Optimizing Supply Chains Through System Dynamics Modelling and Simulation: Lessons From the Navy

Introduction

In critical activities, spanning from business to critical organizations operations, as the military ones for example, decision making in face of complexity has been the growing normal. Such complexity poses difficulties for humans due to cognition limitations, as for instance limited memory and processing capacity, and the inability to follow the behaviour of a set of simultaneous varying parameters over time. Moreover, one can only understand cause-and-effect relationships that are close in time and space, which implies a fair inability to understand and follow causal relationships that are not close in time or space. Therefore, and aiming at better decision-making, we do have to rely on formal models to harness the inner complexity of many problem typologies – among them the dynamics of supply chains. Models can be classified into two broad categories: mental and formal. The former ones are non-shared models and cannot be discussed as they are internal to our minds. The later ones are discussable because being formal they become explicit and can be shared under several formats.

From the several approaches to model complex systems, the System Dynamics methodology, developed at Massachusetts Institute of Technology during the 1950s seems quite suitable to understand and help in the management of complex systems. A System is a grouping of parts that work together for a purpose (Forrester, 1968). Also, according to Forrester (1961) our organization´s decisions and policies are the causes of many unintended problems, among them, undesirable oscillations and other problematic patterns affecting the parameters of interest. Hence, the systems view is an adequate approach to deal with supply chain management, a critical issue for some typologies of organizations where the outcomes are strategic, if not at a national level, at an organizational level. Armed forces among other organizations, fall under this category.

It is critical to bring clarity to the fact that one typically divides systems – sometimes systems of systems – in its composing elements, to cope with their inner complexity and to be able to analyse them. But whatever one divides for purposes of analysis, one must integrate again. Oftentimes a system cannot even be understood when split apart. In the same sense the understanding of a heart outside of the body would not allow for a full understanding of such organ.

One of the main thinking revolutions of the 1950s suggests that one cannot fully understand the nature of systems by analysis, statistical or otherwise. Another method is desirable – synthesis. When trying to understand a system the first thing analysts do is to take it apart; when synthesis advocate exactly the opposite. The defenders of deductive approaches may disagree, however all creativity (hence original ideas) come from induction thinking. Deduction is useful, however, to “validate” or refute a hypothesis or set of hypotheses, for a certain time horizon. Analysts even suggest that the product of analysis is knowledge. Knowledge, however, is not always the same as understanding. Knowledge may even lead to the know how it works, but not know why it works.

When one gives up synthetic thinking for the sake of analysis, one starts missing understandability. Systems thinking, fundamentally grounded on logics, is a blend of analysis and synthesis, leading towards an expansionism instead of reductionism doctrine (Forrester, 1961; Senge, 1990). Such approach suggests that in order to build increasing understanding one has to move towards larger systems. With synthesis, knowledge goes from wholes to larger wholes, not the opposite – from wholes down to parts.

Causality implies determinism, and logic ensures determinism. Aristotle´s first law of logic states that in a logical implication, if one denies the consequences (effects), one must deny the precedence. Hence, it is a fundamental law of logic that nondeterminism cannot ensure causality. When considering supply chains, we face a similar dilemma. From one side it is easier to understand activities in the field at tactical level, however there is a continuum from the whole organization, i.e., system, strategy down to tactics and vice-versa, within feedback control influences, where decisions at the top influence what goes into the tactical level and vice-versa (Figure 1).

Figure 1.

The three domains overlap and interact with each other as a continuum