674 Ross D. Arnold and Jon P.

Wade / Procedia Computer Science 44 (2015) 669 – 678

and acknowledging that systems are important. In other words, systems thinking implies a rather general and
superficial awareness of systems.

The systems thinking that Jay Forrester is writing about here is not our systems thinking. He appears to be using
the term in a different way. However, note Dr. Forrester’s assertion that systems thinking has no clear definition or
usage – this again reinforces the need for a complete, widely accepted definition.

Comparison of definitions

Although the literature definitions have all failed the System Test, parts of these definitions are still very useful
and relevant. In ultimately defining systems thinking as a system, considering and synthesizing important parts of
these definitions is key. Shown in Figure 2 is a diagrammatic comparison of the different aspects of the literature
definitions discussed above.

Fig. 2: Comparison of Systems Thinking Definitions

In all of these definitions, common elements tend to include interconnections, the understanding of dynamic
behavior, systems structure as a cause of that behavior, and the idea of seeing systems as wholes rather than parts.
This is especially true if extra consideration is given to the comprehensive literature review provided by Hopper &
Stave, which covers a vast arena of literature on the topic, and to Peter Senge and Barry Richmond, well-respected
gurus in the field.
 Ross D. Arnold and Jon P. Wade / Procedia Computer Science 44 (2015) 669 – 678 675

3. Definition of systems thinking as a system

As shown above, the systems thinking definitions proposed by many other authors have focused on the elements
of which systems thinking is made, by defining its components (which has been accomplished quite thoroughly), but
have neglected to detail what systems thinking actually is, and perhaps even more importantly, what systems thinking
does; the “essence” of what makes the system what it is. These characteristics seem to lack some abstract but important
element. This element is the system aspect of systems thinking. Following Richmond’s principle to focus on both the
forest and the trees10 (Richmond, 1994), it appears that many of these definitions may have focused on either the forest
or the trees. Some define systems thinking too vaguely, while others have simplified systems thinking too much, and
in doing so. Both approaches have failed to capture the systemic essence of systems thinking.
Therefore, a new definition is proposed – to define systems thinking as a system by identifying its goal and then
elaborating upon both its elements and the interconnections between these elements.
The idea of defining a system by its goal is ubiquitous and can be found in myriad systems around us – a water
treatment system, a house heating system, a power system or a transportation system – these systems are all named
by their purpose. Thusly, systems thinking can also be defined in terms of its purpose. Once this is done, the elements
and interconnections between these elements will be elaborated further. First, the objective definition:

Systems thinking is a set of synergistic analytic skills used to improve the capability of identifying and
understanding systems, predicting their behaviors, and devising modifications to them in order to produce desired
effects. These skills work together as a system.

The elegance of this definition is in its simplicity and utility. With a bit of background on the nature of a system,
this definition could be presented in an understandable way to an audience with no background in systems science.

The terms included in the definition are themselves defined as the following:

x Systems: Groups or combinations of interrelated, interdependent, or interacting elements forming

collective entities.
x Synergistic: Characteristic of synergy, which is the interaction of elements in a way that, when
combined, produce a total effect that is greater than the sum of the individual elements.
x Analytical skills: Skills that provide the ability to visualize, articulate, and solve both complex and
uncomplicated problems and concepts and make decisions that are sensible and based on available
information. Such skills include demonstration of the ability to apply logical thinking to gathering and
analyzing information, designing and testing solutions to problems, and formulating plans.
x Identify: To recognize as being a particular thing.
x Understand: To be thoroughly familiar with; apprehend clearly the character, nature, or subtleties of.
x Predict: To foretell as a deducible consequence.
x Devise modifications: To contrive, plan, or elaborate changes or adjustments.

Now that systems thinking has been defined by its objective, the definition will be expanded in terms of its content
(elements and interconnections). This content contains, in part, the synthesis of definitions taken from the literature
above. The content is presented as a systemigram, shown in Figure 3.
676 Ross D. Arnold and Jon P. Wade / Procedia Computer Science 44 (2015) 669 – 678

Fig 3: Systems Thinking Systemigram

In this systemigram, thick lines represent strong connections, while thin dotted lines represent weaker, but still
important, connections. It is important to note that the system of Systems Thinking as depicted in this systemigram
operates as a series of continuous feedback loops. That is to say, the system does not cease to function at the final
node. Rather, as each of the elements improves and in turn improves connected elements, Systems Thinking itself
continuously improves.
The elements presented in the systemigram are synthesized from the literature definitions, leveraged primarily from
Sweeney and Sterman15 (2000), Hopper and Stave3 (2008), and Plate8 (2014). The main deviations from those two
works lies in two elements: Reducing Complexity by Modeling Systems Conceptually and Identifying and
Understanding Non-Linear Relationships. All of the elements are described below:

1. Recognizing Interconnections:
This is the base level of systems thinking. This skill involves the ability to identify key connections between parts
of a system. Even highly educated adults without systems thinking training tend to lack this ability8 (Plate & Monroe,
2014).

2. Identifying and Understanding Feedback:

Some of the interconnections combine to form cause-effect feedback loops3 (Hopper & Stave, 2008). Systems
thinking requires identifying those feedback loops and understanding how they impact system behavior 8 (Plate &
Monroe, 2014).

3. Understanding System Structure:

System structure consists of elements and interconnections between these elements. Systems thinking requires
understanding this structure and how it facilitates system behavior6, 10 (Ossimitz, 2000; Richmond, 1994).
Recognizing interconnections and understanding feedback are keys to understanding system structure. Although this
element is not specifically referenced in Hopper and Stave’s (2008) or Plate’s (2014) taxonomies, it can be implied as
 Ross D. Arnold and Jon P. Wade / Procedia Computer Science 44 (2015) 669 – 678 677

a combination of the two aforementioned elements and is referenced in other important works 6, 10 (Ossimitz, 2000;
Richmond, 1994).

4. Differentiating Types of Stocks, Flows, Variables:

Stocks refer to any pool of a resource in a system. This could be physical, like the amount of paint in a bucket, or
emotional, like the level of trust from one friend to another. Flows are the changes in these levels. Variables are the
changeable parts of the system that affect stocks and flows, such as a flow rate or the maximum quantity of a stock.
The ability to differentiate these stocks, flows, and other variables and recognize how they operate is a critical systems
thinking skill.

5. Identifying and Understanding Non-Linear Relationships:

This element represents a deviation from both Hopper and Stave’s (2008) and Richard Plate’s (2014) taxonomies 3,
8
(Hopper & Stave, 2008; Plate & Monroe, 2014). This element refers to stocks and flows of a non-linear nature. It
is conceptually possible to group this element under Differentiating Types of Stocks, Flows, and Variables. However,
the latter seems to imply a linear flow. To avoid confusion, non-linear flows are separated out into this element.

6. Understanding Dynamic Behavior:

Interconnections, the way they combine into feedback loops, and the way these feedback loops influence and
consist of stocks, flows and variables create dynamic behavior within a system. This behavior is difficult to grasp or
understand without systems training8 (Plate & Monroe, 2014). Emergent behavior, a term used to describe
unanticipated system behavior, is one example of dynamic behavior. Differentiating types of stocks, flows, and
variables, as well as identifying and understanding non-linear relationships, are both keys to understanding dynamic
behavior.

7. Reducing Complexity by Modeling Systems Conceptually:

This element represents a deviation from both Hopper and Stave’s (2008) and Richard Plate’s (2014) taxonomies 3,
8
(Hopper & Stave, 2008; Plate & Monroe, 2014). Although it sounds similar to Hopper and Stave’s Using Conceptual
Models, this element is different. This element is the ability to conceptually model different parts of a system and
view a system in different ways. Performing this activity extends beyond the scope of defined system models and
enters the realm of intuitive simplification through various methods, such as reduction, transformation, abstraction,
and homogenization16 (Wade, 2011). Research shows that perceptual wholes can reduce the conscious accessibility
of their parts9 (Poljac, De-Wit, & Wagemans, 2012). This theoretically allows the interpretation of greater complexity
as the mind holds less detail about each whole. This skill could also be viewed as the ability to look at a system in
different ways that strip out excess and reduce complexity.

8. Understanding Systems at Different Scales:

This skill is similar to Barry Richmond’s forest thinking8 (Plate & Monroe, 2014). It involves the ability to
recognize different scales of systems, and systems of systems.

Finally, the proposed definition must be subjected to the System Test. It passes Item 1 of the tests, as it contains a
clearly defined, understandable, and relatable goal. It also passes Item 2: its elements are described in detail. It also
passes Item 3, as interconnections and dependencies between the elements are described in the systemigram.
Therefore, this definition is the first that passes the System Test and successfully defines systems thinking as a system.