The power and fun of modeling complexity

Many years ago, when working as a chemical engineer, I was tasked to advance an existing model of the polymerization process of an engineering plastic and the production thereof. The intention of the model was to understand the polymerization, optimize the chemistry (catalysts, building blocks, additives, etc.) and the production process to boost the output. In addition, the model was used to increase the predictability of scaling from laboratory to pilot and production plant to reduce time-to-market. Although successful within a certain operating window, the model was unable to predict ‘run-offs’ that resulted in multiple shut-downs per year and many costly (and annoying) cleaning activities. The problem had been going on for over a decade. Since, we couldn’t model it, we didn’t fully understand it and couldn’t solve the problem. During our many brainstorm sessions, the observation of a colleague in our laboratory led to an insight, which I used to fundamentally adapt the model. Finally, we were able to predict the ‘run-aways’ and not completely unimportant, solving the issue just costed a fraction of the yearly maintenance budget. It wasn’t a job, it was pure fun to solve these complex problems.

Modeling complexity

Although common practice in R&D, many businesses fail to see the importance of truly understanding the underlying levers that determine their performance. Understanding the key levers that drive success and the root causes of lagging performance are equally important. Some organizations struggle to model across silos, resulting in sub-optimizations within disciplines and typically result in chasing conflicting KPI’s. In fact, a thorough diagnostics phase, to create a deep understanding of the business, should be part of any good strategy development process, as argued by Richard Rumelt, a professor at UCLA’s Anderson School of management, in his book ‘Good Strategy, Bad Strategy’.

The Importance of Modeling for Understanding and Problem Solving

As said, "If you can't model it, you don't understand it. If you don't understand it, you can't identify the origin of the problem and therefore not solve it." This principle is particularly relevant in both business and scientific contexts, where the ability to model complex systems is essential for gaining a deep understanding and effectively addressing issues.

The Role of Modeling in Achieving Understanding

In essence, modeling involves creating a simplified representation of a complex system or process. This can take many forms—mathematical equations, computer simulations, or conceptual diagrams, to name a few. The purpose of a model is to break down something intricate and multifaceted into a more manageable form, enabling us to analyze and understand it better. Our Revenue Operations team in Mitsubishi Chemical Group is part of the Strategy Department and we involve ourselves in functional modelling of our business including the context that we are part of on macro, meso and micro level. Modelling of your own business across silos is not enough, you need to put your business in relation to the transactional environment (meso) and the contextual environment (macro level).

In the realm of science, for example, models are indispensable. They allow researchers to grasp the fundamental workings of natural phenomena, like climate patterns, biological processes, or economic systems. By using models, scientists can identify key variables, predict outcomes, and test hypotheses. Without models, our understanding of these complex systems would remain superficial, limiting our ability to draw meaningful conclusions.

In business, modeling is equally crucial. Consider financial modeling—companies rely on these tools to project revenues, costs, and profits under various scenarios. These models help business leaders make informed decisions by providing a clearer picture of potential future outcomes. The ability to model effectively signifies a deeper understanding of the business environment and the factors that drive success or failure. Your financial results are a result reflection of your ability to create value for your customers and capture that value.

Understanding as a Foundation for Problem Identification

Now, why is this understanding so critical? Simply put, without a solid grasp of how a system functions, it becomes nearly impossible to accurately identify where things might be going wrong. In scientific research, for example, a well-developed model of an ecosystem can help pinpoint the exact factors causing environmental degradation. This understanding is essential for diagnosing problems and proposing viable solutions.

The same principle applies in business. Suppose a company is facing declining sales. To address this issue, the leadership must first understand the market dynamics, customer behavior, and competitive pressures that could be contributing to the downturn. A well-constructed model of these factors can reveal the root cause—whether it's a shift in consumer preferences, pricing issues, or a problem with product quality. Without this level of understanding, any attempt to solve the problem would be based on guesswork rather than evidence. Typically, many initiatives are started to address various phenomenon, spreading resources too thin and not addressing the important levers or with too little attention. As a result, many companies turn towards inorganic growth options to compensate the lack of growth. Consistency and focus on the key issues make the difference.

The Path from Problem Identification to Solution

Once the problem has been accurately identified, thanks to a thorough understanding and effective modeling, finding a solution becomes significantly more straightforward. In the scientific world, this often involves refining the model, testing different interventions, and applying the results to resolve the issue. For instance, if a model reveals that a specific pollutant is damaging an ecosystem, efforts can be focused on reducing or eliminating that pollutant.

In business, solving a problem might involve revisiting and adjusting the model, exploring alternative strategies, and implementing changes. The key point here is that the solution is grounded in a deep understanding of the problem, which is made possible through effective modeling.

Take-away

In summary, the ability to model complexity is fundamental to understanding and problem-solving, whether in science or business. Without a model, understanding remains shallow, the ability to identify problems is limited, and solutions are likely to be ineffective. Likewise Richard Rumult states, the inconvenient truth must be addressed. As challenges become more complex, the importance of robust modeling cannot be overstated. It is a lot of hard work, but any company, any management team owes it to themselves, their employees and customers to make the effort and reap the benefits.