Resources, Conservation and Recycling 101 (2015) 2033

Contents lists available at ScienceDirect

Resources, Conservation and Recycling

journal homepage: www.elsevier.com/locate/resconrec

System dynamics models for decision making in product multiple

lifecycles
Farazee M.A. Asif a, , Amir Rashid a , Carmine Bianchi b , Cornel M. Nicolescu a
a
b

Department of Production Engineering, KTH Royal Institute of Technology, Brinellvgen 68, 10044 Stockholm, Sweden
Department of Political Sciences, University of Palermo, Via Bandiera, 29-90100 Palermo, Italy

a r t i c l e

i n f o

Article history:
Received 22 January 2015
Received in revised form 25 April 2015
Accepted 5 May 2015
Available online 28 May 2015
Keywords:
System dynamic
Resource scarcity
Resources policy
Material criticality
Multiple-lifecycle
Resource conservation

a b s t r a c t
The main drivers for adopting product multiple lifecycles are to gain ecological and economic advantages.
However, in most of the cases it is not straight forward to estimate the potential ecological and economic
gain that may result from adopting product multiple lifecycles. Even though many researchers have
concluded that product multiple lifecycles result in gain, there are examples which indicate that the gain
is often marginal or even none in many cases. The purpose of this research is to develop system dynamics
models that can assist decision makers in assessing and analysing the potential gain of product multiple
lifecycles considering the dynamics of material scarcity. The foundation of the research presented in this
paper is laid based on literature review. System dynamics principles have been used for modelling and
simulations have been done on Stella iThink platform. The data used in the models have been extracted
from different reports published by World Steel Association and U.S. Geological Survey. Some of the data
have been assumed based on expert estimation. The data on iron ore reserves, iron and steel productions
and consumptions have been used in the models. This research presents the rst system dynamics model
for decision making in product multiple lifecycles which takes into consideration the dynamics of material
scarcity. Physical unavailability and price of material are the two main factors that would drive product
multiple lifecycles approach and more sustainable decisions can be made if it is done by taking holistic
system approach over longer time horizon. For an enterprise it is perhaps not attractive to conserve a
particular type of material through product multiple lifecycles approach which is naturally abundant but
extremely important if the material becomes critical. An enterprise could through engineering, proper
business model and marketing may increase the share of multiple lifecycle products which eventually
would help the enterprise to reduce its dependency on critical materials.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Economic gain is one of the main drivers for Original Equipment Manufacturers (OEMs) in deciding whether to adopt product
multiple lifecycles (PML)1 approach or not. Ecologic gain could

Corresponding author. Tel.: +46 8 7909076/+46 734612486; fax: +46 8 210851.

E-mail addresses: [email protected] (F.M.A. Asif), [email protected] (A. Rashid),
[email protected] (C. Bianchi), [email protected] (C.M. Nicolescu).
1
Product multiple lifecycle is a novel approach to conserve recourses proposed
by the research framework called Resources Conservative Manufacturing (ResCoM)
initiated at the department of production engineering, at the KTH Royal Institute
of Technology, Sweden (Asif, 2011; Asif et al., 2012; Rashid et al., 2012, 2013).
Product lifecycle is also used to refer to the stages of product development that
includes introduction, growth, maturity, and decline of a product in the market. In
this paper, multiple lifecycles means a product that has several lives by converting used products and components into new products through remanufacturing or
reuse. For standard denition of remanufacturing and reuse see (Hauser and Lund,
2010).
http://dx.doi.org/10.1016/j.resconrec.2015.05.002
0921-3449/ 2015 Elsevier B.V. All rights reserved.

equally be interesting as long as it would not create any conict with economic interest and if that would ensure long term
strategic advantages. In this respect, most of the research efforts
that underlined potential economic and ecological gain resulted
from PML approach mainly emphasized on energy consumption
and emission aspects. For example, Steinhilper (1998) estimated
that remanufactured automotive alternators and starters generally
save 88% and 89% material and 86% and 91% energy, respectively,
compared to new manufacturing. In a similar analysis Sutherland
et al. (2008) have concluded that remanufacturing of automotive
engines can save up to 90% of energy. Hauser and Lund (2012)
reported that in general remanufacturing can save up to 80% of
energy and material. Goldey et al. (2010) concluded that remanufacturing of telecommunication equipment reduces material and
energy consumption which also results in lower greenhouse
gas emissions. In a study Gutowski et al. (2011) argued that
energy saving through remanufacturing is not always obvious and
positive.

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

Nevertheless, studies in which material saving aspects have

been highlighted, it is done purely from the perspective of absolute
ratio of material (i.e. kg/product) saved or reused without considering criticality2 of that particular material. The examples are
often too specic to a product or a case that it cannot be taken
for granted that PML approach would have same impact on similar or other kinds of product. In addition to these, adopting PML
approach would require OEMs to initiate substantial reform in their
strategic and operational areas which means OEMs need to base
their decision on solid ground rather than few abstract examples.
Of course, these kinds of analysis can assist decision making but at
the same time raises questions such as; is it enough just to know
PML approach saves material? Should importance of saving the
Rare Earth material be considered same as saving Diatomite?
In these respects, it is to be understood that the gain is not only
a matter of absolute quantity of material saved but also results of
physical availability, economic importance and supply risk of that
particular material. It means, more facts are needed to make a better
decision and facts should be presented in a form that the decision
makers can assess the inuences of alternative scenarios. But there
is no such generic decision support tool available today to full
these purposes.
The main objective of this research is to develop system
dynamics models that can assist decision makers in assessing and
analysing the potential gain3 of PML considering the dynamics of
material scarcity. The models should also allow creating different
scenarios to make a better decision. To full this objective a twostep analysis mechanism has been developed by incorporating a
two-level model,

21
Acon

Inform
maon
abo
out
problem

Acon

Resu
ult

Informao
on
about
problem

Result

Fig. 1. (a) Open loop impression of the world, (b) closed-loop Structure of the World
(Forrester, 2009).

Moreover, material recycling aspect to supplement the primary

sources of material has not been included in the models for the
purpose of simplicity. The authors have the plan to integrate these
issues later on in a future publication. The model in this paper can be
considered as the core and on top of it dynamics of energy, emission
and solid waste could be added.
As in most cases material price directly inuences the price of
product, for the reason of simplicity in modelling price of product
has been used to represent material price as well.
In this work iron ore, iron and steel have been referred as materials and the models are generally relevant for industries that are
heavily dependent on these materials and currently not considering
alternative materials as the replacement.
Despite the fact that the models presented in this paper take
into consideration a wide range of parameters, interconnections
and feedback loops, they only represent few of the phenomena
related to material scarcity and enterprise dynamics. In addition,
even though the models could be used to simulate many different
scenarios, only two scenarios have been presented and discussed
in Section 7.4.
4. Research method

the rst step is to observe the global trend of material scarcity i.e.
how the supply of a particular material would uctuate and how
the material reserves would eventually run out over time
the second step is to feed the global trend of material scarcity into
the enterprise level model to see what inuences these would
have on an enterprise that largely depends on the material supply
and from that, understand the potentials of adopting PML.
2. Problem denition
The problem of assessing and analysing the potential gain of
PML has been tackled in the research presented in this paper. The
origin of the problem lies in,
Lack of generic method and model that takes dynamics of
resource scarcity into consideration
Lack of simple and interactive analysis tool.
3. Scope, delimitations and limitations
In this research it is considered that the decision whether PML
is more favourable than single lifecycle approach is made based on
the analysis around material production, supply, consumption and
the inuence of material scarcity. Other vital issues such as emission, solid waste and energy have not been included as this would
be impossible to describe all these phenomena in one publication.

2
The term critical material/criticality of material is used in the same context
as in the Report of the Ad-hoc Working Group on dening critical raw materials
published by European Commission (2010).
3
The term potential gain is used to dene an estimated value (caused due to
scarcity of material and price hike due to the scarcity) which is greater than the actual
value (market price) of material. For example, saving one kilogram of material when
its physical availability is estimated to be high (or economic importance is low), is
strategically less valuable than saving the same amount when the material become
scarcer (or economic importance become higher).

The research presented in this work is a qualitative research

based on literature review. Publicly available research publications
and reports have been used to create a research foundation, identify
the gaps in research and develop new theories to ll the gaps. The
review type that has been used resemble to narrative synthesis as
summarized by Denyer and Traneld (2006). To access literatures
Google and Google Scholar search engines have been used. Initially two streams of research have been dened as the scope for
literature review which included, (1) decision making tool (both
mathematical and simulation based models) developed for PML
and (2) resource scarcity. As the literature review progressed the
scope of the literature search has been further narrowed down to
system dynamics models and material scarcity. Along the search
process the following keywords have been used both in single
and combined form: system dynamics, resource scarcity, critical
raw material, PML, decision making tools, remanufacturing, data
on steel and resource conservation. Often references of certain
selected literatures have been examined to nd other relevant literatures and to do so, supporting tool (that connects related articles)
provided by Google scholar and particular journals have been used.
Guidelines of Rahmandad and Sterman (2012) have been followed
to document the models and report the results.
5. System dynamics basics
Before going in to further details, it is important to enlighten
the readers about the basic principles of system dynamics as this
is the main method used to create the models. System dynamics is
a method to study complex systems. When the concern is to make
decision, people tend to think linear i.e. we see a problem, decide
on an action and expect that our action will resolve the problem as
shown in Fig. 1(a).
However, as shown in Fig. 1(b), in reality a problem leads to
action and action produces a result that may create further problem or alter the nature of the problem. As soon as the nature of

22

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

6. Review of relevant research

Fig. 2. Illustration of feedback concept (Forrester, 2009).

the problem changes or a new problem appears, the initial action

does not remain relevant anymore which forces the action to be
changed. This is caused due to systems internal feedback mechanism which can either be positive (also known as reinforcing)
or negative (also known as balancing) in nature. The action, the
result and the decision making is signicantly inuenced by the
nature of these feedback loops. Moreover, every decision making
involves three major elements i.e. (1) result that is to be achieved,
(2) action to get to the result and (3) information that connects
the action and the result. This can be illustrated by a simple example
of lling a glass of water as shown in Fig. 2. The result that is to be
achieved in this example is to have certain amount of water in the
glass. The action to open the water tap is based on the information
that is received through eyes, i.e. the glass is empty. Throughout
the pouring process the information regarding the level of water is
feedback which eventually helps to determine the action i.e. when
to reduce the water ow and nally close the tap.
In system dynamics the above mentioned principles are applied
in the form of articulated problem, for instance, maintaining
the level of water in the glass in this case. The problem is then
reformulated using mental models generally known as Causal Loop
Diagram (CLD) as shown Fig. 3(a) to identify the systems (in this
case, water pouring process) internal feedback. This mental model
is mathematically expressed considering the interaction and feedback among different variables of the model elements which is then
converted to computer simulations for further elaboration and
visualization as shown in Fig. 3(b). The computer simulation models
are represented with three elements, (a) stock, (b) ow and (c) variable that connects all three model elements. Each of the elements
has distinct characteristics which can be expressed mathematically.
The readers are encouraged to go through the books of Sterman
(2000) and Morecroft (2007) for further details about the system
dynamics principles.

As described in the research method, the scope of literature

review has been narrowed down to the research work related to
decision making models in PML based on simulation. The scope of
simulation models has been further narrowed down to only models
that used system dynamics.
Estimating, predicting and modelling of the natural resources
reserve and depletion are not a new topic. A tremendous amount
of work has been done in this area and therefore, mankind is well
informed about reserve of natural resources and their consumption
and production pattern. However, there is very little researcher
work accessible that used system dynamics principles to model
the dynamics of natural resources scarcity and there is no publication apparently available that coupled system dynamics model
of natural resources scarcity in decision making in PML. The most
recent and relevant research in this topic has been published by
Gleiche et al. (2011) which has not been presented from the perspective of decision making in PML, rather to capture the dynamics
of price uctuation that would occur from the gap between supply and demand of non-renewable (material) resources. However,
this price uctuation due to gap in supply and demand can be considered as one of the important phenomena that decision making
would require. The gap in supply and demand can be a consequence
of many factors and resource scarcity is one of the factors that have
been quite widely discussed. In a similar type of study Boudreau
et al. (2009) modelled the dynamics of platinum supplies to understand the challenges that the fuel cell vehicle would face in future.
Van Vuuren et al. (1999) presented a system dynamics model and
simulated long term trends in production and consumption of
iron/steel in relation to impacts such as ore-grade decline, capital
and energy requirements and ows of waste. Laurenti et al. (2015)
through a conceptual model touched upon the issue that reuse,
repurpose, remanufacture and recycle (relates to PML approach to
some extent) would cause raw material extraction and production to decrease. However, their main focus has been to analyse
the effect of incremental improvements in material and energy
efciency on consumption, pollution, negative externalities and
economic equalities.
There are some system dynamics model based work done in the
area of reverse logistics such as Georgiadis and Vlachos (2004) studied the behaviour of reverse supply chain with product recovery,
Rodrigues et al. (2012) among other things modelled reverse supply chain to see the long-term behaviour of reverse supply chain,
Vlachos et al. (2007) examined capacity planning policies of forward and reverse supply chain and Georgiadis et al. (2006) and
Georgiadis and Efstratios (2010) studied behaviour of closed loop
supply chain with remanufacturing. Even though the higher level
objective of the above mentioned contributions can be connected to
resource conservation (addressed by PML approach), their models
do not address the issue of material scarcity.
It is worth mentioning that the rst system dynamics model that
captured the dynamics of material scarcity is The World Model

+ Water level
Flow of water

Water level

Flow of water

Dierence in
water level

Desired water level

+
Difference in water level
Fig. 3. Examples of (a) causal loop diagram b) stock and ow diagram.

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

23

Fig. 4. Causal relationship between material production, material reserves and economic interest of material production.
Fig. 5 Causal relationship between material delivery, supply, delivery delay and deliverability.

presented by the Club of Rome (Meadows et al., 1972) in early 70s.

The World Model has been used to demonstrate the limits that
the prosperity of mankind would face in future. According to the
report, the prosperity of mankind is directly linked with the size of
population, the scarcity of material resources and arable land is the
consequence of the human activities. The authors have discussed
different alternative approaches to handle the growth to avoid or
minimize catastrophe that could otherwise occur due to population
overshoot. One of the important conclusions drawn from the world
model regarding material scarcity is, no matter which policies we
may adopt the material resources will run out as the physical size of
the world is nite. The work of Meadows et al. (1992) and Acharya
and Saeed (1996) are few of the improved and supplementary versions of the world model which were developed later. As a well
celebrated and cited (almost 13,000 times) work the outcomes of
the world model have been validated through regular updates i.e.
the 20-year (Meadows et al., 1992), 30-year (Meadows et al., 2004)
and 40-year (Turner, 2012). Despite critique from works such as
Alcamo et al. (1998) arguing that due to high level of aggregation
and lack of details the world model never made into the main scientic stream, these updates provide enough evidence of the accuracy
of the world model estimations. It also proves that high level of
aggregation is neither an obstacle to a good scientic work nor
a reason behind inaccurate/unreliable results. Hence, these provide enough condence to foresee the potential of applying system
dynamics in decision making in the PML approach in this work.
7. The modelling approach and the models
The ve steps modelling process describe by Sterman (2000)
has been adopted to create the models. The models describe the
problem of material distribution/supply worldwide that would
occur from scarcity and the challenges that any enterprise would
face in general from this scarcity. Dynamics of material scarcity4 and
enterprise dynamics under the inuence of material scarcity are the
two sectors in the simulation model interacting with each other
to demonstrate the effect of global trend on an enterprise. The one
side of the boundary of the model starts with the material reserves
(i.e. the part of the material resource which has been geologically
evaluated and is commercially and legally mineable (European
Commission, 2010) and the boundary ends with an enterprise
shipping product to its customers. The inuence that technological
developments would have on converting the material resources
to material reserves or increase the efciency of the material
processing rate is outside the scope of the model (impact of these

4
The words written in Italic font from this point forward are the terms used in
the actual causal loop diagram and stock and ow model and the denition of some
of these terms has been included in Appendix A.

have been studied for example, by Wils (1998)). Furthermore,

recycling is also an important activity in the material supply chain
and it has its own dynamics with signicant effects in the supply
of materials. Even if, for the sake of the simplicity, models in this
paper have been developed to analyse only the dynamics of the
primary sources of material and it would be very useful to study
the combined dynamics of the primary and secondary source of
material.

7.1. Dynamic hypotheses and assumptions

The economics of non-renewable resources are a complex issue
which one can envisage by only reviewing the rules proposed by
Hotelling (1931). However, the scope of this research is not to
discuss the detail of the theory of economics of non-renewable
resources rather to incorporate generally accepted theories and
practices in this and relevant areas to develop the hypotheses used
in modelling. These hypotheses are further elaborated in this section and in Section 7.2 with relevant references. Following the
xed stock paradigm dened by Tilton (1996), Hottelings rules
of economics of non-renewable resources implies that user cost
will increase if the resources become scarcer (Barbiroli and Focacci,
2009) and this leads to the balancing feedback loop B1 in the model
which is shown in Fig. 4.
As evident, the non-renewable resources are not evenly distributed over the world and it is only few countries in the world that
produce and supply the major portion of the material resources
to the rest of the world (European Commission, 2012). As a consequence, it is not only the physical reserves of a certain material but
also the geopolitical and economic situation that plays an important role in meeting the global demand of certain material. As the
material gets scarcer and becomes economically or technologically
important, the producing countries tend to adopt different strategies such as restricting export to secure their exclusive access
to raw materials (European Commission, 2008, 2010; Bunker,
1994; Kim, 2010; Korinek and Kim, 2010). European Commission
(2008) has reported that there are over 450 export restrictions on
more than 400 different raw materials, which mainly came from
China, Russia, Ukraine, Argentina, South Africa and India. This
phenomenon is captured in the reinforcing feedback loop R1 as
shown in Fig. 5. It means that if material reserves go down, backlog
in material supply will increase which is an indication of possible
scarcity. As the sign of scarcity becomes evident the material
producing countries may restrict export of material resources.
The result of these two above mentioned feedback loops (B1
and R1) would cause the supply of certain material to drop. This
will cause the price of material to escalate and elevate the concern
to look for alternative materials (or alternative approach to replace
the material) both at the local and the global level. As the alter-

24

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

+
+ Material delivery
rate

Material inventory
world wide
B1

Material
deliverability
-

+
Consumable
material inventory
worldwide

R1

Material
deliverability rao Delay in
material supply
+
+
Percepon of
rarity

+
Material
supply rate

Fig. 6. Causal relationship between material deliverability and demand.

native becomes available the demand of that particular material

is expected to stabilize (after an initial increase) over time. These
dynamic relationships are in-line with the belief of the opportunity cost paradigm as discussed by Tilton (1996). Moreover, there
are examples in industrial sectors of switching from one sort of raw
material to another due to unavailability or quest for cheaper alternatives. One such example is capacitor which was mainly Tantalum
based until 1990s and switched to Niobium and Ceramic based
capacitor (Korinek and Kim, 2010) due to the lack of access to Tantalum. These cause and affect relationships have been captured in
the feedback loop B2 and R2 in the models as shown in Figs. 6 and 7.
In addition, the feedback loops B2 and R2 together with R1
would affect any enterprise that uses the particular material in
concern in many ways. However, the two main effects would be
failure to meet the demand of products due to lack of material supply or drop in demand due to increase in price of product which is
linked with the increase in price of material (as mentioned in Section 3, product price has been also used to represent material price).
The possible inuences of these three feedback loops on the supply chain dynamics of an enterprise are illustrated by the balancing
loop B3 as shown in Fig. 8.

+
Demand of B2
material

Material
reserves

B1

+
+

+
+

Material
deliverability
-

Gap in
Raw material
Material inventory
B3 manufacturing
+ inventory
world wide
+
+
B1
Material delivery
Desired material +
rate
R1
purchase rate

Manufacturing
rate

To deal with the combined effect of the reinforcing loops R1,

R2 and balancing loop B2 discussed above, an enterprise can
take different policy measures. The policy measures should primarily focus on reducing dependency on material resources as
suggested by Meadows et al. (1972, 1992, 2004) and Jackson (2009).
One of the policy measures could be to adopt the approach of
Resource Conservative Manufacturing (ResCoM). ResCoM is a proposed framework that suggests PML as an approach to conserve
resources to reduce dependency on material resources (Asif, 2011;
Asif et al., 2012; Rashid et al., 2012, 2013). However, it is obvious
that as long as the material in concern does not become economically or strategically critical, OEMs may not prioritize material
conservation through PML approach. As discussed in the previous section, in the context of the models presented in this paper,
the criticality of the material is a function of the worldwide material reserves and the geopolitical situation in a given time. As a
material gets scarcer and/or price of the material increases beyond
perception it becomes strategically important for any enterprise to
take action to conserve material. Depending on the market size,
level and quantity of material to be recovered and the time when
action to conserve material would be taken can decouple any enterprises dependency on material to some extent. These relationships
have been captured in the balancing feedback loop B4 as shown in
Fig. 9.
As already presented above, there are six feedback loops apparent in the models and the complete overview of their interactions
with each other are shown in Fig. 10. To summarize the discussion
above, the balancing loops B1, B2 and reinforcing loop R1 refer to
the dynamics of material scarcity representing the global model.
The balancing loop B3 refers to the supply chain dynamics of an
enterprise which is connected by the reinforcing loop R2 to the
global model. The reinforcing loop R2 refers to the dynamics of an

Percepon of
rarity
+

Material
inventory world
+
wide

Product
demand

Price of
product

Material
deliverability rao

R2

7.2. Policy alternatives

Demand of
material

+
+ Manufacturing
rate

Fig. 8. Causal relationship between raw material inventory, manufacturing and

material purchase rate (supply chain dynamics of an enterprise).

Material supply backlog

+

B2

Price of
product

Product
demand

Gap in
+ Raw material
B3 manufacturing
inventory
+
+
Desired material
purchase rate
R2

Material delivery
rate

+
Consumable
material inventory
worldwide

R1
+
Material
supply rate

Delay in
material supply
+
Material supply
backlog

Fig. 7. Causal relationship between material deliverability, price of product and

manufacturing.

R2

Material
reserves

B3
Price of
product

R1

+
Material
price stress Material
supply
index
stress index

Material
processing rate
+

+
Time to exhaust
material reserve +

B4

B1

+
Rao of material
reserve

Economic interest +
of processing
material

Perceived
material
recovery rate
+

Potenal
+
gain of PML
+
Fig. 9. Causal relationship between material reserves, price of product, and potential
gain of PML.

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

25

+
Potenal
gain of PML

+
Perceived
material
recovery rate

Material
supply
stress index

B4

Material
price stress
index
+
-

+
+

B1

Price of
+ product

- Material
reserves

Material
processing rate
+

Time to exhaust
material reserve

+
Rao of
material
reserve

Material
inventory
world wide

+
Economic interest of
processing material

Percepon of
rarity

Product
demand

Gap in
Raw material
B3 manufacturing
+ inventory
+
+
Desired material
purchase rate
R2

+
+
Material
deliverability rao Material delivery
+
rate

B2

+ Manufacturing
rate

Material
deliverability
+
Demand of
material

+
Consumable
material inventory
worldwide

R1
+
Material
supply rate

Delay in
material supply
+
Material supply
backlog
+

Fig. 10. An overview of the complete causal loop diagram with six feedback loops (the enterprise level model is illustrated in grey and global model in black).

enterprise under the inuence of material scarcity. Finally, the balancing loop B4 refers to the policy alternative that can be adopted
to combat material scarcity.
7.3. Stock and ow diagram
All the above mentioned cause and effect relationships have
been converted into the stock and ow diagram as shown in
Figs. 11 and 12.

As mentioned earlier most of the data used to dene the

key variables and to run the models are extracted from relevant
databases with few assumed data. Data on steel and iron production and consumption, together with iron ore productions,
consumptions and reserves have been used to run the models.
The sources of data are mentioned in Table 1 below including the
key variables with indication if the variables are endogenous or
exogenous. In addition, the equations of the models are listed in
Appendix B.

Fig. 11. Stock and ow diagram of dynamics of material scarcity.

26

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

Fig. 12. Stock and ow diagram of enterprise dynamics under the inuence of material scarcity.

7.4. Simulation results

The results presented in this paper are not specic to any real
case as the idea of the modelling is to develop a generic (i.e. ts for
all materials and all kind of manufacturing industries) conceptual
model. Besides, the results of the simulations can be read in variety
of ways. A typical way of reading the results of system dynamics
models is to examine the behaviour of different key variables in
reference condition and compare them with possible scenarios. A

reference condition is mainly determined based on real facts and

gures, which is often referred as current state of a system. The
reference condition in the context of this paper is when the model
is run with the data provided in Table 1. Once the reference condition is set, the users of the model can change the parameters of
the key variables to generate scenarios and observe the behaviour
of the system as whole and the behaviour of individual entity. Key
variables in this context are the ones that inuence the system
behaviour, problem and decision making most. Perceived price of

Table 1
List of key variables, sources of data, initial values of the key variables and their units of measure.
Key variables
Endogenous

Material reserves
Material inventory worldwide
Consumable material inventory worldwide
Material demand
Current material processing capacity worldwide
Price of product
Product demand
Raw material inventory

Exogenous

Material demand growth factor

Material processing growth factor
Ratio of material consumption nationwide
Growth factor of nationwide material consumption
Material processing capacity growth factor
Desired material reserve
Desired material deliverability
Ratio of material consumption at enterprise level
Quantity of material per product
Perceived price of product
Manufacturing growth factor
Estimated market size of PML
Material recovery ratio
Weight average
Time to change the capacity, Time to process, Time to delivery,
Time to consume, Time to manufacture, Time to update
product demand
Delay in recognizing drop in reserve
Time to update material demand
Time to update the price
Desired delay in supply

Sources of data used for the

initial value and input

Initial values

Units of the variables

Jorgenson (2012), World Steel

Association (2011)
World Steel Association (2012)

121,429

Million tonne

1186
1186
1205
1205
80
5000
5000

Million tonne
Million tonne
Million tonne/year
Million tonne/year
D/widget
Widget/year
Widget (equivalent material)

0.05
0.06
0.0033
0.02
0.01
56,667
1
0.02
1
400
0.1
0
0.8
0.5
1

Unit-less
Unit-less
Unit-less
Unit-less
Unit-less
Million tonne
Unit-less
Unit-less
kg/widget
D/widget
Unit less
Unit-less
Unit-less
Unit-less
Year

20
3
5
4

Year

Data is assumed

World Steel Association (2012)

Data is assumed

Year
Year

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

Material price stress index (Unit-less)

27

Material supply stress index (Unit-less)

Potenal gain of PML (Unit-less)

1,5
1,45
1,4
1,35
1,3
1,25
1,2
1,15
1,1
1,05
1
1

9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101

Year
Fig. 13. The behaviour of potential gain of PML with respect to material supply stress index and material price stress index.

product, for instance, is a key variable in the presented models and

its parameter has been changed to create a new scenario which is
discussed in details later in this section.
Nevertheless, referring to the main objective as mentioned in
the beginning of this paper, potential gain from PML approach is
the main result that this research and the modelling has aimed for.
Therefore, it is appropriate to start presenting the results by introducing the behaviour of potential gain of PML. It is important to
highlight that the potential gain of PML approach is a consequence
of material scarcity and the inuence of scarcity on the price of
material. In that respect, potential gain of PML is best read in relation to material supply stress index and material price stress index,
and their behaviours are shown in Fig. 13. As apparent in the graphs,
material price stress index is expected to increase sharply within a
short period of time where growth of material supply stress index
is spread overtime. These advocate that material price increase
will affect manufacturing industries more rapidly than physical
unavailability of material. Manufacturing industries that only focus
on cost optimization strategies will face serious issues compared
to industries that may undertake material saving strategy through

PML approach. It is also important to look at the time scale to know

at which point in time it becomes important to start resource conservation to gain highest potential. This time scale could also be
used as an indication to plan when alternative strategies to battle scarcity would become absolutely necessary. In addition, any
enterprise can easily estimate total potential of adopting PML at a
given time if estimated market size of PML and material recovery ratio
is known. The sharp fall and rise (oscillation) before material supply
stress index becomes stabilized again (as shown in Fig. 13) is due
to the fact that at that point of time consumable material inventory
worldwide falls sharply. This causes material demand to stabilize
which allows the consumable material inventory worldwide to grow
again before it is completely consumed.
The other important outcome of the models is to examine how
resource scarcity could affect an enterprise over time. This can
be read through the dynamic behaviours of manufacturing rate in
comparison to product demand. As described in Section 7.1 and
illustrated in the causal loop diagram B3 in Fig. 8, manufacturing
rate of any enterprise is constrained by material availability and
material price hike. The behaviour of manufacturing rate in respect

Product demand (Widget/year)

Manufacturing rate (Widget/year)

700000
600000
500000
400000
300000
200000
100000

97

101

93

85

89

81

73

77

69

65

57

61

53

49

45

41

37

29

33

25

21

17

13

Year
Fig. 14. The behaviour of manufacturing rate in comparison with product demand in the reference condition.

28

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

Product demand (Widget/year)

(a)

Manufacturing rate (Widget/year)

140000000
120000000
100000000
80000000
60000000
40000000
20000000

97

101

93

89

81
85

73

77

61

65
69

53

57

45
49

37

41

33

21

25
29

13

17

5
9

Year
The behaviour of manufacturing rate in comparison with product demand when
constraint of price hike is removed (zoomed out Y-axis)

Product demand (Widget/year)

(b)

Manufacturing rate (Widget/year)

1200000

1000000

800000

600000

400000

200000

101

97

93

89

81
85

73

77

61

65
69

57

53

45
49

41

37

33

21

25
29

17

13

5
9

Year
The behaviour of manufacturing rate in comparison with product demand when
constraint of price hike is removed (zoomed in Y-axis)
Fig. 15. (a) The behaviour of manufacturing rate in comparison with product demand when constraint of price hike is removed (zoomed out Y-axis). (b) The behaviour of
manufacturing rate in comparison with product demand when constraint of price hike is removed (zoomed in Y-axis).

to product demand in the reference condition is shown in Fig. 14.

The graph illustrates that the growth of any enterprise will be
threatened by either the increase in price of material i.e. price of
product goes beyond perceived price of product or the unavailability of material i.e. material deliverability declines, whichever comes
rst. In the reference condition of the models, it is the price increase
that is causing the growth of the enterprise to stop and eventually
decline. It also means that when the price of the product exceeds
the perceived value, the consumers will not demand the product
any longer causing the demand to stabilize over time.
If the constraint of price hike is removed from the models i.e.
increase in price does not inuence the manufacturing cost of
any enterprise signicantly or the price of material will always be
affordable, in that case the only active constraint to the growth is
material availability. Under this constraint the behaviour of manufacturing rate and product demand follows the pattern shown in

Fig. 15(a) and (b) which illustrate the same graph but Fig. 15(b) is
shown with zoomed in Y-axis. As it appears, the growth sustains
better in this case in comparison to growth curves shown in Fig. 14.
However, it should be remembered that in reality it is not possible
to completely avoid the constraint of price increase as it is directly
inuenced by material scarcity but could be managed well if PML
approach is undertaken as discussed in Section 7.2.
It is also important to discuss the off bound phenomenon of
the variable product demand apparent in Fig. 15(a) and (b). In
the models it is only the price of product that controls the product
demand and when it is removed, product demand continues to grow
innitely. In reality, it means that the consumers (who are creating
the demand) will want a certain product even if the price of product
is ridiculously high and they will want the product for eternity
despite the fact that the product may no longer be available in the
market.

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

29

Material reserves (Million tonne)

120000

100000

80000

60000

40000

20000

0
1

9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101

Year
Fig. 16. The behaviour of material reserves.

Apart from the results presented above, it is also interesting to

see the behaviour of material reserves, material inventory worldwide
and consumable material inventory worldwide to understand how
material resources may be supplied and consumed over time. As
shown in Fig. 16, material reserves are expected to decline exponentially. The exponential drop in reserves represents the conventional
phenomenon that if material resources are abundant and easy to
get more will be consumed at a faster rate. However, the consumption gets slower when the reserve is close to its end due to the
opposite phenomenon that is the demand gets lower as it is hard
to get material resources.
It is also interesting to observe, as Fig. 17 illustrates, in future
the consumable material inventory worldwide may be much less than
material inventory worldwide. It is an indication of possible gap in
physical availability of material and actual access to material due
to geopolitical and economic situation of the world in a given time,
which has been illustrated by reinforcing loop R1 and discussed
in Section 7.1. The graph show that at the beginning of simulation the nations that are holding material resources are supplying
materials as per demand from the rest of the world. As the time
passes, material consumption rate escalates which causes material

producing nations to be more concerned about protecting their

material resources. Their actions to protect material resources
cause the gap between available material and actual access to
material resources to increase. It means decision makers should
not make their policies by only estimating physical availability of
materials but also should considered other externalities that may
prevent the material producing nations to export materials to rest
of the world. A drop and a slight rise in consumable material inventory worldwide is caused due to the fact that both the demand and
the supply of material at that point becomes quite low causing
the material producing nations to be a bit generous before they
continue to control the access to material resources again.

8. Discussion
It is appropriate to start by mentioning the classic argument
about system dynamics models which is, system dynamics models
are not to predict the future but to present a range of alternative scenarios that may be available within the given boundary and under
the dened assumptions. Even though system dynamics models

Material inventory worldwide (Million tonne)

Consumable material inventory worldwide (Million tonne)
7000
6000
5000
4000
3000
2000
1000
0
1

9 13 17 21 25 29 33 37 41 45 49 53 57 61 65 69 73 77 81 85 89 93 97 101

Year
Fig. 17. The behaviour of material inventory worldwide and consumable material inventory worldwide.

30

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

can be used for both qualitative and qualitative analysis, the results
of system dynamics model are more often used for qualitative analysis. In this respect, the lessons that should be learned from the
models presented in this paper are:
it is shown once again that the supply of non-renewable materials
is declining exponentially
for any enterprise to continue the business-as-usual in this
condition, emphasis should be on conserving materials at the
enterprise level
however, it is to be noted that conserving material at the enterprise level will not have signicant inuence in the global
resource scarcity problem unless a global effort is put forward.
Even though the models presented in this paper may appear
to be complex, it is quite simplied representation of many complex phenomena. Despite integration of many issues in the models,
they are still quite generic and t any enterprise and any material
of concern. The positive side of such generic model is that it may
be easily tailored to t any specic case which the authors may not
have thought of at this point. In fact, one of the aims of this modelling exercise is to keep it generic so that other important issues
such as emissions, solid waste and energy consumptions of PML
approach can be easily integrated in the models later on.
9. Conclusions
Main purpose of the modelling in this work is to aid decision
making in PML and the models clearly show relationships between
material reserve, supply, enterprises performance and need for
adopting PML approach. As the models advocate, for an enterprise,
it is perhaps not attractive to conserve a particular type of material
through PML approach which is naturally abundant. However, this
approach becomes extremely important if the material in question
is critical. An enterprise could through engineering, proper business
model and marketing may increase the share of multiple lifecycle products which eventually would help the enterprise to reduce
its dependency on material resources. This reduced dependency
on material not only provides strategic advantage but also makes
an enterprise less vulnerable in case of material scarcity and price

volatility. It also gives more time to an enterprise to reform itself in

case the market or technology forces them to switch to a new type
of material.
The other important conclusion of this work is that despite physical availability of a certain material its supply may be restricted due
to unexpected demand, economic importance or geopolitical reasons. It means that the decision making should not only be based
on estimating physical availability of certain material but dynamics
of geopolitical and market environment that controls the supply of
materials should also be considered. For an enterprise or even for
a nation, reducing dependency on material resources is extremely
important to deal with unforeseen uncertainty associated with supply of materials. PML can be an approach to reduce dependency on
material resources but only global efforts can improve the situation
substantially.
The conclusions made here appear to be obvious and perhaps
can be established through literature review as well. Nevertheless,
purpose of the modelling here is to provide a methodical approach
to develop a tool which can be used for analysing more complex
interactions and dynamics in a quantiable manner. Such a tool
should also assist decision makers in creating different scenarios
by tuning different variables of the models which is not possible
if conclusions are made only from literature review. Moreover, the
models should be scalable in terms of analysing several materials,
different enterprises and numerous numerical values. This is the
rst step in creating comprehensive model which will eventually
include other critical issues that inuence resource conservation
and its implications.
Acknowledgement
This work has been conducted as part of the ResCoM project
that has received funding from the European Unions Seventh
Programmes for research, technological development and demonstration under grant agreement No 603843.
Appendix A.
This appendix contains the denition of the terms used in the
models.

Term used in the models

Denition

Material resource

Regroups all identied resources. It is a natural concentration of minerals or a body of rock that is, or may become, of
potential economic interest as a basis for the extraction of a mineral commodity. A resource has physical and/or
chemical properties that make it suitable for specic uses and it is present in sufcient quantity to be of intrinsic
economic interest. It encompasses mineral reserves and reserve base plus other identied resources which could be
exploited in the future if required according to the economic situation
Reserve base includes the mineral reserves plus those parts of the resources that have a reasonable potential for
becoming economically available within planning horizons beyond those that assume proven technology and current
economics (European Commission, 2010)
Is the part of the resource which has been fully geologically evaluated and is commercially available for mining
(European Commission, 2010)
Is equals to Material reserves/Desired material reserve
Is a parameter assumed to be equal to 1/3 of material reserves, it means that if the amount (stock) of material reserves
become lower than the 1/3 of initial amount it will become more difcult to extract material
Is a graph function of ratio of material reserves. It means that if the reserves of material gets less (in this case lower
than 1/3 of initial amount) it becomes more difcult (or costly) to extract material, hence, economically less attractive
to process (in other word extract and convert) material
Is the rate at which material is extracted and converted to its new from. In the models material reserves represents the
reserves of iron ore which is converted to crude steel
Is the inventory of crude steel in different producing countries that stays with the producers (not available for rest of
the world for consumption)
Is the rate at which crude steel is exported to different countries where they are consumed
Is the inventory of crude steel that is ready to be consumed by the countries all over the world
Is the rate at which crude steel is consumed by secondary industries after converting it to different forms of steel
Is same as Material consumption rate
Is the time between a consumer places an order to receive certain amount of material and when the material arrives
to the consumer
Represents a maximum duration (in this case 4 years) that is acceptable by a consumer as the waiting time to get
supply of material

Material reserves (addressed as mineral

reserves in the referred report)
Ratio of material reserve
Desired material reserve
Economic interest of processing material

Material processing rate

Material inventory worldwide
Material delivery rate
Consumable material inventory worldwide
Material consumption rate
Material supply rate
Delay in material supply
Desired delay in supply

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

Term used in the models

Denition

Supply delay ratio

Material deliverability

Is equal to Desired delay in supply/Delay in material supply

Is a graph function of supply delay ratio which represents that if supply delay ratio increases material deliverability will
also increase
Is equals to Desired material deliverability/Material deliverability
Is a parameter assumed to be equals to 1. It is expected that material deliverability and desired material deliverability
is always 1 meaning that material producing countries should deliver materials as per the demand (desired) from the
rest of the world and should not control export of material due to for example, political reasons
Is a graph function of Material deliverability ratio which describes that if Material deliverability ratio increases
Perception of rarity will also increase
Is equals to Product demand-Manufacturing rate. It is basically the difference between supply and demand
Is the rate at which material needs to be purchased in order to ll the gap in manufacturing
Is equals to Material reserves/Material consumption rate
Is a graph function that describes the inverse relationship with time to exhaust material reserve. It is assumed that if
time to exhaust material reserve is higher, material supply stress index will be lower
Is a graph function that describes the positive relationship with price of product. It is assumed that if price of product is
higher, material price stress index will be also higher

Material deliverability ratio

Desired material deliverability

Perception of rarity
Gap in manufacturing
Desired material purchase rate
Time to exhaust material reserve
Material supply stress index
Material price stress index

Appendix B.
This appendix contains the equations of the models (the equations are organized in alphabetic order).
Equations of the stocks and the ows
STOCK:
Consumable material inventory worldwide(t) = Consumable material inventory worldwide(t dt) + (Material delivery rate Material consumption rate) * dt
INIT Consumable material inventory worldwide = 1186
INFLOWS:
Material delivery rate = (Material inventory worldwide/Time to delivery)*Material deliverability
OUTFLOWS:
Material consumption rate = MIN((Consumable material inventory worldwide/Time to consume),Material demand)
STOCK:
Current material processing capacity worldwide(t) = Current material processing capacity worldwide(t dt) + (Change in capacity) * dt
INIT Current material processing capacity worldwide = 1205
INFLOWS:
Change in capacity = (Current material processing capacity worldwide*Material processing capacity growth factor)/Time to change the capacity
STOCK:
Material demand(t) = Material demand(t dt) + (Change in demand rate) * dt
INIT Material demand = 1205
INFLOWS:
Change in demand rate = ((Material demand*Material demand growth factor)/Time to update material demand)*Perception of rarity
STOCK:
Material reserves(t) = Material reserves(t dt) + (Rate of conversion of resources to reserve Material processing rate) * dt
INIT Material reserves = 170,000/1.4
INFLOWS:
Rate of conversion of resources to reserve = 0
OUTFLOWS:
Material processing rate = MIN(Current material processing capacity worldwide,((Material reserves/Time to process) + (Material reserves/
Time to process*Material processing growth factor)))*Economic interest of processing material

STOCK:
Material supply backlog(t) = Material supply backlog(t dt) + (Demand of material Material supply rate) * dt
INIT Material supply backlog = 1
INFLOWS:
Demand of material = Material demand
OUTFLOWS:
Material supply rate = Material consumption rate
STOCK:
Material inventory worldwide(t) = Material inventory worldwide(t dt) + (Material processing rate Material delivery rate) * dt
INIT Material inventory worldwide = 1186
INFLOWS:
Material processing rate = MIN(Current material processing capacity worldwide,((Material reserves/Time to process) + (Material reserves/
Time to process*Material processing growth factor)))*Economic interest of processing material
OUTFLOWS:
Material delivery rate = (Material inventory worldwide/Time to delivery)*Material deliverability
STOCK:
Material resources(t) = Material resources(t dt) + (Rate of conversion of resources to reserve) * dt
INIT Material resources = 1,000,000
OUTFLOWS:
Rate of conversion of resources to reserve = 0

31

32

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033

STOCK:
Price of product(t) = Price of product(t dt) + (Change in price) * dt
INIT Price of product = 80
INFLOWS:
Change in price = (Price of product*Perception of rarity)/Time to update the price
STOCK:
Product demand(t) = Product demand(t dt) + (Change in demand) * dt
INIT Product demand = 50,000
INFLOWS:
Change in demand = ((Product demand*Manufacturing growth factor)/Time to update product demand)*Effect of price on demand
STOCK:
Raw material inventory(t) = Raw material inventory(t dt) + (Raw material purchase rate Manufacturing rate) * dt
INIT Raw material inventory = 50,000
INFLOWS:
Raw material purchase rate = MIN(Possible raw material purchase rate, (Manufacturing rate + Desired material purchase rate))
OUTFLOWS:
Manufacturing rate = MIN((Raw material inventory/Time to manufacture), Product demand)*Effect of price on demand
Equations of the variables
Delay in material supply = Material supply backlog/Material supply rate
Desired material purchase rate = Gap in manufacturing-(Perceived material recovery rate/Potential gain of PML)
Gap in manufacturing = Product demand-Manufacturing rate
Material deliverability ratio = Desired material deliverability/Material deliverability
Perceived material recovery rate = Material recovery ratio*Manufacturing rate*Estimated market size of PML*Potential gain of PML
Perceived price ratio = Price of product/Perceived price of product
Possible raw material purchase rate = (Material consumption rate*Ratio of material consumption nationwide*Ratio of material
consumption at enterprise level)/Quantity of material per product + ((Material consumption rate*Ratio of material consumption nationwide*Ratio of material
consumption at enterprise level*Growth factor of nationwide material consumption)/Quantity of material per product)
Potential gain of PML = Material supply stress index*Weight average + Material price stress index*(1 Weight average)
Ratio of material reserve = Material reserves/Desired material reserve
Supply delay ratio = Desired delay in supply/Delay in material supply
Time to exhaust material reserve = Material reserves/Material consumption rate
Equations of graph functions
Economic interest of processing material = GRAPH(DELAY1(Ratio of material reserve,Delay in recognizing drop in reserve)) (0.00, 0.00), (0.131, 0.015), (0.263, 0.05),
(0.394, 0.1), (0.525, 0.17), (0.656, 0.3), (0.787, 0.42), (0.919, 0.58), (1.05, 0.695), (1.18, 0.795), (1.31, 0.87), (1.44, 0.915), (1.58, 0.945), (1.71, 0.97), (1.84, 0.99),
(1.97, 1.00), (2.10, 1.00)
Effect of price on demand = GRAPH(DELAY(Perceived price ratio,5))(0.5, 1.00), (0.792, 0.99), (1.08, 0.965), (1.38, 0.875), (1.67, 0.745), (1.96, 0.54), (2.25, 0.385),
(2.54, 0.28), (2.83, 0.17), (3.13, 0.105), (3.42, 0.055), (3.71, 0.02), (4.00, 0.00)
Material deliverability = GRAPH(DELAY(Supply delay ratio,5))
(0.00, 0.1), (0.179, 0.1), (0.357, 0.1), (0.536, 0.109), (0.714, 0.172), (0.893, 0.289), (1.07, 0.424), (1.25, 0.559), (1.43, 0.69), (1.61, 0.816), (1.79, 0.915), (1.96, 0.969), (2.14,
0.987), (2.32, 1.00), (2.50, 1.00)
Material price stress index = GRAPH(Perceived price ratio) (0.5, 1.00), (0.792, 1.00), (1.08, 1.01), (1.38, 1.03), (1.67, 1.06), (1.96, 1.12), (2.25, 1.20), (2.54, 1.29), (2.83,
1.36), (3.13, 1.42), (3.42, 1.47), (3.71, 1.50), (4.00, 1.50)
Material supply stress index = GRAPH(Time to exhaust material reserve) (0.00, 1.50), (5.56, 1.50), (11.1, 1.50), (16.7, 1.50), (22.2, 1.48), (27.8, 1.42), (33.3, 1.34),
(38.9, 1.24), (44.4, 1.18), (50.0, 1.13), (55.6, 1.09), (61.1, 1.06), (66.7, 1.03), (72.2, 1.01), (77.8, 1.00), (83.3, 1.00), (88.9, 1.00), (94.4, 1.00), (100, 1.00)
Perception of rarity = GRAPH(DELAY1(Material deliverability ratio,5))
(0.1, 0.00), (0.217, 0.00), (0.333, 0.02), (0.45, 0.055), (0.567, 0.14), (0.683, 0.24), (0.8, 0.365), (0.917, 0.495), (1.03, 0.625), (1.15, 0.775), (1.27, 0.93),
(1.38, 0.98), (1.50, 1.00)

References
Acharya SR, Saeed K. An attempt to operationalize the recommendations of the
limits to growth study to sustain the future of mankind. Syst Dyn Rev
1996;12(4):281304.
Alcamo J, Leemans R, Kreileman E. Global Change Scenarios of the 21st Century,
Results from the IMAGE 2. 1 Model. Oxford: Elsevier Science Ltd; 1998.
Asif FMA. Resource Conservative Manufacturing: A New Generation of
Manufacturing. Stockholm: KTH Royal Institute of Technology; 2011
[dissertation].
Asif FMA, Bianchi C, Rashid A, Nicolescu CM. Performance analysis of the closed
loop supply chain. J Remanuf 2012;2(4).,
http://dx.doi.org/10.1186/2210-4690-2-4.
Barbiroli G, Focacci A. Natural nonrenewable resource development information
and knowledge. In: Principles of Sustainable Development. Encyclopedia of
Life Support Systems. Paris: EOLSS Publisher; 2009. p. 189213.
Boudreau J, Choi E, Datta R, Rezhdo O, Saeed K. Platinum supply and the growth of
fuel cell vehicles. In: Ford A, Ford DN, Anderson EG, editors. System Dynamics
Society: Proceedings of the 27th International Conference. New Mexico:
System Dynamics Society; 2009. p. 6293.
Bunker S. The political economy of raw materials extraction and trade. In: Socolow
R, editor. Industrial Ecology and Global Change. New York, NY: Cambridge
University Press; 1994. p. 43750.
Denyer D, Traneld D. Using qualitative research synthesis to build an actionable
knowledge base. Manage Decis 2006;44(2):21327.

European Commission. Fact Sheet: Iron Ore, Polinares, EU Policy on Natural

Resources, Polinares Working Paper No 4. Dundee: Sponsored by the European
Commission; 2012.
European Commission. Critical Raw Materials for the EU, Report of the Ad-hoc
Working Group on Dening Critical Raw Materials. Brussels: Sponsored by the
European Commission; 2010.
European Commission. Communication from the Commission to the European,
The Raw Materials InitiativeMeeting Our Critical Needs for Growth and Jobs
in Europe. Brussels: Sponsored by the European Commission; 2008.
Forrester JW. Some basic concepts in system dynamics. In: Report No. D-4894.
Massachusetts Institute of Technology, Sloan School of Management; Jan 2009.
Georgiadis P, Efstratios A. The impact of two-product joint lifecycles on capacity
planning of remanufacturing networks. Eur J Oper Res 2010;202(2):
42033.
Georgiadis P, Vlachos D. The effect of environmental parameters on product
recovery. Eur J Oper Res 2004;157(2):44964.
Georgiadis P, Vlachos D, Tagaras G. The impact of product lifecycle on capacity
planning of closed-loop supply chains with remanufacturing. Prod Oper
Manage 2006;15(4):51427.
Gleiche B, Mosig B, Reinwald D. Contributing to knowledge-based decision
support: a system dynamics model regarding the use of non-renewable
resources. In: European Conference on Information Systems. Proceedings of
the 19th European Conference on Information Systems; 2011, p.181.
Goldey CL, Kuester EU, Mummert R, Okrasinski TA, Olson D, Schaeffer WJ. Lifecycle
assessment of the environmental benets of remanufactured
telecommunications product within a green supply chain. In: Sustainable

F.M.A. Asif et al. / Resources, Conservation and Recycling 101 (2015) 2033
Systems and Technology (ISSST). IEEE 2010: International Symposium; 2010.
p. 16.
Gutowski TG, Sahni S, Boustani A, Graves SC. Remanufacturing and energy savings.
Environ Sci Technol 2011;45(10):45407.
Hauser W, Lund RT. Remanufacturingan American perspective. Green
manufacturing. In: ICRM 2010: Proceedings of the Fifth International
Conference on Responsive Manufacturing; 2010. p. 16.
Hauser W, Lund RT. Boston: Boston University; 2012, bu.edu [Internet] [updated
2012 Aug 17; cited 2015 Mar 9]. Available from
http://www.bu.edu/reman/RemanSlides.pdf.
Hotelling H. The economics of exhaustible resources. J Polit Econ
1931;39(2):13775.
Jackson T. Prosperity without Growth: Economics for a Finite Planet. London:
Earthscan; 2009.
Jorgenson JD. Mineral Commodity Summaries. Virginia: U.S. Geological Survey;
2012 (Sponsord by the U.S. Department of the Interior).
Kim J. Recent trends in export. In: OECD Trade Policy Papers No. 101. Paris:
Sponsored by OECD; 2010.
Korinek J, Kim J. Export Restrictions on Strategic Raw Materials and Their Impact
on Trade and Global Supply. Paris: OECD Trade Policy Studies The Economic
Impact of Export Restrictions on Raw Materials; 2010 (Sponsord by OECD).
Laurenti R, Singh J, Sinha R, Pottin J, Frostell B. Unintended environmental
consequences of improvement actions: a qualitative analysis of systems
structure and behavior. Syst Res Behav Sci 2015.,
http://dx.doi.org/10.1002/sres.2330 (Syst. Res. 2015).
Meadows DH, Meadows DL, Randers J. Beyond the Limits. US: Chelsea Green
Publishing Company; 1992.
Meadows DH, Meadows DL, Randers J. Limits to Growth, The 30-Year Update. US:
Chelsea Green Publishing Company; 2004.
Meadows DH, Meadows DL, Randers J, Behrens WW III. The Limits to Growth; A
Report for the Club of Romes Project on the Predicament of Mankind. New
York, NY: Universe Books; 1972.
Morecroft J. Strategic Modelling and Business Dynamics, A Feedback Systems
Approach. West Sussex: John Wiley & Sons, Ltd; 2007.
Rahmandad H, Sterman JD. Reporting guidelines for simulation-based research in
social sciences. Syst Dyn Rev 2012;28(4):396411.

33

Rashid A, Asif FMA, Krajnik P, Nicolescu CM. Multiple life cycles product systems:
redening the manufacturing paradigm for resource efcient production and
consumption. In: Seliger G, Kilic SN, editors. Sustainable Manufacturing. GCSM
2012: 10th Global Conference on Sustainable Manufacturing; 2012. Oct
31Nov 2, 2002, Istanbul, Turkey. CIRP; 2012. p. 16.
Rashid A, Asif FMA, Krajnik P, Nicolescu CM. Resource conservative manufacturing:
an essential change in business and technology paradigm for sustainable
manufacturing. J Cleaner Prod 2013;57:16677.
Rodrigues LL, Motlagh FG, Ramesh D, Kamath V. System dynamics model for
remanufacturing in closed-loop supply chains. In: Husemann E, Lane D,
editors. System Dynamics Society: Proceedings of the 30th International
Conference; 2012.
Steinhilper R. Remanufacturing The Ultimate Form of Recycling. rst ed. Stuttgart:
Fraunhofer IRB Verlag; 1998.
Sterman J. Business Dynamics, Systems Thinking and Modeling for a Complex
World. USA: McGraw-Hill Companies; 2000.
Sutherland JW, Adler DP, Haapala KR, Kumar V. A comparison of manufacturing
and remanufacturing energy intensities with application to diesel engine
production. CIRP AnnManuf Technol 2008;57(1):58.
Tilton JE. Exhaustible resources and sustainable development: two different
paradigms. Resour Policy 1996;22(1/2):917.
Turner GM. On the cusp of global collapse? Updated comparison of the limits to
growth with historical data. GAIAEcol Perspect Sci Soc 2012;21(2):
11624.
Van Vuuren DP, Strengers BJ, De Vries HJ. Long-term perspectives on world metal
usea system-dynamics model. Resour Policy 1999;25(4):23955.
Vlachos D, Georgiadis P, Iakovou E. A system dynamics model for dynamic capacity
planning of remanufacturing in closed-loop supply chains. Comput Oper Res
2007;34(2):36794.
Wils A. End-use or extraction efciency in natural resource utilization: which is
better? Syst Dyn Rev 1998;14(23):16388.
World Steel Association. Fact Sheet Raw Material. Brussels: Sponsored by World
Steel Association; 2011.
World Steel Association. World Steel in Figure 2012. Brussels: Sponsored by World
Steel Association; 2012.