Elsevier Editorial System(tm) for Procedia CIRP

Manuscript Draft

Manuscript Number: PROCIR-D-14-00278R2

Title: Transferability of Self-Healing Principles to the Recovery of Supply Network Disruptions - The
Case of Renesas Electronics

Article Type: SI: CIRP RoMaC2014

Keywords: self-healing; network healing; supply network disruptions; supply chain risk management;
recovery

Corresponding Author: Ms. Marie Katharina Bruening, MSc

Corresponding Author's Institution: Jacobs University Bremen

First Author: Marie Katharina Bruening, MSc

Order of Authors: Marie Katharina Bruening, MSc; J. Henning Buchholz, MSc; Julia Bendul, Prof. Dr.
Manuscript

Available online at www.sciencedirect.com

Procedia CIRP 00 (2014) 000–000

www.elsevier.com/locate/procedia

Robust Manufacturing Conference (RoMaC 2014)

Transferability of Self-Healing Principles to the Recovery of Supply

Network Disruptions – The Case of Renesas Electronics
Marie Brüninga*, J. Henning Buchholzb, Julia Bendula
a
Jacobs University Bremen, Bremen 28759, Germany
b
Deloitte Consulting GmbH, Munich 81669, Germany
* Corresponding author. Tel.: +49-421-200-3077; fax: +49-421-200-3078. E-mail address: m.bruening@jacobs-university.de

Abstract

The topic of supply network disruption recovery has started to receive great attention within the academic as well as the business world. This is
due to increasingly complex supply networks and the rise of natural and man-made disasters in quantity and severity. In this paper, the topic is
approached in an interdisciplinary bio-inspired way. The transferability of biological self-healing principles to the recovery of supply network
disruptions is analysed and first propositions are derived. A single case study of the Japanese microcontroller company Renesas Electronics is
analysed with regard to the developed propositions. A strategically important plant of Renesas was severely damaged by the triple disaster in
Japan in 2011 which led to a disruption of the company‟s supply network. Five out of six propositions regarding the transferability of biological
self-healing are, at least partly, proven. Furthermore, the importance of close collaboration within a network of suppliers, customers and
competitors is emphasized. The results make further research in this academically still underdeveloped field promising, especially with regard
to implications and strategies for supply chain risk managers.
© 2014 The Authors. Published by Elsevier B.V.
Selection and peer-review under responsibility of the International Scientific Committee of “RoMaC 2014” in the person of the Conference
Chair Prof. Dr.-Ing. Katja Windt.

Keywords: self-healing; network healing; supply network disruptions; supply chain risk management; recovery

1. Introduction The second development is related to the rise in quantity

Currently there are two parallel developments in the field
of supply network disruptions, which start to receive great
attention in the international business world and in academia.
Supply network disruptions are defined as “unplanned and
unanticipated events that disrupt the normal flow of goods and
materials within a supply chain” [1].
The first development refers to current business trends in
the area of supply chain management that increase the
vulnerability of supply networks [2]. Prominent examples are
increases in product variety, lean management, contract
manufacturing, reduction of supplier base, global sourcing and
just-in-time deliveries [2,3,4]. These cause highly globalized
and complex supply networks which, in turn, lead to a high
number of risk exposure points, decreased flexibility, and
reduced time and material buffers [3].
and severity of natural as well as man-made disasters, which is
indicated by a number of independent studies [4]. Supply
network disruptions and their ensuing operational and
financial risks are considered to be the single most pressing
concern for top executives. This was revealed by a survey of
the 1,000 largest European and North American companies in
terms of market value [5].
There are numerous possible causes of supply network
disruptions, such as socio-political factors, natural
catastrophes or terrorism. However, recent academic literature
points out that, regardless of the cause of disruption, the
immediate and long-term financial consequences for an
individual company are severe [3,6]. Immediate consequences
include increases in cost and decreases in profitability and net
sales [3] whereas long-term consequences refer to negative
stock market reactions, decreases in customer service level, or
a damaged brand image [1,3,4,7].

2212-8271 © 2014 The Authors. Published by Elsevier B.V.

Selection and peer-review under responsibility of the International Scientific Committee of “RoMaC 2014” in the person of the Conference Chair Prof. Dr.-Ing.
Katja Windt.
2 M. Brüning, J. H. Buchholz & J. Bendul / Procedia CIRP 00 (2014) 000–000
There are several parallels between the biological self-
healing of human skin wounds and the supply network‟s
recovery of disruptions. For instance, one might draw a link
between the structure of supply networks and cell networks,
i.e. the skin. The parallels lead to the assumption that learning
from nature (i.e. from biological self-healing) can be helpful
in improving the recovery of supply network disruptions, both
in terms of time-to-recovery and costs. As a prerequisite, it is
necessary to prove if the concept of biological self-healing can
actually be transferred to the recovery of supply network
disruptions. The objective is to make statements about this
transferability in order to assess if further research in this field
is sensible. The analysis of this transferability, with principles
derived from biological self-healing as basis of analysis, is
therefore the key contribution of this paper to the research
field of self-healing supply networks.
The next section describes the theoretical background,
followed by the methodology section. Parallels between the
biological self-healing and the recovery of supply network
disruptions are described and characteristics of self-healing
networks are presented (section four). In section five, several
underlying principles of biological self-healing are identified.
These form the basis for the development of five propositions.
The case of Renesas Electronics Corporation (“Renesas”) is
described, analysed and evaluated in section six according to
the developed propositions. Section seven contains a
conclusion, limitations of the research as well as suggestions
for further research.
2. Theoretical Background
The topic supply network disruption is mainly affiliated to
supply chain risk management and supply chain resilience
literature. Within this academic literature, there is a
differentiation between preventive and reactive strategies
when dealing with disruptions [7]. Preventive strategies have
the goal to protect companies and supply networks from
disruptions by taking advance actions to minimize the
occurrence of disruptions. For instance, Kleindorfer and Saad
[8] provide a conceptual framework including risk assessment
and risk mitigation activities. Knemeyer et al. [9] deal with
proactive planning for catastrophic events and Norrman and
Jansson [2] analyse the proactive supply chain risk
management of the company Ericsson. Reactive strategies, on
the other hand, deal with actions after a disruption occurred,
thus during the recovery. The focus is to limit the severity of
disruptions regarding time-to-recovery and required costs [6].
Bello and Bocell [3], for instance, highlight socio factors of
supply partners that are relevant to jointly recover from
disruptions. Reactive strategies are closely related to the
concept of resilience which generally refers to “the capability
and ability of an element to return to a stable state after a
disruption”[10]. As supply networks become increasingly
complex and risky, companies are less likely to completely
eliminate disruptions from occurring [1]. Although reactive
strategies are therefore crucial, they receive less attention in
literature compared to preventive strategies [11].
Although several approaches in the field of reactive
strategies exist, collaborative efforts of involved actors are not
emphasized enough. Most academic literature focuses on
effort and capabilities of individual companies. Collaboration
is also required for successful and efficient healing of skin
wounds [12]. Organisms are able to heal wounds not because
of the robustness of individual cells. It is because of the
interactions between large numbers of cells [13]. Therefore, it
is sensible to take an interdisciplinary approach to compare
biological self-healing with the recovery of supply network
disruptions, an approach which does not yet exist.
Currently, interdisciplinary bio-inspired approaches are
promising especially in the field of technical and engineering
solutions. Generally, there is the expectation that transferring
concepts from nature to areas such as engineering or business
can have advantages. This is because natural concepts are
challenged within their environments and, in this way, being
constantly optimized for their respective setting [14]. Thus,
they are validated trough evolution [13]. In the field of
logistics and supply chain management, only few bio-inspired
approaches exist. Gleich et al. [15] state that one would have
expected more effort in this direction, especially with regard
to the increasing complexity in this topic.
The concept of biological self-healing can be found in bio-
inspired literature mainly with regard to technological topics,
e.g. self-healing materials [16]. However, it has not been
addressed much in supply chain management literature.
Cordes and Hülsmann do not elaborate on the biological
process, but they refer to self-healing of supply networks as
„the process of autonomous recovery after the supply network
faces a targeted attack or random damage‟ [17].
In the field of computer science, the concept of self-healing
is transferred to software-based systems [13,18,19]. The
motivation to apply principles of living systems to computer
science is the fact that biological systems exhibit remarkable
adaptation and robustness when confronted with widely
changing environments. In addition, Ghosh et al. [18] stress
that the self-healing procedure of bringing back a system to
normalcy will decrease maintenance time of software-based
systems and thus saving money. George et al. [13] emphasize
that the capacity of organisms to adapt to changing
environments and to heal organs is due to the interaction
between cells. Transferring this capacity to software may lead
to systems that operate adequately even in the face of
disruptions, i.e. catastrophic failures and large scale software
attacks [13].
3. Methodology
As described above, there is only little academic literature
about the collaborative efforts during the recovery of a supply
network disruption. Especially, an interdisciplinary approach
of combining biological self-healing with the recovery of
supply network disruptions does not yet exist. In this respect,
case study research is especially appropriate for
underdeveloped topics and in the early stages of research on a
topic [20]. Consequently, this qualitative exploratory research
approach was chosen to generate further insights in this topic.
As such, this empirical inquiry of a case study „investigates
a contemporary phenomenon in depth and within its real-
world-context‟ [21]. Yin [22] illustrates that in order to verify
 M. Brüning, J. H. Buchholz & J. Bendul / Procedia CIRP 00 (2014) 000–000
a case‟s fit into the theoretical research topic, the case subject
has to meet a number of selection criteria. In this research, the
criteria include that the case subject has been affected by a
disruption, the existence of a supply network and a successful
recovery. With regard to the design of the case study, a
holistic single-case design with a single-unit of analysis is
chosen for a more detailed analysis. The use of single case
studies is sometimes criticized for having little or no
generalizability. It should be noted that with single case
studies analytical, not statistical, generalizations can be
achieved when the proposed claim is represented soundly and
can resist logical challenge [22]. An advantage of single case
study research is the likelihood of generating novel theory
when dealing with relatively underdeveloped topics [20].
Regarding the data collection, different kinds of sources
have been consulted, analysed and triangulated to achieve a
high consistency of findings. The sources included scientific
journal papers as well as documents such as newspaper and
magazine articles, mostly in Japanese to accommodate the
setting of the case study.
With respect to characteristics and the process of biological
self-healing, five underlying principles are derived from
existing scientific research. Based on these, theoretical
propositions are formulated. These are subsequently
compared to the settings of the case subject and its network.
Starting with theoretical propositions implies the drawback of
limiting the ability to make discoveries apart from the
propositions [22]. However, advantages of using a structured
approach prevail. In order to conduct the data collection as
well as the data analysis in a systematic manner, a case study
protocol is derived from the developed propositions. This
procedure increases the reliability of the research [21].
4. From Biological Self-Healing to Supply Network
Healing
Self-healing, in its biological sense, is used in medical
literature in reference to „self-repair‟, „wound repair‟ or
„wound healing‟. The human skin wound healing is referred to
as self-healing because it is a process with no external help
used or required, i.e. medical treatment [23]. This concept of
wounds repairing themselves can be found in literature on
computer science as well. Ghosh et al. [18] describe self-
healing systems as being able to work without human
intervention and to make adjustments to restore themselves to
normalcy. Thus, all resources and capabilities needed to
successfully heal are readily available within the network.
The biological self-healing discussed in this paper is
limited to the wound healing of human skin for two reasons.
Firstly, because this kind of wound healing is covered well in
literature. Secondly, there are several parallels between wound
healing and the recovery of supply network disruptions.
4.1. Parallels between biological self-healing and recovery of
supply network disruptions
First, one parallel refers to the actors (i.e. entities) of these
two concepts. The entities involved in wound healing are cells
that form the tissue. Similarly, a supply network consists of
3
different connected entities [1]. These entities are comprised
of resources belonging to individual companies that are
available within a supply network. These can be different
kinds of resources, such as physical, financial, human
resources or knowledge.
Second, an injury of the human skin and an event that
disrupts a supply network share the characteristic of being
unpredictable [1].
Third, a skin wound is similar to an externally or internally
disrupted part of a supply network. A wound is the result of
an injury and is defined as „a disruption of normal anatomical
structure and function‟ [23]. Similar to disrupted parts of
supply networks, skin wounds may originate from external or
internal forces [23]. In case of supply networks, internal refers
to all resources within the network.
Fourth, healing is described as a complex and dynamic
process that results in a return to normal anatomic structure,
function, and appearance [23]. The recovery of supply
networks refers to the process of restoration to full
functionality after a disruption [6]. Thus both have the goal of
a complete restoration to original conditions. This goal is also
stressed by the literature of self-healing software systems with
a more general description: A self-healing system should
recover from an unhealthy state and return to the healthy state,
and function as it was prior to a disruption [18].
These parallels lead to the expectation that it can be fruitful
to look closer at the process of biological self-healing in order
to be able to analyse the transferability to the recovery of
supply networks.
4.2. The Process of Biological Self-Healing
Wounds normally proceed through an orderly and timely
reparative process. Orderliness refers to a predefined sequence
of events. Timeliness is relative as it is determined by
different factors, namely by the nature and the degree of the
healing process, the status of the host, and the environment
[23]. According to the classification described by Diegelmann
and Evans [24], biological self-healing consists of four
distinct, but overlapping phases, called hemostasis,
inflammation, proliferation and remodeling. The phases can
be divided into two categories, referred to as short-term action
and long-term repair.
The short-term action, consisting of the hemostasis phase,
describes a temporary mending which occurs immediately
after an injury of the human skin occurred [24]. Within this
phase, a clot plugs the wound [12]. The cells involved in the
hemostasis send chemical signals which initiate the next
phases [24].
The long-term repair includes the inflammation,
proliferation and remodeling. The wound is cleaned out, a new
collagen matrix is deposited and becomes cross-linked and
organized [24]. Important within these phases is the cell
migration and the cell division. A specialized cell has the
ability to move into the wound [24] and the ability to divide
itself into two separate cells [13]. Therefore, cells have the
properties of localization, adaptation, adequate redundancy,
and awareness towards the environment [13]. The complete
restoration defines the end of the long-term repair.
4 M. Brüning, J. H. Buchholz & J. Bendul / Procedia CIRP 00 (2014) 000–000
What becomes evident is that the successful and efficient
healing of a skin wound requires collaborative efforts of many
actors involved [12]. George et al. [13] emphasize the
importance of the interaction within the whole network of
cells during the healing process. In the discourse of self-
healing software systems, the related aspect of the
maintenance of systems health is addressed [18]. In case of
supply network disruptions, it is crucial that other parts of the
network continue working. This further reinforces the
importance of focusing not only on individual entities, but on
the perspective of the whole network.
In order to limit the scope of the analysis, not all
similarities between the concept of biological self-healing of
skin wounds and recovery of supply network disruptions are
taken into account. This research is focused on the
resemblance of the surrounding conditions and prerequisites
of the involved processes to initiate an analysis of
transferability. The self-healing process is based on several
principles which, in turn, lead to theoretical propositions.
5. Developing Propositions
To analyse the transferability of biological self-healing to
supply network disruptions, five propositions are developed.
The propositions and the underlying principles, derived from
the concept of biological self-healing, are shown in Table 1.
Several kinds of cells move during the process of wound
healing [24]. This implies the existence of a surplus of cells.
There are enough cells available for the healing process of an
unpredictable injury [13]. When transferring this to the
recovery of supply networks, the proposition states that in a
supply network, a surplus of resources exists.
P1: In a supply network, a surplus of resources exists.
The second principle relates to the actual movement of
cells. The specific (surplus) cells are able to move to the areas
in need [24]. Transferred to supply networks, the proposition
is that resources have the ability to move within a supply
network, for example between different companies.
P2: Within a supply network, resources have the ability to
move.
The principle of adaptation refers to the capability of cells
to perform different functions depending on changes in the
environment [13]. This leads to the proposition that resources
within a supply network can adapt in case of disruption and
are able to take over tasks in disrupted parts of the supply
network. For instance, employees are able to perform new
tasks in order to temporarily substitute the actual workforce of
a damaged plant.
P3: Within a supply network, resources can adapt (i.e. take
over tasks of disrupted parts).
The next principle emphasizes the maintenance of the
network‟s health. While one specific area is recovering from
an injury, it is important that other body functions continue
working as usual [18]. Thus, the proposition states that if a
part of a supply network is recovering from a disruption, other
parts of the network, i.e. suppliers, customers, competitors and
their respective resources, are left unaltered.
P4: If a part of a supply network is recovering from a
disruption, other parts of the network are left unaltered.
Another crucial principle concerns the chronology of the
self-healing process. There is the differentiation between the
short-term action and the long-term repair. Therefore, it is
proposed that during supply network recovery this distinction
does exist as well.
P5: During a supply network recovery, there is a
distinction between short-term action and long-term repair.
The case of Renesas is analyzed in the following section
according to these five developed propositions.
6. The Renesas Case
The situation of the Japanese company Renesas was
chosen as a case study as it strongly aligns with the selection
criteria stipulated in the methodology chapter. First, Renesas‟
supply network was directly affected by the March 11 triple
disaster in Japan in 2011 when eight of its production plants
in Northeast Japan were severely damaged. Second, as a
worldwide leading supplier of semiconductor products to a
number of industries [25], Renesas is closely linked to its
customers, of which the top ten accounted for approximately
50 percent of the consolidated net sales [26]. Third, after the
triple disaster Renesas was capable of completely recovering
its operations and continue business until today. Despite the
severe damages to its main plant, Renesas was able to recover
its supply network and start test running its operations only
one month after the disaster. This presents an unusually short
time-to-recovery that warrants further investigation.
Renesas is a Japanese manufacturer of semiconductor
products. The Tokyo headquartered company employs
roughly 34,000 people worldwide with revenue of
approximately 785 bn Japanese Yen in 2012 [26]. Among its
main products are microcomputers, memories, large-scale
integrated circuits and logic integrated circuits that are used in
a number of industries. Renesas supplies between 20 to 40
percent of automotive microcontrollers worldwide
[27,28,29,30,31]. These microcontrollers are a critical part of
engine-control units, transmission controls, safety devices,
navigation and communications within a car and are highly
customized [27,32]. Both, the criticality of products as well as
their dominant market share [25], make Renesas a key
supplier for global automotive manufacturers [33]. The
company has nine production sites in Japan and seven others
throughout Asia. The distribution of its products is conducted
through its more than 18 subsidiaries in 16 countries and four
continents worldwide. This extensive geographic coverage
was aided by the fact that Renesas originated out of a merger
between NEC Electronics and Renesas Technology in 2010
[34] and also incorporates former operations of Mitsubishi
and Hitachi [30].
Renesas rose to popularity when the March 11 earthquake,
ensuing tsunami and nuclear meltdown wreaked havoc in
Northeast Japan in 2011 [30]. Renesas major chip factory in
Hitachinaka was directly affected and heavily damaged by the
earthquake [35], resulting in production shortages to major
automotive customers and inadvertently causing global supply
disruptions [27,30]. Seven of the company‟s other plants in
the region were not as badly affected but still ran at
diminished capacity due to fragile or broken supply networks
 M. Brüning, J. H. Buchholz & J. Bendul / Procedia CIRP 00 (2014) 000–000
[32]. Resource requirements in terms of electricity also turned
out to be a critical element, whereas a supply of clean water
and gas could quickly be re-established.
In an effort to recover their supply network as soon as
possible, Renesas aimed to use their fab network as a short-
term action [30] to manufacture products at other plants or
through outsourced manufacturing. Renesas‟ other Japanese
plants also suffered from energy shortages and were not able
to cover missing production volumes. Moreover, only very
few of Renesas‟ products manufactured in Hitachinaka were
standardized to a degree which would allow production to
take place in other, internal or external, facilities [36,37]. This
was further complicated by the fact that severely damaged
infrastructure in and around the production vicinities would
not allow for immediate substitute machinery to be delivered.
As a consequence, even figuring in substitute products from
their fab network, Renesas only reached pre-disaster
production volumes in September 2011, six months after their
main production site was damaged [36].
Nonetheless, the company started their long-term repair
efforts only ten days after the triple disaster when it had
drawn up and was starting to execute a recovery plan [38].
The plan foresaw to recover operations by repairing the clean-
rooms used for production, reinstalling production equipment,
and running test operations to ensure sufficient production
quality. At first this was estimated to be a lengthy process as
only restarting microcontroller plants even under normal
circumstances can be a process of up to a week [25,27,39].
The ensuring energy shortages in Japan even further
complicated any outlook in this respect [39]. Renesas
immediately institutionalized a special task force at their
headquarters to monitor the status of individual plants.
Delegates from this task force were also dispatched to assist
individual plants on-site [28,40]. Even though Renesas‟
customers were not completely certain how to assist their
supplier in recovery efforts [2, 27], the on-site task force was
joined by representatives from Renesas‟ major automotive
customers, members of the Japan Automotive Manufacturers
Association as well as individuals from Renesas‟ suppliers
only days after the disaster [38,41]. Automotive customers
clearly expressed their concern for Renesas, putting aside
doubts they might look for new suppliers. At this point,
employees had already ensured the safety of their families and
were also assisting in the recovery efforts, even though they
were initially supposed to be put on mandatory leave. As a
consequence an average of 2,500 supporters would be
involved in recovery activities on-site per day (roughly 80,000
man-days throughout the whole recovery process) [30,31,38].
Actual recovery work was organized jointly by a steering
committee comprised of representatives from all parties of the
supply network. Working groups were temporarily located on
the plant‟s premises to immediately address and resolve
issues. For this, groups comprised of multiple functional
entities worked together on individual work streams (e.g.
restarting production line #1). Visualization of statuses,
progress and open issues on clearly visible white-boards and
posters helped to facilitate an efficient recovery flow [38].
Consequently, Renesas was able to start production test runs
after only one month [31,38,39,42] and to allow its engineers
5
to already resume work. Limited production was started after
another two months [30] - three months faster than initially
anticipated [41].
6.1. Evaluation of Propositions
Preceding propositions regarding the transferability of
biological self-healing onto supply networks have been laid
out in theoretical detail within the previous chapters (see
Table 1). This section will exemplify their application in the
real-world context of the Renesas case.
(1) In the days following the triple disaster Renesas
employees (especially at the affected plants) could not
perform their usual daily business activities due to the severe
damages to all facilities and were involuntarily idle, creating a
temporary surplus of human resources within the supply
network. This was further enhanced by support staff that was
dispatched from other network partners such as customers,
business associations, and suppliers to render assistance to
Renesas. On average this surplus amounted to approximately
2,500 supporters per day [38,30]. In terms of production
machinery as a resource within the supply network, there was
a clear deficit as numerous machines and other production
equipment was destroyed during the disaster. Contrary to this,
in biological systems there is always an adequate redundancy
available. Self-healing aims at allocating the surplus cells
currently most suited to facilitate tasks to reach the status quo.
(2) The aforementioned surplus of human resources within
the supply network was deliberately moved and redirected by
the overall joint steering committee based on prioritized tasks.
Once a task was completed they would support other teams,
even if they lacked specialized skills, but would then focus on
performing generalist tasks. Once an employee‟s usually
performed tasks had been reinstated, they would resume their
original tasks. The ability to move also holds true in case of
know-how, which was proactively shared between all
involved parties to facilitate the flow of information and ease
of understanding complex problems. Visualization and
distribution of information was even allowed up to the degree
of confidential information to support the goal of recovering
operations as soon as possible. In terms of machinery, there
was no ability to move as the majority of specialized machines
had been destroyed within the disaster. Undestroyed machines
within the supply network were not able to produce the
customized products. Whereas for biological self-healing,
specialized cells have the ability to move and replicate to
mend wounds, within supply networks this capability requires
standardization in terms of machinery and products (e.g. when
outsourcing production) or availability in terms of information
(e.g. when substituting tasks).
(3) Within the allocation of surplus human resources,
individual employees could only adapt to perform new tasks
as long as they did not require specialist knowledge. Based on
the timeliness required in this process lengthy retraining was
not sensible. In terms of machinery, the customized products
and little experience with outsourced manufacturing prevented
an efficient utilization of the fab network. Considering
biological self-healing, individual cells have the ability to
adapt to new functions. In terms of supply network elements
6 M. Brüning, J. H. Buchholz & J. Bendul / Procedia CIRP 00 (2014) 000–000

this might be an aspirational goal in terms of standardization formulation of a recovery plan and institutionalization of the
and availability of information as already indicated. joint steering committee. In essence, Renesas completed steps
that would provide a temporary solution to prevent worsening
Table 1. Overview of developed principles, propositions and results from the of the situation but also laid the foundation for and completed
case study.
a long-term recovery process. This is also reflected in the
Biological Proposition Transfer Examples from Renesas case differentiation between short-term action and long-term repair
principle -ability
in terms of biological self-healing in a similar fashion.
Existence of a In a supply Partly - Surplus of human resources due
surplus of network, a proven to daily tasks being rendered
cells surplus of redundant and redirection of 6.2. Discussion of Results
resources exists resources.
- Deficit in terms of production Summarizing these findings on the basis of the stipulated
machinery resources due to losses
during the disaster. propositions the following picture emerges. First, the
existence of a surplus of resources within a supply network
Cells have the Within a supply Partly - Allocation of human resources only holds true for human resources, but not for the severely
ability to network, proven based on prioritized tasks and
move resources have utilization for generalist topics.
damaged machinery resources and could, therefore, only
the ability to - Deliberately free flow of know- partly be proven. Second, the ability to move could also only
move how to facilitate understanding partly be proven as human resources were only utilized within
and efficiency. a generalized context without requiring specialized
Cells have the Within a supply Partly - Adaption capabilities of human knowledge. The flow of information was merely intended for
capability to network, proven resources limited to generalist efficiency improvement purposes within the given
adapt resources can topics. circumstances. Third, the adaptability of resources was
adapt (i.e. take - Production machinery resources
over tasks of can cover only limited volumes limited to general tasks for human resources and non-
disrupted parts) due customization. customized products in case of machinery resources and,
therefore, only partly holds true. Fourth, the proposition that
Emphasis on If a part of a Not - No formalized alterations of the
the supply network proven supply network are occurring
other parts of a disrupted network are left unaltered could not
maintenance is recovering (relationships stay the same). be proven as ripple effects were clearly causing distress for
of the from a - Ripple effects throughout the other involved network parties. Fifth, in terms of chronology
network‟s disruption, whole supply network (lost sales
of healing the proposition that there is a clear distinction
health other parts of for suppliers, missing inventories
the network are for customers). between short-term actions and long-term repair could be
left unaltered proven, as has become evident by the attempted utilization of
the fab network as well as Renesas‟ recovery plan.
Chronology of During a supply Proven - Short-term action includes
healing: network intended utilization of fab
Differentiation recovery, there network and externals. 7. Conclusion and Further Research
between short- is a distinction - Long-term repair focuses on
term action between short- establishing a recovery plan and
and long-term term actions executing in alignment with The analysis of the single case study of Renesas shows that
repair and long-term supply network partners four of the five illustrated principles identified in human skin
repair wound self-healing can, at least partly, be transferred to the
recovery of a supply network disruption. Transferability of
(4) Even though there were no formalized alterations to the principles is, however, often dependent on the type of
supply network during the recovery, individual network resource (e.g. human resources). Only the principle related to
elements had to be altered and adjusted to enable maintenance the chronology of healing can be entirely transferred. In
of the overall network. In the case of human resources, these contrast to this, the principle concerned with the maintenance
were temporary dispatches from all involved parties. As for of a network‟s health cannot be transferred at all in the case of
machinery, other plants or external contract manufacturers had Renesas. The mixed results of this single-case study,
to take over as much production volumes as possible. nonetheless, warrant further analysis in this research field.
Generally, the disruption in one part of the network caused a Especially with respect to the management of supply network
ripple effect throughout other network elements (e.g. supply disruptions, decisions taken as well as techniques to improve
shortage for customers or lost sales for suppliers). In terms of collaboration (e.g. visualization and sharing of information)
biological self-healing, the maintenance of one part of the many more examples can be drawn from the Renesas case
network does, in most cases, not cause these ripple effects as it that would exceed the scope of this current paper.
is as a rather isolated event. Being able to isolate individual Limitations of the presently discussed contents pertain to a
disruptions to limit the effect they have on the overall supply differentiation of whether a cell temporarily or permanently
network would indeed be an aspirational feature. fulfils a certain function and what implications this entails,
(5) Shifting of production volumes to other plants as well something that is evident in the Renesas case. In this respect,
the utilization of contract manufacturers to take over not much attention was given to what types of cells act at any
production as much as possible was the immediate short-term given time and what characteristics they exhibit. This missing
action taken by the Renesas management, fully aware that all segmentation is in turn causing difficulties in clearly defining
other measures would take considerably more time. the degree of externality of actual supply networks.
Nonetheless, these other measures were started given the
 M. Brüning, J. H. Buchholz & J. Bendul / Procedia CIRP 00 (2014) 000–000
The analysis‟ results regarding the transferability reveal a
potential which makes further research in this field sensible.
For instance, looking into the biological processes and
principles on a more detailed level to derive even more
beneficial insights from biological self-healing could be a
next step. As indicated by the results of this case study, a
higher level of detail may also be relevant regarding the
different types of resources or external network elements and
their functions within a supply network. Other potential topics
in this research field may be to analyse the way in which cells
and resources are communicating, i.e. sending and receiving
signals, and analysing how the principles change with respect
to the importance of individual network elements in relation
to other network elements.
Regarding the research approach, single case study
research has several limitations especially with regard to its
generalizability. Therefore, further research should analyse a
larger quantity of cases, including cross-case studies.
Furthermore, additional information sources, for instance
derived from interviews, should be consulted. In the long-
term, after setting a comprehensive qualitative base,
quantitative data analysis to validate results might be sensible.
Additionally, a long-term goal would be to operationalize the
principles in order to define concrete recommendations
regarding strategies for supply chain risk management.
An important general outcome of the conducted research is
that a recovery from a supply network disruption can be more
successful in terms of time-to-recovery if the actors involved
are focusing on collaborative approaches rather than
individually-focused actions. This is evident both in the case
of Renesas as well as in biological self-healing. Thus, the
complexity of a supply network can amplify the severity of a
disruption, but it can at the same time help to heal the
disruption if managed correctly.
References
[1] Craighead CW, Blackhurst J, Rungtusanatham MJ, Handfield RB. The severity
of supply chain disruptions: design characteristics and mitigation capabilities.
Decision Sciences 2007;38(1):131-156.
[2] Nikkei. Japan cars: Carmakers face long road to recovery. March 28th 2011.
[3] Bello D, Bovell L. Collaboration analysis: Joint resolution of problems in
global supply networks. Information Knowledge Systems Management
2012;11(1/2):77-99.
[4] Tang CS. Robust strategies for mitigating supply chain disruptions. Int J of
Logistics: Research & Applications 2006;9(1): 33-45.
[5] Green M. Survey: Executives rank fire, disruptions top threats. Best's Review
2004;Sept:105.
[6] Simchi-Levi D, Schmidt W, Wei Y. From superstorms to factory fires. Harvard
Business Review 2014;92(1/2):97-101.
[7] Tandler S, Essig M. Conceptual framework of supply chain safety. In: Essig M,
Hülsmann M, Kern E-M, Klein-Schmeink S, editors. Supply chain safety
management - security and robustness in logistics. Berlin, Heidelberg:
Spinger-Verlag; 2013. p. 3-40.
[8] Kleindorfer PR, Saad, GH. Managing disruption risks in supply chains.
Production & Operations Management 2005;14(1):53-68.
[9] Knemeyer AM, Zinn W, Eroglu C. Proactive planning for catastrophic events
in supply chains. J of Operations Management 2009;27(2):141-153.
[10] Bhamra, R., Dani, S., Burnard, K. Resilience: the concept, a literature review
and future directions. International Journal of Production Research
2011;49(18): 5375-5393.
[11] Behdani B, Adhitya A, Lukszo Z, Srinivasan R. How to Handle Disruptions in
Supply Chains – An Integrated Framework and a Review of Literature.
Available at SSRN 2012.
7
[12] Martin P. Wound Healing - Aiming for perfect skin regeneration. Science
1997;276:75-81.
[13] George S, Evans D, Marchette S. A biological programming model for self-
healing. In: Proceedings of the 2003 ACM workshop on Survivable and Self-
Regenerative Systems 2003:72-81.
[14] Passino KM. Biomimicry for optimization, control, and automation. London:
Spinger; 2005.
[15] Gleich A, Pade C, Petschow U, Pissarskoi E. Potentials and trends in
biomimetics. Heidelberg: Springer; 2010.
[16] Nachtigall W. Bionik - Grundlagen und Beispiele für Ingenieure und
Naturwissenschaftler. Heidelberg: Springer-Verlag;2002.
[17] Cordes P, Hülsmann M. Self-healing supply networks - a complex adaptive
systems perspective. In: Essig M, Hülsmann M, Kern E-M, Klein-Schmeink S,
editors. Supply chain safety management - security and robustness in logistics.
Berlin, Heidelberg: Spinger-Verlag; 2013. p. 217-230.
[18] Ghosh D, Sharman R, Raghav Rao H, Upadhyaya S. Self-healing systems -
survey and synthesis. Decision Support Systems 2007;42(4):2164-2185.
[19] Mikic-Rakic M, Mehta N, Medvidovic N. Architectural style requirements for
self-healing systems. Proceedings of the first workshop on Self-healing
systems. ACM, 2002.
[20] Eisenhardt KM. Building theories from case study research. Academy of
management review 1989;14(4):532-550.
[21] Yin RK. Case Study Research: design and methods. 5th ed. USA: Sage
Publications Inc; 2014.
[22] Yin RK. Applications of Case Study Research. 3rd ed. USA: Sage
Publications Inc; 2012.
[23] Lazarus GS, Cooper DM, Knighton DR, Margolis DJ, Percoraro RE,
Rodeheaver G, Robson MC. Definitions and guidelines for assessment of
wounds and evaluation of healing. Wound Repair and Regeneration 1994;
2(3):165-170.
[24] Diegelmann RF, Evans MC. Wound healing: an overview of acute, fibrotic
and delayed healing. Frontiers in Bioscience 2004;9(1):283-289.
[25] Nikkei. Japan cars: Aftershocks to Japan automakers 'to last months'. April
15th 2011. [20] Eisenhardt KM. Building theories from case study
research. Academy of management review 1989;14(4):532-550.
[26] Renesas Electronics website http://am.renesas.com/
[27] Nikkei. Japan cars: The assembly line is rolling again, tenuously, at Honds.
April 19th 2011.
[28] Nikkei Electronics. Kore wa mou naosenai. April 30th 2012
[29] Park YW, Hong P, Roh JJ. Supply chain lessons from the catastrophic natural
disaster in Japan. Business Horizons 2013;56(1):75-85.
[30] The Wall Street Journal. Quake-rocked Renesas feels renewed, more relevant.
Retrieved on February 23rd, 2012. [31] Todo Y, Nakajima K, Matous P. How
do supply chain networks affect the resilience of firms to natural disasters?
Evidence from the Great East Japan Earthquake. Discussion paper; 2013.
[32] The Wall Street Journal. Chip shortage: auto makers‟ biggest threat?.
Retreived on March 28th, 2011.
[33] Nikkei. Japan electricity: Factories close as power is cut. March 14th 2011.
[34] Nikkei. Japan consumer products: NEC Electronics see net loss widen. July
29th 2009.
[35] Greve, H., et al. (2013). Network advantage: how to unlock value from your
alliances and partnerships, John Wiley & Sons
[36] Nikkei Electronics. Runesasu ga shinsai kara manda mono „gaibu seisan
kanou na shourui wo fuyasu‟. June 27th 2011.
[37] George S, Evans D, Marchette S. A biological programming model for self-
healing. In: Proceedings of the 2003 ACM workshop on Survivable and Self-
Regenerative Systems 2003:72-81.
[38] Nikkei Electronics. Seizou furoo, zenbu misemasu. May 14th 2012.
[39] Nikkei. Japan hardware: Japanese chip makers restart. March 25th 2011.
[40] Fujimoto T, Park YW. Balancing supply chain competitiveness and
robustness through „virtual dualsourcing‟: Lessons from the Great East Japan
Earthquake. Int J Production Economics 2013;147:429-436. [41] IEEE
Spectrum. How Japanese chipmaker Renesas recovered from the earthquake.
August 23rd 2011.
[42] Nikkei. Japan components: Supply chains bounce back, but for how long?.
July 1st 2011.
Detailed Response to Reviewers

Reviewers' comments:

Reviewer #1: The authors discuss similarities between self-healing processes and recovery from supply
chain disruptions. A case study is presented to support their argumentations.

The topic is extremely important and relevant. The case study is very interesting.

In my opinion the paper however lacks some important key-points which require major improvements
and clarifications:

1) The paper should be better contextualized w.r.t. the literature about supply chain risk management
and, more importantly, supply chain resilience.
Some additional references could be considered, e.g.,
Sodhi, M., Son, B., Tang, C.: Perspectives on Supply Chain Risk Management. Production and Operations
Management 21(1), pp. 1-13, 2011
Kleindorfer, P., Saad, G.: Managing Disruption Risks in Supply Chains, Production and Operations
Management 14(1), pp. 53-68, 2005
Bhamra, R., Dani, S., Burnard, K.: Resilience: the concept, a literature review and future directions,
International Journal of Production Research, 49, pp. 5375-5393, 2011

Authors’ comments:
 Contextualization of the paper has been slightly amended within the “theoretical background”
section, taking into consideration the areas proposed by the reviewer.

2) The authors should clearly state the paper objectives and contribution (which are unclear in the
current version) in the introduction. The introduction is too long. I suggest to split it into two sections,
respectively, dedicated to the (1) introductory part, objective, etc. and the (2) state of the art/theoretical
background.

Authors’ comments:
As suggested by the reviewer, the introduction section has been divided into an introductory part and
the theoretical background

3) It seems to me that the authors discuss only analogies between self-healing principles and recovery
after supply-chain disruptions. Mechanisms and methods to quantitatively evaluate these analogies
(transferability aspects) are not addressed. Would it be possible?

Authors’ comments:
 The emphasis on quantitative evaluations has been slightly increased in the conclusion section.
However, due to the limited scope of this paper, no concrete approaches how to quantitatively evaluate
the analogies are discussed, as mentioned throughout the paper.

4) What could the practical implications for and applications of the transferability principles to supply
chain (risk) management (and, if possible, of quantitative methods to operationalize them) be? Would
they be limited to "recommendations regarding implications and strategies" (see conclusions)? A further
elaboration on these points would be very beneficial.
Authors’ comments:
 Practical implications have been slightly further elaborated in the conclusion following the reviewer’s
comments, but are still limited due to the limited scope of the paper, no concrete approaches on
operationalization are discussed at this point.

5) The section methodology is limited only to the single case-based research. I would have expected to
find explanations on how the authors have conducted the research and the process of relating self-
healing principles to supply chain recovery.

Authors’ comments:
 This has been discussed within the “Methodology” section, but has been made more explicit now,
illustrating the use of Japanese language resources.

6) In section 3, I do not find any elaboration on the fact that supply chains are typically multi-objective
and multi-decision maker environments. This fact may entail a differentiation of the approach based on
self-healing principlea between supply chains and production networks or supply chains coordinated by
a focal company.

Authors’ comments:
 The multi-objectivity of supply chains is briefly touched upon in the “Conclusion and further research”
section, stating that a higher level of detail may be relevant to distinguish different types of resources or
even external elements of the network and their functions within a supply network. Based on the authors’
knowledge, production networks are usually a part of supply chains and a clear separation in terms of
self-healing principles would exceed the scope of this paper. Also with regard to the “coordination by a
focal company”, assuming the reviewer is referring to the importance of individual network elements and
respective implications for the whole supply network, this aspect has now been elaborated slightly more
in the conclusion.

7) In section 4, it would have been nice to find further elaborations on the implications for the upstrean
supply network.

Authors’ comments:
 Implications for the upstream supply network and related network elements has been briefly touched
upon in the “Conclusions” section, but is clearly limited due to limitations of this paper. The authors are
well aware of the necessity to further distinguish these perspectives.

Again, the topic is very interesting but I found hard to clearly understand the objective, contribution and
potential implications for industrial practices or further research. I encourage the authors to improve the
presentation of these (and previous) aspects as well as to consider in the future also more quantitative
applications.
Reviewer #2:
The paper analyzes the interesting analogy between the self healing of a wound and an analogue process
in a supply chain, based on the case study of the recovery of production by the Renesas Corporation
after the triple desaster in Japan.

The paper is well written, and the logic is absolutely clear, and well integrated into the existing literature
on the subject. At this early stage of this new field of research, the paper does not generate too much
new scientific knowledge, but it is certainly a good start to an innovative way of using bio - processes to
inspire improvements in supply chain management and production logistics. Therefore it is deemed a
valuable contribution to the conference.

It is suggested that the authors consider the following items for a minor revision:
1) As mentioned in the paper, a significant difference betweeen the healing of wounds and the case
study is, that to the best of the authors knowledge there was no redundancy in the equipment needed
for starting up production after the earth quake. In general, microelectronic factories have several
identical pieces of equipment available for each process step, and such equipment is usually from an
equipment manufacturer which has delivered the same models to competitors. To bring in and start up
new equipment as a (temporary) replacement for damaged equipment seems feasible, going by general
experience. It would be good to mention why this was not possible in this case.

Authors’ comments:
 In the case of Renesas, production machinery used in the plant in question was highly customized by
Renesas itself, up to a degree where this was the only plant within Renesas production network that was
able to produce the respective specialized products. Even more so, in this case production machinery was
not available to any competitors (explained also by the strong market leadership of Renesas).
Nonetheless, this aspect has now been explained slightly more within the case study.

2) An earthquake is most critical for the lithography equipment, and it is suspected that the equipment
that was only available at the main production site (as mentioned in the paper) was an extreme UV
wafer stepper. Even if for these most advanced equipment models it might have been difficult to find a
quick replacement, it should have been at least feasible to transport the wafers (instead of the
equipment) for those special processes to another factory, e.g. one of the wafer foundries in Taiwan (e.g.
TSMC) or Dresden (Global Foundries), and then back again. To transport wafers to a different place for
selected processes to be carried out is not uncommon in the industry.

Authors’ comments:
 While the authors acknowledge the fact that transporting semi-finished goods to other production
networks is a common practice within the industry in question, in the case of Renesas the damaged
infrastructure around the actual production site as well as restrictions of the scope of their production/
fab network (e.g. limitation of non-Japanese partners) seems to have been a restricting element for this
strategy.

3) It is general experience, that an earth quake will affect the Ultra Clean Water and Chemical/Gas
Distribution network in a factory quite badly, this potential shortage of material needed for many
production processes is not mentioned in the paper, so it should be stated whether this potential
infrastructure problem was not present, or whether there was at least no information about it.
Authors’ comments:
 The authors acknowledge the fact that this might be a common issue. In case of Japan, however,
water and gas supplies which are usually transported through underground networks are controlled to a
higher degree where supplies are temporarily suspended once earthquake are registered by sensors
throughout the country, preventing severe infrastructure damages. In contrast to this, electricity grids
and lines which are usually run above ground in Japan are not as prone to falling victim to earthquakes as
poles and cables can be damaged through extensive swinging. This fact has now been amended and
explained in brief within the case study.

4) The concept of "containment action" as the first step for a disruption (corresponding to the blood
clogging and thus sealing an open wound) is also a key requirement in Quality Management in
connection with process problems, in particular those problems which will directly affect customers. It
might be a good idea, to point out this fact.

Authors’ comments:
 The authors appreciate the reviewers comment and acknowledge its general validity. Nonetheless,
due to the limited scope of this research paper and currently missing links into the direction of Quality
Management processes, the authors would like to reserve this reference for future research.