The Engineering Mindset Part 3: Complex or Complicated? Adapting Strategies

Chris and Penny Hamlin on how a complexity-based approach can be helpful in avoiding and mitigating unintended consequences

IN COMPLEX systems, unintended consequences and unpredictable outcomes are not only likely within the system itself, but also inevitable in interconnected systems and the wider environment. These effects can be positive or negative, can often be anticipated, and can be managed to some degree.

Using C-THRU research into greenhouse gas emissions we will illustrate how different boundaries and frames of reference produce surprisingly different and sometimes contradictory conclusions. As an engineer, how do you reconcile this and how do you determine which boundaries are appropriate?

Should you focus only on what’s within your direct control, or consider the wider system and find ways to influence it? This paradox is common in many complex challenges. We explore how collaboration can expand your influence and ability to act in the most holistically beneficial manner. The key takeaway is understanding how to direct your energy for the greatest impact.

Span of control

In any professional context, it is important to recognise the scope of your control or influence, which typically falls into one of four buckets:

1. Direct control: Factors you can manage and act upon directly
2. Influence: Areas where you can guide or affect outcomes, though without full control
3. No control: Situations where you are unable to exert any influence due to external constraints or scale
4. Uncontrol: Events or conditions that cannot be prevented, only responded to when they occur

Tasks under your direct control are relatively straightforward; you can take deliberate action that typically produces predictable results. However, areas of influence are more complex; while you can steer these toward a desired result, there is no guarantee of success, as the outcome depends both on external variables beyond your command and your ability to act through others.

For factors beyond your control or influence, the challenge lies in recognising that immediate intervention is not feasible. In such cases, acceptance is necessary, though circumstances may evolve over time, or the problem can potentially be decomposed into smaller, manageable parts.

Finally, there are events or conditions that are inevitable and uncontrollable but have direct impact on you and your context.

In these instances, the best approach is to be ready to react as proactive control is impractical.

System boundaries

The degrees of control and influence you have are closely linked to the boundaries of the system in which you operate. To recap, in a complex system the boundaries are ill-defined, unclear and porous with information and resources flowing across them.

When we define systems boundaries within any engineering context, it is crucial to consider what flows across these borders – whether it is materials, people, knowledge, money, or energy.

It can be useful to think about the boundaries that define the extent of your control, influence and irrelevance, and consider how your actions impact the flows. This provides valuable insight into where unintended consequences, both positive and negative, might arise.

However, it is essential to appreciate that perceptions of whether a situation is good or bad depend on the perspective of those impacted. For instance, the implications of sea-level rise will be experienced very differently in Tuvalu, a low-lying island nation, compared to Denver, the Mile-High City.

Ultimately, the evaluation of these consequences depends on who gets to choose and interpret them, highlighting the importance of inclusivity in assessing engineering actions and their broader impacts.

C-THRU example

Mitigating greenhouse gas (GHG) emissions from plastics is clearly a complex issue, and the C-THRU study revealed that the impact varies significantly depending on where you draw your system boundaries.

One piece of research, based on the US market in 2021, evaluated the GHG emissions of 16 plastic products and their current alternatives.

Using a cradle-to-grave approach, the study considered each product’s production, transportation, use, and end-of-life phases.

Each stage contributes to overall emissions, and the C-THRU team demonstrated how strategies to reduce emissions differ depending on the chosen boundaries of the system under analysis.

Innermost boundary

In the production phase, the emissions from plastic products are, on the whole, higher than their respective alternatives. The charts below show that emissions from sprayed polyurethane foam (SPF) are higher than those of fiberglass insulation. Similarly, production emissions from high-density polyethylene (HDPE) hand soap containers are significantly higher than glass alternatives. For industrial drums, the manufacture of HDPE and steel options are largely equivalent with plastic and actually slightly better.

Intermediate boundary (production, transportation, and use)

If we extend the boundary to include transportation and use phases of the same products we see a different picture.

Fiberglass insulation leads to higher use emissions because it allows more heat to pass through, requiring more energy to be used for heating and cooling over time compared to the impermeable SPF insulation.

With hand soap there are higher transportation impacts from using 15-20 HDPE bottles but no use-phase emissions. Washing and refilling the glass bottles 15-20 times with flexible polypropylene pouches has significant use-phase emissions but overall impact is still 25% lower than HDPE. Refurbishment of steel drums generates higher use-phase emissions than HDPE drums, reinforcing the case for the latter.

The message is now quite different with emissions from plastic products higher only in one chart. Reducing production emissions is still important, but there are also other factors and mitigations that need consideration.

Widest boundary (including end-of-life)

However, for a complete picture of the emissions of a product, the end-of-life phase must be taken into account – this encompasses landfill, waste-to-energy and recycling.

For building insulation, since the model assumes both materials are landfilled at the end of life, there is little change in emissions, with SPF still having significantly lower overall emissions than fiberglass. For liquid soap, while HDPE bottles’ waste-to-energy emissions are offset by recycling, their overall emissions remain higher than refilling glass containers. However, if HDPE bottles are reused with polypropylene pouches (like those used to refill glass bottles), emissions drop by another 15%, making plastic better overall than glass in this case. For industrial drums, steel drums are recycled at a higher rate than HDPE drums, and using recycled steel reduces demand for virgin steel, resulting in lower emissions. Over ten years, steel drums have 25% less impact than HDPE drums. The end-of-life processes and the extent of recycling options available therefore significantly impacts the overall emissions of plastic products and their alternatives. As you can’t always know the end-of-life outcome, how can you know which action is the best to take?

System behaviour

The C-THRU research highlights how different perspectives and system boundaries can shape the outcomes of decisions. Switching from plastics to alternative materials might reduce emissions during the production phase but could significantly increase overall emissions during the use or end-of-life stages. A comprehensive understanding of the impact across all stages of a product’s life cycle enables more informed and effective actions.

Clearly, although C-THRU focused on GHG emissions, the impacts of plastics on oceanic pollution and biodiversity loss are also important areas for similar consideration. The research highlights the importance of examining the impacts and unintended consequences of individual policy interventions and understanding the potential trade-offs required to combat planetary crises.

Collaborations and coalitions

In a complex system, your spans of control and influence directly affects your ability to identify and respond to unintended consequences. To increase both awareness and impact, it is essential to form coalitions and collaborate with those who have overlapping and adjacent areas of influence.

For example, an HDPE manufacturer can take steps to reduce emissions within its own production process and may even influence its supply chain and transportation logistics. However, its ability to impact the use phase of products made from its materials is more constrained, with limited insight into end-of-life management. By partnering with other stakeholders, such as retailers or municipal waste services, the manufacturer can gain greater understanding and visibility. Coordinating efforts across these sectors could reveal new avenues for influence, drive innovation, and potentially lead to alternative business models that support broader systemic change.

Being aware of impacts beyond your immediate control is key to reducing negative effects and maximising positive ones. Building partnerships across different systems with diverse groups expands the areas where you can have influence and discover new possibilities. Together, the group can act more effectively in uncertain situations, overcoming limitations that individuals face when working alone. Areas that once seemed out of your control can become tangible and manageable, turning previously insurmountable problems into opportunities for action.

Since no one can fully understand the larger system, how do you know your coalition is good enough? In truth, you can’t know, but collaborating with like-minded partners creates a more dynamic and far-reaching system than you might expect. In a future article in this series, we will describe a model for monitoring and proactively intervening to shape system-wide outcomes following a process that we call Guided Evolution.

By focusing on what you can control and trusting coalition partners to handle what is within their reach, everyone can better identify and respond to unintended consequences. Expanding collaboration also opens up the chance to consider a whole new range of unforeseen outcomes, enabling the complex system to self-organise and creating space for the emergence of coherent and mutually reinforcing solutions.

The greater the diversity and reach of the expanded complex system created through collaboration and cooperation, the greater the potential for aligned and coherent impact.

Conclusion

We know that complex systems are often nested within other complex systems. While individuals and organisations within these systems have limited control and influence, their actions can have widespread effects throughout the system of systems.

Being aware of this helps us anticipate where unexpected and unintended consequences might arise and consider where we have influence or control over these. It gives us insight into what we need to monitor and where we need to look to see if the system is behaving as we had anticipated and expected. When unexpected or unintended outcomes are identified, we can assess whether and how we mitigate their negative effects or amplify the positive ones.

By recognising our own boundaries, we can use partnerships with others to expand our influence and control through collective, coordinated action. As we already know, in complex systems, the focus must be on guiding principles and general direction, rather than fixed outcomes, and this is true for the collaborations and coalitions that we develop.

In conclusion, to operate effectively in complex systems as an individual engineer or organisation you must:

• understand your boundaries of control and influence
• identify the potential for unintended and unexpected consequences and establish monitoring processes to detect these
• build diverse coalitions with others who have a similar and aligned collaborative mindset to expand your ability to incite change and amplify or mitigate emergent system behaviour