2.3. System resilience

The ability of a system to absorb and recover from disturbances, i.e. system resilience, is important for the continuity of system functioning.

The stability of the system's structure and behaviour over time does not, however, imply a high level of resilience. In fact, it is almost the opposite: the rigidity of the system's structure and the invariability of its functioning in the face of environmental change make the system fragile and prone to collapse when it encounters new and unforeseen disturbances.

Different disciplines have given slightly different definitions for resilience. In engineering, resilience can be defined as the speed with which a (technical) system recovers after a failure: the faster the recovery, the higher the resilience of the system. The resistance of a system, i.e. its ability to withstand an external disturbance without changing its behaviour, can also be considered a component of resilience, especially in engineering sciences.

When dealing with larger and more complex systems, such as a healthcare system, resilience is more difficult to define as a measurable quantity. Complex systems have many internal processes that generate constant change. Large systems are also influenced by a large number of changing factors external to the system. Complex systems therefore never reach a fully stable state, which makes measuring resilience difficult.

Ecosystems are particularly complex systems, and variation is an integral part of their normal functioning, as the example of vole cycles in the previous section illustrates. Canadian ecologist C.S. Holling made a clear distinction between ecosystem stability and resilience, defining resilience as the maintenance of interactions within a system despite changes in the system.

Of course, for the interactions in a system to persist, the elements of the system must also persist, since without the elements there are no interactions between them. As long as fluctuations in the system do not threaten the survival of its elements, the (ecological) system can be considered resilient.

In the case of ecological systems, resilience is therefore about, among other things, the conservation of species and the interactions between them (such as predation). System stability, on the other hand, was defined by Holling as the ability of a system to recover quickly to a state of equilibrium after a disturbance. The system of voles and predators is therefore highly unstable, as the numbers of individuals of the different species are constantly fluctuating, meaning that there is no stable equilibrium. However, the system is resilient - as long as fluctuations in population sizes do not threaten the survival of the species.

Resilience is thus defined differently in different scientific disciplines. In the engineering sciences, which study relatively simple man-made systems, resilience refers to the ability of a system to withstand and recover quickly from disturbances. In the sciences that study complex social or ecological systems, resilience refers to the ability of a system to withstand disturbances and to change and reorganise itself in a way that preserves its essential functions and structures.

Diversity increases resilience

The diversity of actors in a system increases the resilience of the system by reducing the vulnerability of the system to occasional failures.

For example, in diverse ecosystems there are many species that function in essentially the same way: there are many photosynthetic plants, many plant-eating animals, many animal-eating predators, many microbes that decompose dead organisms, and all of these have many parasitic organisms that depend on them. If one species is lost from an ecosystem, for example through the spread of a new disease, the ecosystem will not change much if there are other disease-resistant species in the ecosystem that can replace the lost species. (However, it should also be noted that not all species can be replaced. In many ecosystems there are species of particular importance - so-called keystone species - whose loss or decline will cause major changes in the functioning of the system). Similarly, inter-individual variation, that is, the diversity of individuals, protects the functioning of the system when not all individuals of a species are equally sensitive to the same disturbances. 

This also serves as an example of resilience provided by the modularity of a system, already mentioned earlier in the course: when a system is composed of discrete units (such as different species or individuals), perturbations (such as diseases) are less likely to spread throughout the whole system.

Thus, diversity at different levels of the system increases the resilience of the system if (1) actors differ in their sensitivity to different perturbations (response diversity) and (2) the actors' roles in the system are similar enough to allow them to substitute for each other in the functioning of the system (functional redundancy). However, diversity can also increase the variability within the system (e.g. fluctuations in population sizes), because the number of possible non-linear interactions increases with the number of different actors.

From the above, it is easy to understand that a reduction in diversity reduces the resilience of the system. A system that is simplified to the extreme is one in which each function is performed by only one - or one type of - actor. If a failure knocks out any one of these actors, the entire interdependent system collapses.

Efficiency and resilience

Complex systems, where each task is performed by several independent actors, are highly resilient to disturbances, but not necessarily very efficient. Under stable conditions, differences in efficiency between actors lead to a system with many actors being less efficient than a system where each task is performed only by the most efficient actor (under the prevailing conditions).

In production systems, efficiency is therefore often increased by simplifying the system, i.e. by eliminating redundancies. The remaining structures and actors are naturally chosen so as to be as efficient as possible. Protection against disturbances can be sought by keeping the conditions as favourable as possible for the functioning of an efficient system. However, if conditions change undesirably, the enhanced system will have a very low capacity to withstand and respond to disturbances.

In tightly coupled systems, such as public administrations, diversity and several overlapping actors also impose significant management and coordination costs, and such systems are also not very agile to change when change is needed.

On the other hand, where the system operates without central coordination - as in a market economy - the system's composition of multiple parallel actors is precisely what allows for rapid change, again because of its modularity, i.e. the relatively weak interconnectedness of the actors. In a changing environment, the best performing solutions will become dominant and the worst ones will be crowded out.

However, in a complex system, the diverging goals of separate actors or competition between actors can create friction within the system and reduce its efficiency. Such competitive inefficiencies are found in many social systems, where different people or organisations pursue their own goals, which may not be consistent with the goals of the system as a whole.

Competitive inefficiencies are also found in experimental ecosystems. For example, decomposer fungi competing for the same resources inhibit each other's activity through chemical compounds, with the result that multi-species fungal communities decompose wood more slowly than communities with fewer species.

Since system efficiency and resilience are to some extent conflicting goals, the appropriate trade-off between these conflicting goals must be chosen for each system (of course, systems also have other characteristics that must be taken into account.) Sound design should start from the premise that the more important the needs served by the system and the longer the lifespan of the system, the more emphasis should be placed on resilience rather than on short-term efficiency.

Disruptions to systems serving essential needs - such as food and energy systems - are particularly serious and need to be particularly well protected against. System lifespan, on the other hand, means that the system is likely to face disturbances and changing conditions over time, making resilience and adaptability essential attributes.

A current example of a system that was not designed to withstand future changes is the energy system in central Europe. Central Europe is heavily dependent on a single supplier and source of energy, namely Russian fossil energy (oil and natural gas). This dependency became concrete in spring 2022, when the EU was unable to disconnect from Russian energy as part of the sanctions against Russia following the war in Ukraine.

Would you like to comment something on this section? Voluntary.