Toolkit: Complex Systems Toolkit.

Author: Onyekachi Nwafor (KatexPower).

Topic: Emergence in complex systems.

Title: Understanding emergence in complex engineering systems.

Resource type: Knowledge article.

Relevant disciplines: Any.

Keywords: Available soon.

Licensing: This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. 

Downloads: A PDF of this resource will be available soon.

Who is this article for?: This article should be read by educators at all levels in higher education who are seeking to provide students with an overall perspective on complex systems in engineering. 

Related INCOSE Competencies: Toolkit resources are designed to be applicable to any engineering discipline, but educators might find it useful to understand their alignment to competencies outlined by the International Council on Systems Engineering (INCOSE). The INCOSE Competency Framework provides a set of 37 competencies for Systems Engineering within a tailorable framework that provides guidance for practitioners and stakeholders to identify knowledge, skills, abilities and behaviours crucial to Systems Engineering effectiveness. A free spreadsheet version of the framework can be downloaded. 

This resource relates to the Systems Thinking and Critical Thinking INCOSE competencies. 

AHEP mapping: This resource addresses several of the themes from the UK’s Accreditation of Higher Education Programmes fourth edition (AHEP4): Analytical Tools and Techniques (critical to the ability to model and solve problems), and Integrated / Systems Approach (essential to the solution of broadly-defined problems).  

Learning and teaching resources:

• Santa Fe Institute Complexity Explorer 

• NetLogo Models Library

• Complexity Labs

• Insight Maker

• International Society for the Systems Sciences 

• Complex Systems Society

Premise:

Engineering systems today are increasingly complex, interconnected, and adaptive. To understand and manage them effectively, engineers must move beyond reductionist thinking where systems are broken into isolated parts and adopt systems thinking, which views systems as wholes made up of interacting components. 

At the heart of this perspective lies emergence, a defining characteristic of complex systems. Emergence refers to properties or behaviours that arise from interactions among components but cannot be predicted or understood by examining those components in isolation. Appreciating emergence helps engineers anticipate how individual design decisions can produce system-level outcomes, sometimes beneficial, sometimes negative and unintended. 

This article introduces the concept of emergence as one key characteristic of complex systems, situates it within systems thinking, and provides practical guidance for recognising and managing emergent behaviours in engineering practice.

1. What is a system?:

A system can be defined as “a set of interconnected elements organised to achieve a purpose” (Meadows, 2008). Systems possess structure (components), relationships (interactions), and purpose (function). Engineering systems such as aircraft, power grids, transport networks, or data infrastructures are composed of numerous subsystems that depend on each other. 

Crucially, systems thinking emphasises interdependence and feedback. The behaviour of the whole cannot be fully explained by the behaviour of the parts alone. Properties such as resilience, adaptability, and emergence result from interactions within the system’s structure and environment. Recognising these relationships is essential to understanding how system-level behaviours arise.

2. Understanding emergence:

Emergence describes the appearance of new patterns, properties, or behaviours at the system level that are not present in individual components. These properties are often irreducible: they cannot be explained solely by analysing each part separately (Holland, 2014). 

Researchers distinguish between: 

• Weak emergence – behaviours that are theoretically predictable if all component interactions were known but are practically impossible to compute due to complexity (e.g. traffic flow patterns). 

• Strong emergence – properties that are fundamentally novel and irreducible to component-level descriptions (e.g., consciousness in biological systems). 

In engineering, most emergent behaviours are weakly emergent: complex yet explainable with sufficient data and computational tools such as agent-based modelling or system dynamics. 

A key caveat is that emergence depends on perspective and system boundaries. What seems emergent at one scale (e.g., the stability of a power grid) might appear straightforward when viewed at another. Therefore, engineers must define boundaries and assumptions clearly when analysing emergence. 

3. Why emergence matters in engineering:

Emergence shapes how engineering systems behave, evolve, and sometimes fail. It can produce both desired outcomes (like adaptability or resilience) and undesired ones (like instability or cascading failure). 

Understanding emergence enables engineers to: 

• anticipate how local interactions scale up to global system behaviour; 

• design feedback loops and architectures that promote stability; and 

• identify potential points for intervention when emergent behaviour becomes undesirable. 

For instance, in cyber-physical systems, emergent coordination can enhance efficiency, but it may also create unpredictable vulnerabilities if feedback loops reinforce errors. Engineers therefore must not only observe emergence but learn how to influence it through design and governance. 

4. Recognising and managing emergent behaviour:

• Recognising emergence 

Engineers can identify emergence by looking for: 

• • System-level patterns that do not trace directly to any single component (e.g. global traffic flow or collective oscillations in a power grid). 

• • Unexpected behaviours, such as new failure modes or self-organising phenomena. 

• • Scale-dependent properties, where behaviour changes qualitatively as the system grows or interacts with its environment. 

• • Adaptive or learning responses, where the system adjusts without explicit central control.

• Intervening in emergent systems 

Not all emergence is beneficial. Engineers often need to mitigate unwanted emergent behaviours such as instability or inefficiency while reinforcing desirable ones. Effective approaches include: 

• • Redesigning interactions rather than individual components, focusing on how feedback and connectivity shape outcomes. 

• • Introducing constraints or buffers to dampen runaway feedback loops. 

• • Enhancing diversity and modularity so subsystems can adapt locally without propagating failures globally. 

• • Monitoring system states continuously, using sensors, data analytics, or digital twins to detect emergent behaviour early. 

Managing emergence requires humility: complex systems cannot be fully controlled, only influenced. The goal is to guide system dynamics toward safe and productive outcomes. 

5. Illustrative examples of emergence in engineering systems:

• Network systems 

The Internet exemplifies emergence: billions of devices follow simple communication protocols, yet collectively create a resilient, adaptive global network. No single node dictates its performance; instead, routing efficiency and viral content propagation arise from local interactions among routers and users. 

• Transportation systems 

Urban traffic patterns such as congestion waves, spontaneous lane formation, and adaptive rerouting emerge from individual driver behaviour and infrastructural design. Traffic engineers use simulation models to study how simple decision rules generate complex city-wide flows. 

• Energy systems 

Electrical grids maintain frequency and voltage stability through distributed interactions among generators, loads, and controllers. Emergent synchronisation enables reliability, but loss of coordination can cause cascading blackouts showing both beneficial and harmful emergence. 

• Manufacturing systems 

In smart factories, machines and sensors collaborate autonomously, producing system-wide optimisation in scheduling and quality control. Adaptive algorithms and feedback loops create emergent flexibility beyond what central planning alone could achieve. 

6. Practical guidance for engineers and educators:

For engineers, the key is to design with emergence in mind: 

• focus on local rules that encourage desirable global behaviour; 

• incorporate feedback and sensing to detect changes early; and 

• use modular, diverse architectures to enhance resilience. 

For educators, teaching emergence provides an opportunity to bridge theory and practice. Software such as NetLogo and Insight Maker allows students to visualise emergent behaviour through agent-based and system-dynamics models. Linking engineering examples to ecological, social, or digital systems helps learners appreciate the universality of emergence. 

Conclusion:

Emergence is not an anomaly to be avoided but a natural attribute of complex systems. It challenges traditional engineering by revealing that system behaviour often arises from relationships, not components. 

Understanding emergence equips engineers to recognise interdependencies, design adaptive solutions, and work with complexity rather than against it. By embracing systems thinking, engineers can create technologies that are not only functional but resilient, sustainable, and aligned with real-world dynamics.

References:

• Holland, J.H. (2014). Complexity: A Very Short Introduction. Oxford: Oxford University Press. 

• Johnson, S. (2001). Emergence: The Connected Lives of Ants, Brains, Cities, and Software. New York: Scribner.

• Mitchell, M. (2009). Complexity: A Guided Tour. Oxford: Oxford University Press.

• Bar-Yam, Y. (2003). Dynamics of Complex Systems. Cambridge, MA: Perseus Publishing.

• Ottino, J.M. (2004). Engineering complex systems. Nature, 427(6973), 399. 

• Helbing, D. (2013). Globally networked risks and how to respond. Nature, 497(7447), 51-59. 

Any views, thoughts, and opinions expressed herein are solely that of the author(s) and do not necessarily reflect the views, opinions, policies, or position of the Engineering Professors’ Council or the Toolkit sponsors and supporters.  

Toolkit: Complex Systems Toolkit.

Author: Nafiseh M. Aftah, PhD Candidate (University of Kansas).

Topic: Why integrate complex systems in engineering education? 

Title: Complex systems in a transformational era.

Resource type: Knowledge article.

Relevant disciplines: Any.

Keywords: Available soon.

Licensing: This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. 

Downloads: A PDF of this resource will be available soon.

Who is this article for?: This article should be read by educators at all levels in higher education who are seeking an overall perspective on teaching approaches for integrating complex systems in engineering education. 

Related INCOSE Competencies: Toolkit resources are designed to be applicable to any engineering discipline, but educators might find it useful to understand their alignment to competencies outlined by the International Council on Systems Engineering (INCOSE). The INCOSE Competency Framework provides a set of 37 competencies for Systems Engineering within a tailorable framework that provides guidance for practitioners and stakeholders to identify knowledge, skills, abilities and behaviours crucial to Systems Engineering effectiveness. A free spreadsheet version of the framework can be downloaded. 

This resource relates to the Systems Thinking and Critical Thinking INCOSE competencies. 

AHEP mapping: This resource addresses several of the themes from the UK’s Accreditation of Higher Education Programmes fourth edition (AHEP4):  Analytical Tools and Techniques (critical to the ability to model and solve problems), and Integrated / Systems Approach (essential to the solution of broadly-defined problems).  

Premise:

Engineering education is undergoing a fundamental transformation. The convergence of technological, social, and environmental challenges demands that future engineers move beyond procedural problem-solving toward complex thinking – a mindset capable of navigating uncertainty, interdependence, and dynamic change. This shift has been accelerated by advances in Artificial Intelligence (AI), which have redefined both the nature of engineering practice and the competencies students must develop to thrive in it. 

For scientists and engineers, understanding complex systems is critical for the ability to apply knowledge and techniques across diverse contexts. This is particularly visible in fields such as bioengineering, which depends on advances in chemistry, physics, computing, and other engineering disciplines. Such integration requires designing subsystems where engineering expertise can be meaningfully applied. Complex systems also involve human interaction, introducing unpredictability, feedback loops, and uncertainty. Modern AI-enabled systems—ranging from autonomous vehicles to smart grids and biomedical devices—cannot be fully understood through a single traditional discipline. These systems are not simply complicated; they are interconnected, dynamic, and often nonlinear (Jakobsson, 2025). 

What this means for engineering education and educators:

Across the globe, educators have turned to Problem-Based Learning (PBL) as a central strategy for cultivating systems-oriented thinking. For instance, Tauro et al. (2017) and the case study conducted at Tishk International University demonstrate that integrating PBL within mechatronics education enhances students’ ability to connect theory with practice, encouraging collaboration and creativity in addressing multifaceted engineering problems. Similarly, Watters et al. (2016) show that industry–school partnerships transform classrooms into real-world laboratories, reinforcing the value of experiential learning and knowledge transfer between academia and professional practice. These initiatives reflect a broader movement toward authentic, interdisciplinary engagement, a necessary foundation for understanding and designing complex systems. 

However, adopting PBL and interdisciplinary methods is not only a pedagogical improvement but also an epistemological necessity. As Stegeager et al. (2024) emphasise, educators themselves must evolve from instructors to facilitators, cultivating reflective and adaptive learning environments that mirror the complexity of professional engineering contexts. Mynderse et al. further highlight that when students are given responsibility for solving open-ended problems, they report higher satisfaction and deeper conceptual integration. These outcomes suggest that active learning approaches foster the kind of complex, interconnected reasoning required for contemporary engineering practice. 

In parallel, the AI-driven classroom is transforming the educational landscape. Emerging evidence shows that generative AI tools support personalised learning and immediate feedback, freeing educators to focus on mentorship and creativity (Jaramillo, 2024). Yet this technological advancement also underscores the limits of automation. Machines can model and predict, but they cannot interpret ethical implications, reconcile trade-offs, or integrate human and ecological perspectives. This is where complex thinking becomes indispensable: it enables learners to understand AI not merely as a computational tool but as a component within broader sociotechnical systems. 

The need for complex systems understanding is especially acute in fields such as bioengineering and mechatronics, where technologies intersect with living systems and social contexts. The defining feature of complex systems is the interaction among multiple components that produce emergent, often unpredictable behaviour. For engineering students, grasping these principles means developing the ability to think beyond linear causality and to engage with feedback loops, uncertainty, and adaptive design. 

The imperative to transform engineering education:

In traditional engineering education, students get topics presented in discrete classes. They get trained in thermodynamics and fluid mechanics and they often forget what they have learned by the time they are at the control systems course where there is an opportunity to bring together skills from prior knowledge. This modularised model is already losing its effectiveness in preparing the students for encountering real-world problems. As the adage says, “In theory, theory and practice are the same; in practice, they are not”. Understanding the role of noise, measurement errors, simplifying assumptions and computational errors play an essential role. To this end, it is crucial to centre complex system design and embrace interdisciplinarity to develop a competency that supports life-long, adaptive learning.  

As an example, Aalborg University in Denmark stands as a global exemplary of systems-oriented engineering education. Its PBL model is not an add-on; it is the spine of the entire curriculum. Every semester, students tackle a new problem – often tied to societal needs such as urban planning, environmental sustainability, or healthcare. Students must identify relevant knowledge areas, work collaboratively across disciplines, and reflect on both process and outcome. Faculty report that this structure promotes holistic thinking, resilience, and a sense of professional identity early on the students’ journeys (Kolmos et al. 2008). 

On the undergraduate level, capstones are a common part of engineering education which happens at the late stages of the student’s studies. At Rowan University (New Jersey, USA), Engineering Clinics provide a different but equally powerful model. Students work across all four years on interdisciplinary teams, contributing to faculty research or industry-sponsored projects. These clinics are embedded in the curriculum and require students to engage deeply with current research problems, often involving complex technical and human systems. A junior clinic project, for example, might involve the optimisation of a renewable energy system integrating mechanical, electrical, and computer engineering principles. Therefore, students learn to navigate ambiguity, collaborate with experts, and see the relevance of their disciplinary knowledge in a broader context by confronting the messy nature of real data. 

These are two of many examples where systems thinking is cultivated. Students gain exposure to open-ended problems and practice seeking connection across domains as they encounter the limits of their knowledge. In this fast-moving era, crossing disciplines empowers students for lifelong adaptation, allowing them to incorporate their experiences into any new technological developments. It also encourages treating learning as a collaborative social process, rather than a solo race to secure the first job. 

Educators must do more than just deliver content; they also need to act as facilitators and learn alongside their students. By redesigning the curriculum around design-oriented problems that mirror real-world changes, higher education will better prepare future engineers to face upcoming systemic global challenges.  

Looking ahead:

As artificial intelligence and automation continue to reshape industry, engineering education must also evolve. Integrating complex systems into teaching offers students the opportunity to engage directly with the data-driven ecosystem they will encounter in practice. The goal is not only to produce technically skilled engineers, but also thoughtful stewards of technology who can navigate its broader social and ethical dimensions. 

One ongoing challenge is that independent projects often vary in quality and can be difficult to assess. Without intentional design, students may default to trial-and-error approaches instead of drawing on knowledge from prior courses. At the same time, the pressure to cover extensive technical material can make it difficult to provide the broader systems context essential for modern engineering. Yet when learning is reinforced across the curriculum, students are better prepared for future careers that demand systems-based thinking. 

Experiential, self-directed projects play a crucial role in this preparation. They allow students to choose their own path while working closely with advisors and industry partners. Whether developing a product, designing a system, or engaging with professionals, students gain a perspective that feels different from traditional coursework. This process offers them a glimpse of what it means to think and act like real engineers, fostering both confidence and adaptability as they transition from the classroom to the workplace.

References:

• Jakobsson, Eric et al. (1999) ‘Complex systems: Why and what?’, New England Complex Systems Institute. Available at: https://necsi.edu/complex-systems-why-and-what (Accessed: 16 July 2025). 

• Jaramillo, J., Chiappe, A. (2024) ‘The AI-driven classroom: A review of 21st century curriculum trends’, Prospects. Available at: https://link.springer.com/article/10.1007/s11125-024-09704-w (Accessed: 1 August 2025). 

• Kolmos, A., Du, X., Holgaard, J., Jensen. (2008) ‘Facilitation in a PBL environment. Available at: https://www.researchgate.net/publication/338698720_Facilitation_in_a_PBL_environment (Accessed: 30 July 2025) 

• Mynderse, James A., Shelton, Jeffrey N., (2014) ‘Implementing problem-based learning in a senior/graduate mechatronics course’, Paper presented at 2014 ASEE Annual Conference & Exposition, Indianapolis, Indiana. Available at: https://peer.asee.org/implementing-problem-based-learning-in-a-senior-graduate-mechatronics-course (Accessed: 1 August 2025). 

• Stegeager, N., Traulsen, S., Carvalho Guerra, A., Telléus, P., Du, X. (2024) ‘Do good intentions lead to expected outcomes? Professional learning amongst early career academics in a problem-based program’, Education Sciences, 14(2), p. 205. Available at: https://doi.org/10.3390/educsci14020205. 

• Tauro, F., Cha, Y., Rahim, F., Rasul, M.S., Osman, K., Halim, L., Dennisur, D., Esner, B., Porfiri, M. (2017) ‘Integrating mechatronics in project-based learning of Malaysian high school students and teachers’, International Journal of Mechanical Engineering Education. Available at: https://journals.sagepub.com/doi/full/10.1177/0306419017708636 (Accessed: 1 August 2025). 

• Watters, J., Pillay, H. and Flynn, M. (2016) ‘Industry-school partnerships: A strategy to enhance education and training opportunities. Australia’, Queensland University of Technology. Available at: https://eprints.qut.edu.au/98390/  (Accessed: 30 July 2025). 

Any views, thoughts, and opinions expressed herein are solely that of the author(s) and do not necessarily reflect the views, opinions, policies, or position of the Engineering Professors’ Council or the Toolkit sponsors and supporters.  

Toolkit: Complex Systems Toolkit.

Author: Dr Raja Toqeer CEng, MIET, CMgr, FCMI, FHEA, iPEER (University of Sheffield).

Topic: Methods and tools used to embed complex systems in engineering education. 

Title: High-level overview of complex systems methods and tools.

Resource type: Guidance article.

Relevant disciplines: Any.

Keywords: Available soon.

Licensing: This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. 

Who is this article for?: This article should be read by educators at all levels in higher education who are seeking an overall perspective on teaching approaches for integrating complex systems in engineering education. 

Related INCOSE Competencies: Toolkit resources are designed to be applicable to any engineering discipline, but educators might find it useful to understand their alignment to competencies outlined by the International Council on Systems Engineering (INCOSE). The INCOSE Competency Framework provides a set of 37 competencies for Systems Engineering within a tailorable framework that provides guidance for practitioners and stakeholders to identify knowledge, skills, abilities and behaviours crucial to Systems Engineering effectiveness. A free spreadsheet version of the framework can be downloaded. 

This resource relates to the Systems Thinking, Systems Modelling and Analysis, and Critical Thinking INCOSE competencies. 

AHEP mapping: This resource addresses several of the themes from the UK’s Accreditation of Higher Education Programmes fourth edition (AHEP4):  Analytical Tools and Techniques (critical to the ability to model and solve problems), and Integrated / Systems Approach (essential to the solution of broadly-defined problems). Additionally, this resource addresses the Problem Analysis theme.  

Downloads: A PDF of this resource will be available soon.

Learning and teaching resources:

• Glossary: This article refers to many concepts and terms which are more fully described and explained in this companion resource.

• Introduction to the Modeling and Analysis of Complex Systems – Open Textbook Library – Hiroki Sayama’s open textbook (2015).

• Complexity Explorer – Free courses and tutorials on complexity science (Santa Fe Institute).

• Systems Innovation Resources – Guides and tutorials for applying systems thinking and complexity:

• • Systems Thinking Resources: Best Books, Articles, and Videos to Get Started – IDEO U

• • Systems Thinking Hub | OpenLearn – Open University 

• Accessible modelling/simulation/visualisation tools:

• • Insight Maker | Build simulations and models for free – Free web-based system dynamics and agent-based simulation tool. 

• • NetLogo – Free agent-based modelling platform with an extensive model library

• • Gephi – The Open Graph Viz Platform – Open-source network analysis and visualization software. 

• • NetworkX

• • Stella

• • Vensim

Premise:

From smart cities and power grids to global supply chains, complex systems undeniably form the backbone of modern engineering challenges, integrating diverse technical and human domains to deliver resilient solutions that are capable of addressing emerging global demands. Traditional engineering approaches are limited in their ability to address increasingly complex and nonlinear problems, as they often fail to consider systems holistically. Complex systems exhibit dynamic behaviours and patterns that emerge from interactions within the whole, offering insights that go beyond what can be deduced from individual components (Martin, 2025).  

However, recognising complexity alone is insufficient. To engage meaningfully with such systems, engineers and educators require systematic methods and analytical tools that make the structure, behaviour, and evolution of complex systems more transparent and tractable. Methods such as system dynamics, network analysis, agent-based modelling and causal loop mapping enable the identification of affected points, feedback mechanisms and unintended consequences providing a structured way to explore “what if” scenarios and support informed decision making. Without these tools, understanding remains largely intuitive and fragmented, limiting the capacity to model interactions, predict emergent behaviours or design resilient interventions.  

There are many different ways to model complex systems, each suited to exploring particular types of interactions, timeframes, or behaviours. The following sections outline several commonly used tools and illustrate the contexts in which they can be effectively applied within engineering education. This guidance therefore focuses on the practical application and pedagogical integration of key complex systems methods and tools, with the aim of equipping engineering educators to embed systems thinking effectively in their teaching and practice. 

Systems thinking and mapping tools:

Systems thinking provides a holistic perspective for students to explore the interdependencies, feedback loops, and emergent behaviours that characterise complex engineering challenges. A range of mapping and modelling tools can be used to visualise and analyse system structures and behaviours. These tools can be broadly categorised into three categories: qualitative mapping tools (such as rich pictures and influence diagrams) that support shared understanding and problem framing; causal modelling tools (such as causal loop diagrams) that reveal feedback structures and dynamic behaviour; and quantitative simulation tools (such as system dynamics models) that enable experimentation and testing of hypotheses. 

Rich pictures, influence diagrams and causal loop diagrams are adaptable for both conceptual exploration and analytical modelling in engineering education. Each offers distinct advantages and limitations. Rich pictures are highly flexible, enabling diverse stakeholders to collaboratively capture multiple perspectives of a system. Their visual and narrative style promotes inclusivity and creativity but can lack analytical precision and consistency between users. Influence diagrams provide a more structured representation by showing directional relationships between variables, supporting clearer causal reasoning and decision making. However, they do not capture feedback or temporal dynamics, which limits their use in modelling evolving systems. Causal loop diagrams offer an advantage as they explicitly map, reinforcing and balancing feedback loops, giving powerful insights into system behaviour over time. However, these can become complex and difficult to interpret without adequate guidance and their qualitative nature may oversimplify quantitative relationships. When used in sequence, these tools can scaffold students’ systems thinking skills from exploratory mapping (rich pictures), through structural reasoning (influence diagrams), to dynamic analysis (causal loop diagrams). Embedding this progression in engineering education not only enhances students’ critical and reflective capabilities but also enables them to identify leverage points, anticipate unintended consequences and design resilient solutions that respond effectively to the complexity of real-world complex systems. 

Figure 1 presents a product causal loop diagram illustrating how product quality, sales, investment and profitability interact through reinforcing and balancing feedback loops. Two reinforcing loops (R1 and R2) show how profitability and product quality can drive self-sustaining growth: higher profits enable reinvestment in sales, while improved quality enhances customer satisfaction and market demand, both improving overall performance. In contrast, two balancing loops (B1 and B2) act as stabilising forces. When rapid sales growth strains production capacity, quality declines, prompting corrective investment to restore standards (B1). Meanwhile, as quality improves, it eventually reaches a maximum threshold where further gains lead to diminishing returns (B2), reflecting real-world technological and resource limits. Together, these loops demonstrate the dynamic interaction between growth and constraint in complex systems. The model highlights how feedback processes shape organisational performance and underscore the value of systems thinking for anticipating unintended consequences and supporting sustainable decision making in educational contexts where understanding system dynamics enhances learning and design practice. 

Figure 1. Product causal loop diagram (Credit: Creately)

System dynamics modelling:

System dynamics (SD) models simulate system behaviour over time by representing key elements such as stocks, flows, feedback loops, and time delays. This approach is particularly useful for understanding long-term patterns and testing interventions in complex contexts, such as modelling energy demand, tracking carbon emissions, or optimising supply chain dynamics. By using accessible tools like Stella, Vensim, or Insight Maker, educators can create interactive learning experiences that allow students to experiment with ‘what-if’ scenarios, deepen their understanding of dynamic behaviours, and develop the skills needed to make informed, data-driven decisions. Figure 2 illustrates a dynamic stock-and-flow diagram of a model for new product adoption. The diagram demonstrates how stock and flow structures can capture accumulations and delays within a system, providing insights into how adoption rates evolve over time in response to feedback processes. 

Figure 2. Dynamic stock and flow diagram of model New product adoption (taken from Wikipedia: model from article by John Sterman 2001 - True Software) 

Agent-based modelling:

Agent-Based Modelling (ABM) analyses complex systems by simulating the actions and interactions of many individual “agents” each following simple behavioural rules. Agents can represent people, vehicles, organisations or even machines depending on the context and their collective behaviour gives rise to larger system patterns that are often unexpected or counterintuitive. For example, in a traffic flow model, each car (agent) follows basic rules for acceleration, braking and lane changing. While these rules are simple in isolation, their combined effects can lead to emergent phenomena such as traffic jams or wave-like congestion patterns, behaviours not explicitly programmed into the system. Similarly, in a disease transmission model, each agent might represent a person whose movement and interactions influence infection spread across a population, providing valuable insight into intervention strategies. 

ABM is particularly useful in systems where differences among agents and local interactions matter. Whereas System Dynamics (SD) captures aggregate feedback through mathematical relationships, ABM reveals the distributional and spatial dimensions of system behaviour by modelling individual actions and decisions. Educators may choose ABM to help students see how microscale decisions lead to macroscale outcomes, reinforcing the concept that system-level order often emerges from local and uncoordinated interactions. Open-source platforms such as NetLogo provide accessible environments for teaching these principles, offering pre-built models that allow students to experiment with agent rules and parameters. Through such interactive exploration, engineering students can observe how small behavioural changes can cascade into large-scale effects deepening their understanding of emergence, adaptability and complexity in real-world complex systems. Figure 3 presents a schematic of an agent-based model, illustrating how interactions among individual agents within an artificial environment can lead to emergent system-wide patterns. 

Figure 3. Schematic of an agent-based model, showing how interactions between agents lead to emergent phenomena within an artificial world (Credit to Agent-Based Modeling and the City: A Gallery of Applications, Crooks, A., et al 2021). 

Network analysis and modelling:

Network analysis looks at how the pattern of connections within a system affects how it behaves, performs, and recovers from disruption. Instead of focusing on individual parts, this approach studies the relationships between elements whether they are people, machines, or data points and how these connections shape the overall outcome of the system. In network science, two important ideas help describe how a network is organised: degree distribution and clustering coefficients. Degree distribution shows how many connections (or “links”) each element, known as a node, has. If most nodes have a similar number of links, the network tends to behave in a steady and predictable way. However, if a few nodes have many more connections such as major airports in a flight network, the system can operate very efficiently but may also become more vulnerable if one of those key nodes fails. Clustering coefficients measure how connected a node’s neighbours are to each other. A high clustering coefficient means that a node’s connections are also well connected, forming strong local groups. This structure can improve communication and resilience within the network, though it may also limit flexibility or slow the spread of new information. 

By analysing these features, students learn that the way parts of a system are connected is just as important as the parts themselves. Real-world complex systems examples include power grids, transport networks, and organisational systems, where understanding connectivity helps engineers identify weaknesses and design for greater robustness. Tools such as Gephi and NetworkX make it possible to visualise and measure these network properties, helping turn complex data into clear, interpretable diagrams. Figure 4 shows the structure and properties of a technological network, illustrating how node connectivity and clustering together influence the system’s overall resilience. 

Figure 4. Composition and properties of technological network (Credit to Network Resilience: Definitions, approaches, and applications by Xiaoyu Qi and Gang Mei). 

Conclusion:

Understanding and managing complexity is now an essential skill for modern engineers. By gradually introducing students to different systems thinking tools from qualitative mapping to dynamic simulation and network analysis, educators can help them build a deep and transferrable understanding of how complex systems behave. Each tool offers a different perspective: mapping tools encourage exploration and shared understanding, dynamic models reveal feedback and time-based behaviour, and network analysis exposes structural patterns and resilience. Taken together, these approaches form a developmental pathway that strengthens students’ ability to think critically, reason systematically, and make informed design and management decisions. Embedding this progression within engineering education cultivates curiosity, adaptability, and a mindset equipped to tackle the interconnected social, environmental and technological challenges of the future. In doing so, educators prepare graduates not just to work with complex systems, but to improve and transform them.  

References: 

• Avison, D. E., Golder, P. A. and Shah, H. U. (1992) ‘Towards an SSM toolkit: rich picture diagramming’, European Journal of Information Systems, 1(6), pp. 397–408. doi.org/10.1057/ejis.1992.17 

• Barabási, A.-L. & Albert, R. (1999) ‘Emergence of scaling in random networks’, Science, 286(5439), pp.509-512. doi: 10.1126/science.286.5439.509 

• Bonabeau, E. (2002) ‘Agent-based modeling: Methods and techniques for simulating human systems’, PNAS, 99 (Suppl_3), 7280–7287. doi.org/10.1073/pnas.082080899 

• Checkland, P. (1981) Systems Thinking, Systems Practice, Chichester: John Wiley & Sons. 

• Forrester, J.W. (1997) ‘Industrial dynamics’,  Journal of the Operational Research Society, 48(10), pp.1037-1041.  

• Martin, P. (2025) ‘The EPC Complex Systems Toolkit for Engineering Education – Greater Than the Sum of Its Parts’, Quanser Blog, 18 June 2025. Accessed online 27 August 2025: https://www.quanser.com/blog/engineering-education/complex-systems-toolkit-for-engineering-education/ 

• Sterman, J.D. (2000), Business Dynamics: Systems Thinking and Modeling for a Complex World, Boston, MA: Irwin/McGraw-Hill. Accessed online 27 August 2025: https://dspace.mit.edu/handle/1721.1/102741 

Any views, thoughts, and opinions expressed herein are solely that of the author(s) and do not necessarily reflect the views, opinions, policies, or position of the Engineering Professors’ Council or the Toolkit sponsors and supporters.  

Toolkit: Complex Systems Toolkit.

Author: Dr. Rebecca Margetts (Nottingham Trent University).

Topic: The importance of teaching and learning about complex systems.

Title: The real world is a complex system.

Resource type: Knowledge article.

Relevant disciplines: Any.

Keywords: Available soon.

Licensing: This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. 

Downloads: 

• PDF version

Learning and teaching resources:

• Glossary: This article refers to many concepts and terms which are more fully described and explained in this companion resource.

• Ask an MIT Professor: What Is System Thinking and Why Is It Important? – by MIT xPRO on September 14th, 2022 

• What is a Complex System? (2025) by James Ladyman and Karoline Wiesner: An overview from Bristol University (with James Lambert) here, or the full book published by Yale University Press on JSTOR (which should be accessible via your institution).

Who is this article for?: This article should be read by educators at all levels in higher education who are seeking an overall perspective on teaching approaches for integrating complex systems in engineering education. 

Related INCOSE Competencies: Toolkit resources are designed to be applicable to any engineering discipline, but educators might find it useful to understand their alignment to competencies outlined by the International Council on Systems Engineering (INCOSE). The INCOSE Competency Framework provides a set of 37 competencies for Systems Engineering within a tailorable framework that provides guidance for practitioners and stakeholders to identify knowledge, skills, abilities and behaviours crucial to Systems Engineering effectiveness. A free spreadsheet version of the framework can be downloaded.

This resource relates to the Systems Thinking and Critical Thinking INCOSE competencies.

AHEP mapping: This resource addresses several of the themes from the UK’s Accreditation of Higher Education Programmes fourth edition (AHEP4):  Analytical Tools and Techniques (critical to the ability to model and solve problems), and Integrated / Systems Approach (essential to the solution of broadly-defined problems). 

Premise: 

We live in a complex world. Complexity is a key challenge, captured in leadership terms by the VUCA framework: volatile, uncertain, complex and ambiguous (Lanucha 2024). Engineers have the privilege of creating products and processes for humans to use in this landscape. Each of these likely has numerous parts which interact, as well as interacting with the environment, people, and needing to meet a host of safety, quality, sustainability, ethics, and financial obligations. Traditionally, engineers analyse problems by breaking them down into simple parts. This helps understanding and makes calculations feasible, but it’s easy to lose understanding of the whole system. Any change can easily create a problem elsewhere. From a technical viewpoint, engineers need to understand this interconnectedness in order for their creations to work. In a wider sense, ‘systems thinking’ is a skill central to engineering quality and management techniques, which seek to rationalise the complexity of entire organisations and their ever-changing market pressures.  

The case for understanding systems: 

Systems is perhaps one of the most misunderstood words in engineering. It is often found combined with mathematical modelling or control – topics often perceived as challenging – and is used in other fields like Computer Science, where tools and models are different. In all cases, the idea revolves around a group of interacting or interrelated elements which form a unified whole. Those elements can be physical or information, hardware or software, or any combination of mechanical, electrical, and other engineering domains. Thinking in terms of systems can therefore be thought of as a holistic approach.  

The Engineering Council UK’s AHEP criteria include a systems approach: C/M6 – “Apply an integrated or systems approach to the solution of complex problems.” Several other AHEP criteria also reference complexity and complex problems, which they define as having “no obvious solution and may involve wide-ranging or conflicting technical issues and/or user needs that can be addressed through creativity and the resourceful application of engineering science. The Systems Thinking Alliance (2025) gives a broader definition of complexity as referring to “the condition of systems, objects, phenomena, or concepts that are challenging to understand, explain, or manage due to their intricate and interconnected nature. It involves multiple elements or factors that interact in unpredictable ways, often requiring significant information, time, or coordinated efforts to address.” For these, there is no ‘one-size-fits-all solution’ (Ellis 2025). This is the reality that engineers need to manage by understanding the potential effects on all parts of the system. 

In order to analyse, engineers dissect complexity into manageable components, and educators teach these simple components before moving onto more complex systems. For example, students initially learn basic electrical components, simple beams, rigid bodies, etc. before bringing these together in case studies, and then moving onto topics like mechatronic systems. Historically, engineers specialised on graduation, perhaps becoming a stress engineer or fluid dynamicist in dedicated offices and functional teams.  A design decision by one team could have unintended consequences for another, as well as additional uncertainty. The advent of cross-functional project and ‘matrix’ organisations mitigated against this, and companies have moved towards attribute teams which can consider the balance of behaviour. Even so, some uncertainty remains in the form of assumptions in calculations, changes in material properties with temperature or stress, or small variations in composition and manufacturing tolerances, which can all accumulate. Any parts which are bought ‘off-the-shelf’ or made by other companies under license must be carefully specified. Relationships can be nonlinear – or even chaotic – and contain feedback loops which can amplify changes (Kastens et al 2009). This all increases the risk of a product’s comfort, performance, and safety being impacted in ways that weren’t anticipated. Any problem that doesn’t come to light until the testing phase – late in the design process – represents costly redesigns and delays. In the unlikely event that a problem isn’t captured during testing either, the outcome could be disastrous. 

Systems engineers will bring the product together and establish these complex behaviours through models and testing. Identifying potential problems early in the design phase can save significant money and facilitate better designs. This can be challenging, especially for systems using novel materials or operating in extreme environments, which aren’t accurately captured by standard calculations. Models may be linearised, neglect external forcing, or be derived for an assumed air density or ambient temperature which may not be valid. In recent decades, the engineering industry has moved towards model-based design and virtual prototyping, facilitated by advances in computer tools. These are increasingly sophisticated, but models still need to be built by engineers with an appreciation of complexity and the mechanisms by which a problem could arise. As humans develop new materials and technologies, and explore the limits of what is possible, engineering techniques and calculations need constant revision, and software tools are frequently updated to facilitate this.  

That holistic view of problems has benefits outside of designing engineering artefacts. The manufacturing process is itself a complex system with potentially long supply chains. As is the organisation, which is comprised of numerous people operating in a landscape of financial pressures, employment law, politics and culture. Quality guru William Deming’s 14 Points for Management (Deming 2018) can be viewed as a systems approach to handling this complexity, by breaking down barriers between departments and instigating continuous improvement. Once a product is produced, it exists in a wider world and continues to interact with it. From a sustainability viewpoint, this can be the user and surrounding community, the environmental impact over a product’s lifecycle, and the financial markets which dictate whether a product is viable. It can also be the social, political, and legal landscapes: these can place direct constraints in the forms of laws governing safety and emissions (such as the UK’s legally binding target of net zero by 2050), or through embargos, tariffs, and subsidies. Each country has its own regulations, which can necessitate multiple variations of a product: a good example is cars, which need to be produced in both left- and right-hand drive, satisfy varying safety and emissions regulations, and cater for differing personal and cultural preferences for size, noise, usage and driving styles. Even when not legislated, a company might choose to support fair trade, lead the way in sustainable practices, or refuse to do business with suppliers or regimes they find objectionable – potentially making this a key part of their brand.  

An engineer’s ability to appreciate and understand the wider social and business landscape is a reason why finance and management consultancy companies can often be seen recruiting engineers at student careers fairs. The Sainsbury Management Fellowship (SMF) scheme notably develops UK engineers as industry leaders, and fellows have made a major contribution to the UK’s economic prosperity (RAEng 2025). 

Conclusions:

Complex systems are the “real world” that engineers attempt to understand and design for. They are complicated, interconnected, changing, and uncertain. The well-known part of engineering is analysis: breaking systems into understandable parts. There needs to be a parallel operation where those parts are assembled or integrated into a whole, and that whole interacts with everything around it. This is where unforeseen problems can occur. Systems models and a holistic systems thinking approach can mitigate this risk. A systems approach and ability to manage complexity is a key skill for engineers, and positions them well for other fields like management.   

References:

• Deming, W. Edwards (2018) Out of the Crisis, The MIT Press. Available from: https://doi.org/10.7551/mitpress/11457.001.0001  

• Ellis, Jordan (2025) ‘Making Sense of Chaos: The Role of Systems Thinking in Understanding Complexity’, The Systems Thinking. Available from: https://thesystemsthinking.com/making-sense-of-chaos-the-role-of-systems-thinking-in-understanding-complexity/ (Accessed 2nd September 2025). 

• ESCAP (2019) Introduction to Systems Thinking Principles and Analytical Tools, United Nations ESCAP. Available from: https://www.unescap.org/sites/default/files/Introduction%20to%20systems%20thinking%20tools_Eng.pdf (Accessed 2nd September 20250. 

• Kastens, K., Manduca, C., Cervato, C., & Frodeman, R., Goodwin, C., Liben, L., Mogk, D., Spangler, T., Stillings, N., Titus, S. (2009). How Geoscientists Think and Learn. Eos, Transactions American Geophysical Union. 90. 10.1029/2009EO310001.

• Lanucha, K. (2024) ‘Navigating VUCA: Unraveling the Dynamics of Volatility, Uncertainty, Complexity, and Ambiguity in Organizational Challenges’, University of Cambridge Online. Available from: https://advanceonline.cam.ac.uk/blog/what-is-vuca-and-what-does-it-mean-for-you-and-your-international (Accessed 24/10/2025).

• RAEng (2025) ‘Sainsbury Management Fellowship’, Royal Academy of Engineering. Available from: https://raeng.org.uk/smf (Accessed 2nd September 2025). 

• Systems Thinking Alliance (2025) ‘What is Complexity?’. Available from: https://systemsthinkingalliance.org/glossary/complexity/ (Accessed 24th October 2025). 

Any views, thoughts, and opinions expressed herein are solely that of the author(s) and do not necessarily reflect the views, opinions, policies, or position of the Engineering Professors’ Council or the Toolkit sponsors and supporters.