules of their interaction, and the environment they operate in.The behaviour of the entire system (at the macro-level) is determined by the numerous strategies of individual agents who imitate each other, "infect" each other with ideas and rules, and thus create the emergent behaviour phenomenon.The computational power available today allows to describe agents' actions in nuances, and build sophisticated models.E.g. consumer behaviour is studied taking into account rational and irrational decision-making aspects (cultural and religious ones), multi-criterial and context-based choice situations, etc. (Katalevsky, 2015).Agent-based modelling allows to visually trace how small, and seemingly secondary factors which determine players' behaviour and interaction lead to significant social consequences (Wilensky, Rand, 2015).Anthropocentric paradigm.This is the only approach to complex systems focused not on the complex processes as such, or on adaptive ecosystems, but on the individual who makes the decisions, and their motives.In our opinion, this approach seems to be the most objective in comprehending complex social systems, and serves as the basis for a realistic assessment of their development.It blends the achievements of economics, sociology, psychology, management science, and political science.The essence of individual and collective human behaviour is studied, along with the specifics of people's interaction with the environment, and the logic behind their choices (Simon, 1972;Deming, 2000;Lindblom, 2001;Schelling, 1978;Ackoff, 1978;Mintzberg, 2013;Akerlof, 2000).The growing popularity of the anthropocentric paradigm is in line with economists' growing interest in studying the substance and motives of human behaviour (Kahneman et al., 1982;Thaler, 1994;Sunstein, 2014;Ariely, 2008;etc.).Economic processes are perceived as emerging social phenomena determined by group interaction (Andersen, Nowak, 2014).Sociologists call such phenomena "constructing social reality" (Berger, Luckmann, 1966).Several levels of systems analysis are typically distinguished: micro-level (individual choice), meso-level (group decisions),1 and macro-level (the entire economic system) (Dopfer, 2004).In the first case, the combined decisions determine the behaviour of a person, in the second of a group, and in the third of the entire macrosystem.The process evolves along the chain from the micro to the macro level, and is described by the unintentional segregation model (Schelling, 1978).It has been proven that individual behaviour is not always rational; its nature is much more complex than previously thought (Simon, 1972;Kahneman et al., 1982).Since the early 2000s the identity theory was gaining popularity, which emphasizes the importance of the social group an individual identifies with (Akerlof, Kranton, 2010).The perception of stories (narratives) determines individual economic strategies which affect the macroeconomic system's behaviour as a whole.The main characteristics of the four approaches described above are structured in Fig. 1.Their features largely determine the range of solutions they offer, and their limitations.The choice of approach in many ways defines the result.

Limitations of the Considered Approaches
The limitations of the technical paradigm based on the "Fix it!"logic are due to the fact that a complex system can only be "fixed" a posteriori, i.e. after a "breakdown".And often it is impossible to get even the first idea in which part of the system a problem will arise.Neither the human factor, nor the socio-cultural context are taken into account.Effective organisations tend to apply interconnected, highly integrated processes and routines which allow complex work to be completed on time.But if an error "penetrates" such a structure, it rapidly "infects" the entire system.An analysis of 80 complex technical system failures in the UK showed that the more a hierarchical organisation strives for order based on bureaucratic procedures, the more prone it is to errors (Turner, 1978).Excessive ordering of business processes increases the likelihood the work will be done according to plan, but at the same time errors will be reproduced and replicated throughout the system.Thus complex systems' failures can be caused both by violating the order, and excessively increasing it.A healthy organisational management process is achieved by avoiding excessive control, building a less rigid hierarchy, coordinating autonomous teams' operations, encouraging diversity of opinion, and flexibility in decision-making (Weick, 1998).The arsenal of the natural science paradigm primarily includes complex mathematical tools (chaos theory, correlation, time series, etc.).Its limitation is that individual behaviour and motives cannot be mathematically calculated.This approach can be used to describe certain phenomena such as, e.g., group behaviour during emergency evacuation, or price fluctuation patterns in financial markets.But since the human factor with its complex motives is removed from these models, they do not allow to holistically interpret complex phenomena.
The evolutionary biological paradigm is actively applied in present-day strategic management, because it offers effective analogy models and "working" strategies (such as co-evolution, "competitive cooperation" (co-opetition), etc.).Since even large companies find it difficult to compete on their own, the "joining the pack" approach appears to be promising.Organisations create their own ecosystem, or join the dominant one.However, this model also greatly reduces the choice of strategies, since adapting is not always the only right way, or it does not guarantee long-term survival, which is confirmed by numerous historical examples.Real life is much richer, and offers a wide variety of options.The anthropocentric paradigm proceeds from the understanding that a person's actions are determined by their identity and by socio-cultural factors, so it proposes to focus on designing social systems (hence its notional slogan "Design!").It is quite popular with institutional economists who pay particular attention to the norms, laws, and culture which determine economic behaviour.2Other approaches, with the exception of the anthropocentric one, prefer "not to see" individuals, which is reflected in their terminology (Mc-Closkey, 1993).E.g. the technical and econophysical paradigms "animate" complex systems: according to them the latter "adapt", "develop", "interact", etc. Com- Katalevsky D., Figure 1.Main scientific paradigms for complex systems' perception Source: author.

Mechanistic (engineering)
Bottom line: social systems as complex technical mechanisms which require design, maintenance, and repair Slogan: "Fix it!"Tools: system dynamics, discrete-event simulation, system engineering tools Scientific fields: general systems theory, systems engineering, simulation modelling

Evolutionary biological
Bottom line: complex social systems' behaviour is described by biological and ecological laws Slogan: "Adjust!" Tools: biological and ecological laws applied to social systems, agent-based simulation Scientific fields: evolutionary economics, management (strategic, innovative)

Natural science
Bottom line: complex systems' behaviour is determined by the universal laws of natural sciences Slogan: "Calculate!" Tools: mathematical modelling, physics and chemical laws as applied to social systems Scientific fields: econophysics, synergetics, catastrophe theory, etc.

Anthropocentric
Bottom line: complex social systems' behaviour as a result of complex, multidirectional activities of people Slogan: "Design!" Tools: sociological, behavioural, and psychological laws applied to social systems Scientific fields: decision theory, organisational behaviour theory, behavioural economics, identity economics, constructivism, narrative theory plex systems are presented as self-sufficient objects endowed if not with reason, then with high autonomy.Man as a manager is excluded from consideration.Still, the technical paradigm indirectly implies that complex systems' mechanisms must be designed, maintained and, at least occasionally, repaired by someone.I.e. the individual is the subject, while the complex system is the object of management.It seems to us that the mechanistic, natural science, and evolutionary paradigms of complex systems' perception, despite using different perspectives still do not adequately describe social systems.An exclusive reliance on them will inevitably lead to methodological errors when it comes to analysing individuals' complex behaviour mediated by the social context, specific worldviews, traditions, etc.Unlike the other approaches, the anthropocentric paradigm blends the widest layer of interdisciplinary research in economics, sociology, psychology, and management, and thus in our opinion allows for the most realistic modelling of human behaviour in complex systems.A good example of the anthropocentric approach is the "garbage can theory" (Cohen et al., 1972), which has greatly influenced economics, sociology, and management.It challenged the dominant at the time rational decision-making paradigm by offering the most realistic description of the process.Among other things it was applied to explain the causes of technical disasters (Sagan, 2020).From the point of view of managing complex socioeconomic systems, analysing the system built around the production of Boeing passenger aircraft, and the specifics of the US aircraft industry regulation is of great practical interest.This case, viewed through the prism of anthropocentric approach, shows how the evolution of complex relationships between various influence groups inside and outside Boeing has led to high-profile technical disasters.

Boeing 737 MAX Case Study: A System Error which Cost 346 Human Lives
At the end of the last decade, two dramatic accidents happened within several months from each other, both related to Boeing, a long-term global aircraft industry leader.In the fall of 2018 a Lion Air plane crashed, and in the spring of 2019, an Ethiopian Airlines one.In both cases the aircraft were of the Boeing 737 MAX series, approved by the US Federal Aviation Administration (FAA) as safe to fly two years earlier.Together, the two tragedies claimed 346 lives.By the special FAA order of 13 September, 2019 all Boeing 737 MAX aircraft in the United States were grounded pending the end of the investigation; three months later Boeing suspended the production of this aircraft series and fired the CEO.What is of interest in this story is not so much the technical causes of the disaster, as answers to the questions what systemic reasons have led to these events, when did these reasons arise, how did they evolve, and was it possible to prevent the tragedy?We are talking about a complex socio-economic system which is subject to one of the most stringent, thorough, and technically advanced regulation in the world.How did the actions of various interest groups lead it to gradually evolve into a "orphan system" which has failed on a massive scale?Our analysis is based on the findings of an independent investigation of the incidents in question, conducted by the US authorities (HCTI, 2020).

Technical explanation of the causes of the disaster
According to experts, the Boeing 737 series is a global civil aviation "bestseller": over 15 thousand aircraft have been sold in total.The Boeing 737 MAX modification was Boeing's answer to its main rival Airbus's plans to launch an improved version of the A320 aircraft -the A320neo, 14% more fuel-efficient than its predecessors.To match the rival design, the 737 MAX series was given larger, upgraded engines.The company positioned this aircraft as being similar to the flagship model ( 737), which made it unnecessary to retrain pilots for flying it.However, the use of larger engines necessitated structural changes, which in certain cases caused the aircraft to destabilise during flight.In an attempt to eliminate this factor, Boeing developed the Manoeuvring Characteristics Augmentation System (MCAS), which automatically corrected plane's position in the air.When the aircraft was under manual control, the MCAS was supposed to be activated by the pilot.But, as it turned out during the investigation of the disasters, in some cases the system failed: many times it engaged on its own.And it was impossible to turn it off, or put the aircraft into manual control mode (HCTI, 2020).After the first crash Boeing blamed the pilots for being underqualified.Only after the second accident the company acknowledged the problems with the MCAS.In a critical situation the pilots were expected to deal with it by switching the aircraft to manual control mode.But as was noted, no additional pilot training was carried out.Also it turned out that switching to manual mode was impossible in principle, since during the MCAS operation this mode was turned off by default.Errors in the MCAS design violated the main design regulation, according to which automated systems' operation must not hinder the actions of the pilot (HCTI, 2020).After the two accidents Boeing made a number of technical improvements to the MCAS: more sensors were added, the possibility of its spontaneous engagement during the flight was eliminated, opportunity to switch to manual control ensured, and additional pilot training conducted.However, the aircraft design and software faults do not fully explain why things "went wrong".To understand the reasons, one must look at deeper corporate culture issues.

Features of the Boeing corporate culture
Established over a hundred years ago by the experienced pilot William Boeing, the company quickly began to receive orders from the US Navy, which facilitated its subsequent rapid growth. 3The founder created a culture that could be described as "a community of engineers dedicated to building excellent aircraft" (Frost, 2020).It was based on a philosophy of increased attention to detail, in line with the belief that neglecting cause-and-effect relationships leads to incorrect interpretations, which, in turn, results in making wrong decisions. 4In 1960-1970 the US air transportation industry was heavily regulated, the market didn't grow particularly rapidly, and the competitive pressure on Boeing wasn't high.However, with the deregulation of the sector and the rise of competition (Fig. 2), the company faced the need to optimise costs.In 1997 Boeing merged with McDonnel Douglas.At the time such a deal seemed to be a perfect solution for both parties: Boeing was the leader in civil aircraft construction, and McDonnell Douglas got an opportunity to make a leap to the top on the strength of the partner's competencies.Otherwise, developing a new competitive aircraft would require $30 billion and 10 years of work, provided that competitors would not make any progress during that time (Callahan, 2020).This has changed the corporate culture: the focus on solving complex engineering problems was replaced by the desire to increase financial gains.However, in the clash of the two parties' corporate cultures the philosophy of the smaller McDonnell Douglas prevailed.As a consequence, Boeing has shifted from its emphasis on solving complex technical problems and conducting costly breakthrough research to increasing profits by cutting costs, and abandoning radical innovation in favour of upgrading older models (Frost, 2020).
Boeing employees had a hard time adapting to the new philosophy, which contradicted their main value: "making excellent aircraft" (Greenberg et al., 2010).The focus on minimising costs and maximising profits has created a "fertile" ground for "replicating" technical errors.Industry experts estimate that in 2011 the cost of designing a new aircraft would be $10 billion, while repurposing the 737 MAX from the 737 NG series model only cost $3 billion.In seven short years, the gradual effect of these destructive forces has led to two major disasters.What seemed to be just an unacceptable engineering error (an MCAS problem), actually had deep roots: different motivation focused on shortterm financial results.However, the landscape of possible causes of the crash would not be complete without examining how the technical issues were overlooked by the key industry regulator, the Federal Aviation Administration (FAA).

Miscalculations of the industry regulator
The 737 MAX development team was under intense pressure from the management to get the aircraft to market faster.As a result, a "concealment culture" developed in the company, which amounted to misinforming the FAA -the agency responsible for certification of all aviation equipment supplied to the US market.Since the FAA didn't have sufficient human resources to independently perform all the necessary tasks, it had the right to delegate some of its certification responsibilities to qualified third-party professionals (Fig. 3).These professionals, known as "designated engineering representatives", were employed (and paid) by Boeing, but reported not to the Boeing management but to the FAA supervisors.They were the FAA's "eyes and ears" in the field, thoroughly familiar with the intricacies of the certification process and believed to take an unbiased approach to certification.This practice was first implemented by the FAA in the 1950s, and has since evolved towards gradual expansion of the FAA field representatives' powers (Fig. 3).This approach was applied to well-known, low-risk technological solutions.It allowed the FAA focus solely on assessing high-risk technologies (projects critical to safety, or radical innovations).However, in reality this "strategy" led to ignoring a number of certification requirements, which also contributed to the Boeing aircraft disasters.Fig. 4 shows the gap between the rate at which new technologies subject to certification are introduced by the industry, and the FAA's "throughput capacity" (internal resources to process applications).In the case of the Boeing 737 MAX in 2013, the FAA delegated 28 of the 87 certification operations to the company itself.By the end of 2016 this ratio was already 79 to Katalevsky D., 1975 1980 1985 1990 1995 2000 2005 2010 2015 Source: author, based on: https://finance.yahoo.com/news/1997merger-paved-way-boeing-090042193.html,accessed 17.02.2023.
Boeing Airbus 91.According to the findings of the investigating panel, the FAA "outsourced" to the aircraft manufacturer too many certification responsibilities (OIG, 2020).Plus, seemingly minor changes were made to the delegation regulations in 2005, which, as it turned out later, had a significant impact on the certification process and its results (Fig. 5).Under the previous system, designated engineering representatives, despite being fully funded by Boeing, reported directly to the FAA.With the introduction of the new system, Boeing itself gained the right to appoint such experts (Fig. 5); they handed the information over to their managers, who processed it and the passed on to the FAA (a similar system was approved by the FAA itself).A few months before the first crash, Boeing and the FAA jointly collected and published statistics according to which in 2010-2018 civilian air carriers had just a single fatal accident.Overall, US civil aviation fatalities (per 100 million passengers carried) over the past 20 years have fallen by 95%. 5 Excessive complacency with such a picture has led to a gradual relaxation of the FAA control of the certification process.However, the MCAS was not the first technical issue the FAA missed.A few years earlier problems with spontaneous combustion of lithium-ion batteries in the Boeing 787 Dreamliner series were discovered, during commercial operation of the aircraft.As with the MCAS later, all aircraft in the series were grounded pending the completion of the investigation.During the certification one of the FAA engineers suggested putting the batteries in a steel case, but Boeing rejected this recommendation, and the FAA officials went along.Only after several spontaneous combustions and the complete termination of all Dreamliner flights the steel casing idea was implemented (HCTI, 2020).Thus even before the Boeing crash, the FAA's supervision delegation system was failing.However, these failures were seen as rare occurrences, so the general certification procedure remained largely unchanged.There is another critical factor in the process under consideration: a conflict of interest in the form of the manufacturing company's pressure on the FAA experts.As of 2013 the FAA began to survey its designated representatives, and in 2016 the company got involved too.Many respondents reported being pressured, to varying degrees, by the Boeing management to speed certification up.Distorted communication between Boeing and the FAA (information was funnelled through a double filter) served as an aggravating factor.For this reason the FAA was unable to adequately assess the risks associated with technical flaws in the design of the MCAS.An analysis of the chain of factors that led to the disaster allows to build a hierarchy of underlying causes of the system's degradation.The main factor was the shift in Boeing's values hierarchy (flight safety waned into the background in favour of maximising financial results), which led to a "shortening" of strategies' timespan.The changes in the FAA certification system also made a contribution (fig.6).

Boeing 737 MAX production race
The report by the experts who investigated the causes of the tragedies highlights the "production race" for the Boeing 737 MAX assembly.The company sought to deliver the aircraft to customers as quickly as possible.If in 2010 the output was just over 30 aircraft per month, in 2014 this figure reached a record value for Boeing at that time at 42, and shortly before the first disaster it was planned to increase it to 57.The focus on stepping up the financial indicators replaced the more important goal of introducing the most advanced safety technologies, and innovations in general, leading to an increased load on production facilities. 6aken together, all this caused a serious degradation of the flight safety system.

Complex Systems' Evolution, or Why Failures are Inevitable
We've described the "orphan system" phenomenon using the development of Boeing 737 MAX aircraft series as an example.Any complex system has internal contradictions of some kind due to its inherent multidimensionality and multiple "tension points" arising because of various internal and external forces.The system becomes "orphan" when its key players refuse to perceive it as a whole and take responsibility for its long-term sustainability.Instead, the problem is passed on to another player (and sometimes to subsequent generations, as, e.g., in the case of the natural environment).This system state becomes a natural outcome when internal contradictions gradually intensify  in the course of the system's drift.The drift happens as follows: first, there is a slow accumulation of errors ("mutations") which is hard to analyse objectively; values and fundamental principles deform, the planning scope shifts from long to short term.The danger is that the organisational system misses this process altogether due to its gradual and prolonged nature.As a result, the system loses the "owner", and starts to change under the influence of the dominant "pressure" vector.Several factors have become critical in the process of the complex US civil aircraft manufacturing regulation system turning into a "orphan" one.The system drift was mainly driven by the in the company's culture, the shift of the motivational attractor from the long term (passenger safety) to short (annual financial result), and the FAA's new aircraft certification policy which created a conflict of interest with the company management.Communication distortions due to the fact that information from the could only reach the FAA after passing through several "filters" made an additional contribution.As a result, the regulator simply couldn't detect technical problems in time.

Principles of "orphan" systems' evolution
Complex systems can be in a stable or unstable state.
The transition from the first to the second happens in several stages (Fig. 7): 1. Creation and launch.The system's foundation is laid, its development vector is set, and links between the elements established.
2. Inertia.The system develops in accordance with its basic value principles.

Drift (accumulation of mutations).
The system gradually begins to change under the influence of internal and external stakeholders, accumulating "mutations".Its elements deform, while the links between them and the basic principles are eroded (at Boeing, this process began in the 1990s).
4. Aggravation of contradictions.The system gradually moves away from its basic principles.Conflicts between goals, objectives, and values increase, communications get distorted. 7Instability arises, contract and project deadlines are no longer met, even intermediate objectives are failed.All resources are thrown to find quick solutions, which worsens the situation further: each element is only interested in "saving itself ".The system comes to the pre-collapse stage.5. Collapse.The accumulated contradictions lead to a major failure which causes partial or complete destruction of the system.As our analysis shows, it was "programmed", inevitable, albeit delayed.6. Intervention.The system is reconfigured and updated.After that, if the lessons have not been learned, the cycle described above repeats.This process is common for all types of organisations and socio-economic systems.The larger the system, and the tighter its elements are interlinked, the more prone it is to become "orphan".Holistic thinking, and understanding complex systems' behaviour allows to detect the emergence of a destructive drift in time, and take steps to turn the system in the desired direction.

Conclusion
The paper analyses the little-studied "orphan" systems phenomenon.More complex socio-economic systems imply an increased number of participants, and stronger interconnections between them.At the same time key system participants' basic values deform, while communications, and links in the original system architecture get distorted.The system begins to change under the influence of multidirectional pressure vectors from interest groups, gradually drifting towards collapse.Only a timely and adequate managerial intervention can reconfigure the system, and direct it along the desired path.

References
The complexity science offers a number of concepts which, individually, can to some extent help companies and organisations of different sizes cope with increasing complexity and its effects.However, a more effective approach is skilfully blending them, which allows to holistically address the entire system, identify the origins of potential crises and catastrophes (which would otherwise remain "hidden"), and outline relevant preventive measures.The paper presents an overview and a comparison of complex systems' perception paradigms extrapolated to the field of management.Using multilayer causal analysis, the case of two resonant disasters with Boeing aircraft is considered, which most vividly illustrates the emergence of "orphan" systems.But no matter how destructive the effects of the system's degeneration in the course of its implicit and extended transformation are, it is always possible to reconfigure and improve it with the help of holistic thinking and understanding the nature of complex systems.

Figure
Figure 2. Comparison of Airbus and Boeing output growth

Figure 4 .
Figure 4. Gap between the rate of new technology development and FAA certification potential

Figure 3 .
Figure 3. Development of the FAA Certification Delegation System

Figure 5 .
Figure 5.Previous and new FAA delegation procedures