Systems Resilience / Aiming to build a body of knowledge for building resilient systems

Research Outline

In the Great East Japan Earthquake of March 11, 2011 and the Fukushima Daiichi Nuclear Power Station incident that followed, the word “unexpected” was used often. In order for our society to be sustainable, it must be able to cope flexibly to various external events. In this regard, we must focus not only on systems that do not break, but also on the important concept of how quickly they can recover if they do break. The word resilience refers to the capability of flexibly recovering even if there is a temporary loss of function in response to a major change in the environment.

In the long history of the earth, living organisms have responded to many events. Even when certain species became extinct, other species filled the gap to maintain the ecosystem. Not only living organisms, but also human-made items such as the Internet, social systems such as economies and finance, and companies and organizations all maintain their activities by responding flexibly to external disturbances.

Do these resilient systems have any common characteristics? Can we construct a body of knowledge for designing and operating resilient systems based on such characteristics, or on scientific principles? Through our researcher network spanning a wide range of fields, we will answer these questions. In addition, with this body of knowledge, we should be able to make human societies safer and more sustainable. (Project Director: Hiroshi Maruyama (Institute of Statistical Mathematics))

Electricity Demand and Supply in Yokohama City.
Compartmentalized Power Grid.
Flood map of Bangkok, Thailand.

Project Objectives

In this project, “Systems Resilience,” our ultimate goal is to construct a body of knowledge (BOK) for designing and operating resilient systems. To do so, we believe that we must have the following three items.

Project Framework

Our first task in this project is to collect insights on resilience from the widest variety of fields possible and synthesize them. In this way, we will elicit a clearer definition of what resilience is, and metrics for it. At the same time, we will formularize the insights we obtain in an executable computation model as an approach for shedding light on the essence of resilience. Our project will therefore proceed with research on the following four subthemes.

Introduction of Subthemes

1. Resilience integration strategy

We are not interested in small ordinary disturbances. We can handle small disturbances adequately using well-known methods such as control engineering or reliability engineering. Of the two aspects of resilience, this corresponds to the concept of resistance.

Meanwhile, we are unable to cope well with rare massive events such as the Great East Japan Earthquake using the concept of resistance only. This is because the cost of resistance would become extremely high. For example, building a costal defense barrier high enough to withstand the tsunami that devastated the Tohoku area after the earthquake would be unrealistically expensive, even if it were possible from an engineering perspective.

Economist Kei Takeuchi in his book, “What is Coincidence—Its Positive Meaning,” says that with respect to extremely rare events, we should assume they will not happen, and if they do occur, we should reallocate their negative effects in society. We must consider the optimal meta strategy on how we should distinguish between the strategies of resistance and recovery.

In this subproject, we aim to create a meta strategy index for how much cost should be respectively allocated to resistance and recovery in order to build a resilient system. (Research Leader: Hiroshi Maruyama (Institute of Statistical Mathematics))

2. Resistance in living organisms

Biological systems need to have built-in robustness to enable them to endure constant exposure to stress and repeated dangerous situations. We focus on the following three aspects in biological systems, which appear mutually contradictory at first glance, to explain the resistance and recovery mechanisms in living organisms. (Research Leader: Hiroshi Akashi (National Institute of Genetics)

3. Resilience in Social Systems

Order formation in social systems is usually considered to be carried out in a top-down manner by the state through the definition of a system of laws. However, in complex modern social systems, in order to respond more flexibly to unexpected events, it is necessary to review this top-down order formation hypothesis. Recognizing that only the people suffering because of a problem possibly have the best knowledge to solve it; it is necessary to introduce bottom-up type rule formation that can respond flexibly to changes in a situation.

Our focus is on the concept of co-regulation, which is receiving more emphasis recently, particularly in Europe. Co-regulation aims at a public-private collaborative system formation method, where regulating for new social problems is carried out by prioritizing a response based on voluntary regulation by corporations and other operating in the field concerned, with the government compensating for risks and imperfections.

In this subproject, we look mainly at fields of the legal system such as cyber security and privacy protection and aim to construct a methodology for bottom-up type order formation using co-regulation to realize resilient social systems. (Representative Researcher: Hitoshi Okada (National Institute of Informatics))

4. Resilience Computation Model

To evaluate the resilience of complex and large-scale systems, it is essential to have a computational methodology for revealing the various dynamic characteristics in a general mathematical model. We will focus particularly on large-scale systems, treating them as various network models (Boolean, Constraint, Bayesian networks, etc.) to devise modeling techniques that efficiently express dynamic changes in the system composition with respect to unexpected events.

Further, in the network model, we aim to construct a computational theory for analyzing dynamics, such as 1) sensitivity to external disturbance and system variables, 2) impossibility of predicting system status, 3) introduction of the concept of grammatical and semantic gap between system states, 4) network configuration renewal to satisfy soft constraints as far as possible, and 5) multipurpose optimization problem with multiple agents. In doing so, we aim to clarify the general principles for designing resilient systems. (Research Leader: Katsumi Inoue (National Institute of Informatics))

Related articles

Research View 025

Harnessing the Power of Resilience

[Systems Resilience] Hiroshi Maruyama (Professor, The Institute of Statistical Mathematics)

Prof. Hiroshi Maruyama of the Institute of Statistical Mathematics has led our organization's research project, “Systems Resilience," for four years. Ahead of the end of the project in March 2016, it co-hosted

Research View 017

Evolutionary Strategies of Diverse Wildlife

[Systems resilience] Jun Kitano (National Institute of Genetics - Project Associate Professor)

The Great East Japan Earthquake that occurred in 2011 (Heisei 23) resulted in a giant tsunami that struck the Tohoku Region of Japan.

Research View 001

How Should Science Deal with the Unexpected?

[Systems Resilience] Hiroshi Maruyama (Assistant Director and Professor, The Institute of Statistical Mathematics)

The Great East Japan Earthquake occurred on March 11, 2011. Since then, corporations and others have made advances in their measures to see how they can use Information and Communication Technology (ICT) to assist in disaster recovery.