Introduction to Modeling Sustainable Development in Business Processes

In this initial chapter, we discuss the main drivers behind the sustainability transition that many businesses are currently facing.

In this chapter, we discuss the optimization of sustainability problems with one single objective that is subjected to one or more constraints. This kind of problem is called a Single Objective Optimization Problem, abbreviated as SOOP. Linear programming (LP) is the mathematical technique we use for solving a SOOP. In linear programming, ‘linear’ refers to the fact that all the mathematical functions, i.e. the objective function and constraints, are required to be linear. ‘Programming’ refers to planning problems. In this book, linear programming is used to solve business process planning problems such as the maximization of profit or the minimization of environmental or societal impact.

In this chapter, we discuss the differences between a MOOP and a SOOP and present the weighted sum approach and the ε-constraint method to solve convex MOOPs formulated as multi-objective linear optimization problems (MOLPs). Moreover, we highlight the fundamental principles of multi-objective optimization and present a way to explore the Pareto optimal front, a function that contains all the optimal solutions for the MOOP.

The economic pillar of the triple bottom line is commonly quantified by minimizing costs or maximizing profits (Seuring and Müller 2008). For supply chain optimization, minimizing costs often involves a tradeoff between costs in different stages of the supply chain, such as factory inventory, transportation, warehousing, manufacturing, etc. The economic optimization of supply chains was the first to be studied. In this chapter, we will discuss some commonly used economic impact quantification methods for business processes in general, while paying special attention to supply chain management.

In this chapter, we focus on quantifying the environmental pillar of the People-Planet-Profit triple bottom line by introducing life cycle assessment (LCA). Life cycle assessment is rooted in life cycle thinking, which approaches products, services, and production systems holistically in terms of impact on the environment. Life cycle thinking advocates taking into account the environmental impact of every stage in the life cycle, from the extraction of raw materials, material processing, transportation, distribution, consumption, reuse/recycling, to disposal. Life cycle assessment is a well-established analytical method for quantifying the environmental impact of a product, service, or production process. This method is traditionally used to study four types of problems: (i) assessment of individual products to understand their environmental impact, (ii) comparison of process paths in the production of substitutable products or processes, (iii) comparison of alternatives for delivering a given function (Jacquemin et al. 2012), and (iv) to analyze the phases of the product or service life cycle that have a more considerable environmental impact – also known as “hotspots” (Piekarski et al. 2013). Several industries, companies, and associations are actively developing LCA approaches. The European Union (EU) is one of the leading global regions advocating life cycle thinking and assessment to ensure the identification of the best environmental outcome.

A commonly accepted definition for the social dimension is not yet available, mainly because there is no consensus on the meaning of the term ‘social’ (Lethonen 2004). The social dimension is immaterial and, therefore, difficult to analyze quantitatively (Lethonen 2004; Munda 2004). Since many social indicators cannot be quantified, qualitative ranking and scoring are currently used alongside quantitative measures (Klöpffer 2008). A popular multi-criteria decision making (MCDM) method that can be used to quantify such qualitative comparisons is the analytic hierarchy process (AHP) (Saaty 1980). The AHP method requires the social criteria of interest to be selected and rated using pairwise comparisons.

With respect to sustainable development, two currently applicable paradigms need to be challenged. The first one is about how to model the complex world we live in. The Cartesian-Newtonian way of thinking (after René Descartes and Isaac Newton) is still dominant. Cartesian-Newtonian thinking, also known as analytical thinking, is based on the premise that a complex problem can be reduced to a set of separate smaller problems. By understanding this set of smaller problems, we can understand the complex problem.

However, Cartesian-Newtonian thinking has some drawbacks. By breaking up complex problems into a set of smaller, more manageable problems, the interaction between the parts gets lost. Moreover, Cartesian-Newtonian thinking is, in general, not very suitable when dealing with nonlinearities, which are a common feature of many real-life environmental and socioeconomic problems.

The second paradigm to be challenged is called the paradigm of economic growth, a term introduced by the ecological economist Herman Daly (1972) to characterize the belief in unlimited growth by mainstream economists.

System dynamics looks at how various elements interact within a system over time and captures the dynamic aspects by incorporating concepts such as stock, flows, feedback, and delays. It provides an insight into the dynamic behavior of a system over time (Tang and Vijay 2001). In this chapter, we will discuss how Causal Loop Diagrams (CLDs) are used to depict the feedback structure of systems. In the next chapter, we will discuss the fundamental modes of behavior of dynamic systems using CLDs to represent their underlying structure.

You may have already observed that many different systems behave the same way. For example, there is a similarity between global population growth over the last 200 years and the annual accumulation of money on a saving account given a fixed yearly net interest rate. Both systems show the same exponential growth behavior because they have a common underlying structure somewhere. The behavior of systems is generally determined by their structure. That structure may consist of feedback loops, stocks and flows, or nonlinearities created by the structure’s interaction with the decision-making processes (Sterman 2000). In this chapter, we discuss the three basic modes of behavior in dynamic systems and their underlying structure: exponential growth, goal seeking, and oscillation. Then we discuss the behavior of more complex dynamic systems and how they interact with some basic structures: S-shaped growth, S-shaped growth with overshoot, S-shaped growth with overshoot and collapse, and tragedy of the commons. These so-called archetypes of system behavior are often observed in the domain of sustainability, as we will show in the following sections.

In the previous chapter, we discussed the structure and state of the dynamic behavior of simple archetype systems. We used CLDs to formalize the discussion about their behavior. We will go one step further in this chapter. By introducing new concepts like stocks, flows, and delays, we will be able to build models that enable us to analyze the dynamic behavior of systems.

In Chap. 10, we discussed three major elements of system dynamics to model real system behavior: stocks, flows, and feedback loops. In this chapter, we will discuss the fourth and last major element, delay. Whenever they become substantial, delays can create instability and oscillation in systems. Since a delay is a process whose output lags behind its input, at least one stock must be involved to decouple the input from the output of the stock. We recognize two types of delays: material and information delays. Within these two types of delays, we can recognize additional differences. First, we will discuss material delays, then information delays, before finally discussing the dynamics of two archetype system behaviors using delays: oscillating behavior and limits to growth behavior.

The behavior of many real-life systems can be described using nonlinear functions. Therefore, nonlinear behavior and its modeling are of major importance in system dynamics. In this chapter, we discuss four important functions that are commonly used to model nonlinear behavior: the table function, MIN and MAX functions, and the IF…THEN, ELSE function. Many more nonlinear functions are available in Vensim^® and are waiting to be explored.