From Complexity in the Natural Sciences to Complexity in Operations Management Systems

Preface

The word complex is used in many contexts, be it at the level of social sciences, biology, chemistry and physics or in our professional and private environments. Any time we cannot understand a situation, we try to escape the challenge of feeling doubt and uncertainty, because we have the impression that we lack methods and techniques (in one word, capabilities) to address the issues involved. Within this framework, we decide to give up and convince ourselves that we are right to do so because we are overwhelmed by complexity. Complexity is an idea, fabric of our daily experience.

When a phenomenon seems simple to us, it is because we perceive that one object and one action are involved in spite of the fact that reality may be much more intricate. This simplification is enough for making us cognize the ins and outs of the situation we experience. In contrast, when a great number of interacting elements are involved, we perceive the situation as complex.

Economic systems and human relationships are complex. Macroscopic situations may appear simple because the microscopic underlying states are hidden. We perceive averages without knowing the detailed states of the components of a whole.

During the second half of the 20th Century, developments in the thermodynamic theory of irreversible processes, the theory of dynamical systems and classical mechanics have converged to show that the chasm between simple and complex, order and disorder, is much more reduced than thought.

Biology is acknowledged as complex, as it is associated with living organisms whose chemical functioning relies on the interactions of many subsystems. The idea of complexity is no longer restricted to biology and has undergone a paradigm shift. It is invading physical as well as social sciences.

The purpose of this book is to describe the main results reached in natural sciences (physics, chemistry and biology) to come to terms with complexity during the second half of the 20th Century and how these results can be adapted to help understand and conduct management operations.

It is divided into three main chapters, namely Complexity and Systems Thinking, Agent-based Modeling for Human Organizations and Complexity and Chaos.

The purpose of the first chapter, Complexity and Systems Thinking, is to give an overview of the way the concept of system has been instrumental in interpreting phenomena observed in natural sciences as well as in emotional behaviors of the human system, as a human being is a system in itself.

The second chapter is devoted to complexity and human organizations. Analyzing existing organizations is a difficult exercise because human relations are intricate. Making them explicit entails the help of a relevant model, including cognitive features. The BDI (Beliefs, Desires, Intention) agent model that will be elaborated meets this requirement.

The third chapter deals with complexity and chaos that is associated with disorder. We will examine how the concepts developed in physical sciences can be used in the field of human organizations for understanding their behavioral evolutions, especially when change management in organizations is pushed by fast-evolving technologies and their consequences in terms of interacting collaboration and cooperation between their human actors.

Jean-Pierre BRIFFAUT

January 2019

Dedication

This book has been written on the occasion of the fortieth anniversary of the foundation of the Institut Frederik Bull (IFB) by Bull. It operates as a think tank with working groups studying the societal impacts of informatics and economy digitalization. The working group linked to this book investigates the complexity of Systems of Systems (SoS).

IFB has developed collaborations with management and engineering schools (EMLV, ESILV, IIM) located at Le Pôle Universitaire Leonard de Vinci in Paris-La Défense.

1

Complexity and Systems Thinking

1.1. Introduction: complexity as a problem

Perception of the world brings about the feeling that it is a giant conundrum with dense connections among what is viewed as its parts. As human beings with limited cognitive capabilities, we cannot cope with it in that form and are forced to reduce it to some separate areas which we can study separately.

Our knowledge is thus split into different disciplines, and over the course of time, these disciplines evolve as our understanding of the world changes. Because our education is conducted in terms of this division into different subject matters, it is easy not to be aware that the divisions are man-made and somewhat arbitrary. It is not nature that divides itself into physics, chemistry, biology, sociology, psychology and so on. These silos are so ingrained in our thinking processes that we often find it difficult to see the unity underlying these divisions.

Given our limited cognitive capabilities, our knowledge has been arranged by classifying it according to some rational principle. Auguste Comte (1880) in the 19th Century proposed a classification following the historical order of the emergence of the sciences, and their increasing degrees of complexity in terms of understanding their evolving concepts. Comte did not mention psychology as a science linking biology and the social sciences. He did not regard mathematics as a science but as a language which any science may use. He produced a classification of the experimental sciences into the following sequence of complexity: physics, chemistry, biology, psychology and social sciences.

Physics is the most basic science, being concerned with the most general concepts such as mass, motion, force, energy, radiation and atomic particles. Chemical reactions clearly entail the interplay of these concepts in a way that is intuitively more intricate than the isolated physical processes. A biological phenomenon such as the growth of a plant or an embryo brings in again a higher level of complexity. Psychology and social sciences belong to the highest degree of human-felt complexity. In fact, we find convenient to tackle the hurdles we are confronted with by establishing a hierarchy of separate sciences. As human beings, biology is of special interest for us because it studies the very fabric of our existence.

In physics, the scientific method inspired by the reductionistic approach from Descartes’ rule has proved successful for gaining knowledge. Chemistry and biology can rely on physics for explaining chemical and biological reactions; however, they are left with their own autonomous problems. K. Popper shares this point of view in his intellectual biography (Popper 1974):

"I conjecture that there is no biological process which cannot be regarded as correlated in detail with a physical process or cannot be progressively analysed in physiochemical terms. But no physiochemical theory can explain the emergence of a new problem… the problems of organisms are not physical: they are neither physical things, nor physical laws, nor physical facts. They are specific biological realities; they are ‘real’ in the sense that their existence may be the cause of biological effects".

1.2. Complexity in perspective

1.2.1. Etymology and semantics

The noun complexity or the adjective complex are currently used in many oral or written contexts when some situations, facts or events cannot be described and explained with a straightforward line of thought.

It is always interesting to investigate the formation of a word and how its meaning has evolved in time in order to get a better understanding of its current usage.

Complex is derived from Latin complexus, made of interlocked elements. Complectere means to fold and to intertwine. This word showed up in the 16th Century for describing what is composed of heterogeneous entities and was given acceptance in logic and mathematics (complex number) circa 1652. At the turn of the 20th Century, it became closer to complicated and used in chemistry (organic complexes), economics (1918) and psychology (Oedipus complex, inferiority complex – Jung and Freud 1909/1910). Complicated is derived from Latin complicare, to fold and roll up. It was used in its original meaning at the end of the 17th Century. Its current usage is analogous to complex: what is complex – theory, concept, idea, event, fact and situation – is something difficult to understand. It is related to human cognitive and computable capabilities, which are both limited.

A telling instance is the meaning given to the word complex in psychology: it is a related group of repressed ideas causing abnormal behavior or mental state. It is implicitly supposed that the relations between these repressed ideas are intricate and are difficult to explain to outside observers.

In the Oxford dictionary, complex is described by three attributes, i.e. consisting of parts, composite and complicated. These descriptors are conducive to exploring the relationships of the concept of complexity with two well-established fields of knowledge, i.e. systems thinking and structuralism. That will be done in following sections.

1.2.2. Methods proposed for dealing with complexity from the Middle Ages to the 17th Century and their current outfalls

Complexity is not a new issue in quest of what is knowable to humans about the world they live in.

Two contributors from the Middle Ages and the Renaissance will be considered here, William of Ockham and René Descartes in their endeavors to come to terms with complexity. Their ideas are still perceptible in the present times.

1.2.2.1. Ockham’s razor and its outfall

William of Ockham (circa 1285–1347) is known as the More Than Subtle Doctor, English scholastic philosopher, who entered the Franciscan order at an early age and studied at Oxford. William of Ockham’s razor (also called the principle of parsimony) is the name commonly given to the principle formulated in Latin as "entia non sunt multiplicanda praeter necessitatem" (entities should not be multiplied beyond what is necessary). This formulation, often attributed to William of Ockham, has not been traced back in any of his known writings. It can be interpreted as an ontological principle to the effect that one should believe in the existence of the smallest possible number of general kinds of objects: there is no need to postulate inner objects in the mind, but only particular thoughts, or states of mind, whereby the intellect is able to conceive of objects in the world (Cottingham 2008). It can be translated into a methodology to the effect that the explanation of any given fact should appeal to the smallest number of factors required to explain the fact in question. Opponents contended that this methodological principle commends a bias towards simplicity.

Ockham wrote a book, Sum of Logic. Two logical rules, now named De Morgan’s laws, were stated by Ockham. As rules or theorems, the two laws belong to standard propositional logic:

1) [Not (p And q) ] is equivalent to [Not p Or Not q].

2) [Not (p Or q)] is equivalent to [Not p And Not q].

Not, And and Or are logic connectors; p and q are propositions.

In other words, the negation of a conjunction implies, and is implied by, the disjunction of the negated conjuncts. And the negation of a disjunction implies, and is implied by, the conjunction of the negated disjuncts.

It can be figured out that Ockham, who was involved in a lot of disputations, felt the need to use the minimum of classes of objects in order to articulate his arguments in efficient discussible ways on the basis of predicate logic. Predicate logic allows for clarifying entangled ideas and arguments, and producing a rational chain of conclusions that can be understood by a wide spectrum of informed people.

Different posterior schools of thought can be viewed as heirs apparent to the principle of Ockham’s razor, among many others, the ontological theory and Lévi-Strauss’ structuralism.

The word ontology was coined in the early 17th Century to avoid some of the ambiguities of metaphysics. Leibniz was the first philosopher to adopt the word. The terminology introduced by 18th Century came to be widely adopted: ontology is the general theory of being as such and forms the general part of metaphysics. In the usage of 20th-Century analytical philosophy, ontology is the general theory of what there is (MacIntyre 1967).

Ontological questions revolve around:

– the existence of abstract entities (numbers);

– the existence of imagined entities such as golden mountains/ square circles;

– the very nature of what we seek to know.

In the field of organization theories, ontology deals with the nature of human actors and their social interactions. In other more abstract words, ontology aims to establish the nature of entities involved and their relationships. Ontology and knowledge go hand in hand because our conception of knowledge depends on our understanding of the nature of the knowable.

The ontological commitment of a theory is twofold:

– assumptions about what there is and what kinds of entities can be said to exist (numbers, classes, properties);

– when commitments are paraphrased into a canonical form in predicate logic, they are the domains over which the variables are bound to the theory range.

When it comes to complexity, the ontological description of an entity should refer to its structure (structural complexity) and its organization (organizational complexity). This is in line with the mindset in German culture to describe a set of entities by two concepts, i.e. Aufbau (structure) and Ablauf (flows of interactions inside the structure). An entity can be a proxy that represents our perception of the world. According to the purpose, a part of the world can be perceived in different ways and can turn out to be modeled by different sets of ontological building blocks.

Lévi-Strauss was a Belgian-born French social anthropologist and leading exponent of structuralism, a name applied to the analysis of cultural systems in terms of the structural relationships among their elements. Lévi-Strauss’ structuralism was an effort to classify and reduce the enormous amount of information about cultural systems to ontological entities. Therefore, he viewed cultures as systems of communication and constructed models based on structural linguistics, information theory and cybernetics to give them an interpretation. Structuralism is a school of thought which evolved first in linguistics (de Saussure 1960) and did not disseminate outside the French-speaking intellectual ecosystem.

1.2.2.2. René Descartes

René Descartes (1596–1650), French philosopher and mathematician, was very influential in theorizing the reductionistic approach to analyzing complex objects. It consists of the view that a whole can be fully understood in terms of its isolated parts or an idea in terms of simple concepts.

This attitude is closely connected to the crucial issue that science faces, i.e. its ability to cope with complexity. Descartes’ second rule for properly conducting one’s reason divides up the problems being examined into separate parts. This principle most central to scientific practice assumes that this division will not dramatically distort the phenomenon under study. It assumes that the components of the whole behave similarly when examined independently to when they are playing their part in the whole, or that the principles governing the assembling of the components into the whole are themselves straightforward.

The well-known application of this mindset is the decomposition of a human being into the body and the mind localized in the brain. It is surprising to realize that Descartes’ approach to understanding what a human being is and how (s)he is organized remains an issue discussed by philosophers of our time. The issue of mind–body interaction with the contributions of neurosciences will be developed in another section.

The argument supporting this approach is to reduce the complexity of an entity in reducing the variety of variables to analyze concomitantly. It is clear that this methodology can be helpful in a first step but understanding how the isolated parts interact to produce the properties of the whole cannot be avoided. This type of exercise can appear very tricky. This way to approach complexity contrasts with holism. Holism consists of two complementary views. The first view is that an account of all the parts of a whole and of their interrelations is inadequate as an account of the whole. For example, an account of the parts of a watch and of their interactions would be incomplete as long as nothing is said about the role of a watch as a whole. The complementary view is that an interpretation of a part is impossible or at least inadequate without reference to the whole to which it belongs.

In the philosophy of science, holism is a name given to views like the Duhem–Quine thesis, according to which it is the whole theories rather single hypotheses that are accepted or rejected. For instance, the single hypothesis that the earth is round is confirmed if a ship disappears from view on the horizon. However, this tenet presupposes a whole theory – one which includes the assumption that light travels in straight lines. The disappearance of the ship, with a theory that light-rays are curved, can also be taken to confirm that the earth is flat. The Duhem–Quine thesis implies that a failed prediction does not necessarily refute the hypothesis it is derived from, since it may be preferable to maintain the hypothesis and instead revise some background assumptions.

The term holism was created by Jan Smuts (1870–1950), the South African statesman and philosopher, and used in the title of his book Holism and Evolution (Smuts 1926). In social sciences, holism is the view that the proper object of these sciences is systems and structures which cannot be reduced to individual social agents in contrast with individualism.

As a mathematician, Descartes has developed what is called analytical geometry. Figures of geometric forms (lines, circles, ellipses, hyperboles, etc.) are defined by analytical functions and their properties described in terms of equations in Cartesian coordinates measured from intersecting straight axes. This is an implicit way to facilitate the analysis of complex properties of geometric forms along different spatial directions.

1.3. System-based current methods proposed for dealing with complexity

1.3.1. Evolution of system-based methods in the 20th Century

All current methods used to deal with complexity have been evolved in the 20th Century within the framework of what is called the system theory.

The system concept is not a new idea. It was already defined in the encyclopedia by Diderot and d’Alembert published in the 18th Century in Amsterdam to describe different fields of knowledge. In astronomy, a system is assumed to be a certain arrangement of various parts that make up the universe. The earth in Ptolemy’s system is the center of the world. This view was supported by Aristotle and Hipparchus. The motionless sun is the center of the universe in the Copernicus’ system. In the art of warfare, a system is the layout of forces on a battlefield