Engineering Sustainability Volume 166 Issue ES5 Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
ice | proceedings
Proceedings of the Institution of Civil Engineers Engineering Sustainability 166 October 2013 Issue ES5 Pages 258–271 http://dx.doi.org/10.1680/ensu.12.00040 Paper 1200040 Received 10/11/2012 Accepted 12/06/2013 Keywords: environment/sustainability
ICE Publishing: All rights reserved
Scenarios analysis through a futures performance framework Dexter V. L. Hunt MEng, PhD School of Civil Engineering, College of Engineering and Physical Sciences, University of Birmingham, Birmingham, UK
Ian Jefferson BEng, PhD, DIS, FGS School of Civil Engineering, College of Engineering and Physical Sciences, University of Birmingham, Birmingham, UK
Christopher D. F. Rogers EurIng, BSc, PhD, CEng, MICE, MIHT School of Civil Engineering, College of Engineering and Physical Sciences, University of Birmingham, Birmingham, UK
Determining an urban environment’s built form, whilst engineering the underlying infrastructure upon which it depends, has all too often been predicated on historical trends, legacies and hindsight. In the 21st century increased interconnectivity and interdependence requires management of complex systems (i.e. coupled human, natural and technological systems) where decision making and planning for the future are far from straightforward. Moreover, a greater awareness of sustainability and resiliency issues (e.g. engineering within resource constraints and climate change such that engineering solutions continue to deliver their intended benefits in the face of change) now requires consideration of a range of possible city changes and assessment of their short- and long-term impacts. Decision making in the face of such uncertainty requires interdisciplinary foresight tools that facilitate understanding and communication between diverse ranges of stakeholders. To facilitate this process, a newly developed future performance framework (FPF) is proposed. When used in conjunction with the long-term future scenario-based urban futures (UF) methodology and the UF (Excel-based) tool, the FPF demonstrates how to embed futures thinking within an earth systems engineering framework.
1.
Introduction
Balancing the biological, social, and machine elements of modern cities is crucially important when creating environmentally sustainable, emotionally satisfying urban centres of the future (Bugliarello, 2001). This is a complex challenge. Engineers can provide unique perspectives on how to address complexity issues to help those responsible for decision-making where a vast number of interactions and interdependencies exist. Earth systems engineering (ESE) is defined as the study and practice of engineering human technology systems in such a way as to provide the required functionality while facilitating the active management of the dynamics of strongly coupled fundamental natural systems (Allenby 1999, 2001, 2002, 2005; Hall and O’Connell 2007; Schneider 2001).
As such, ESE provides a useful context for examining the potential for sustainable, and resilient, urban adaptation to global change. Resilience herein is defined as the ability of an engineering solution to continue to deliver its intended benefits in the face of perhaps radically changing future contexts (Rogers et al. 2012a). Humans exist in complex systems and it could be argued that an individual action taken in isolation at household scale will not have global ramifications that impact 258
on the earth. However, where multiple households exist, for example at the city scale, the future impact will be more significant. The top-down approach considered as part of existing city models, for example city-wide analyses (integrated assessment model of cities and its infrastructure; see Becker et al. (2013), Do¨llner et al. (2006)), is very much part of ESE thinking. All these models can be informed by a bottom-up approach such as that adopted here, where an individual building or multiple buildings are considered first. In such an approach a consideration is needed of material flows and infrastructure requirements inside and outside the home, so that a detailed analysis of future human actions, such as behaviour changes, and technological changes can be considered. This is vital for policymakers who are tasked with identifying the future impacts of policy interventions, not least those that will move society towards a more carbon-neutral, global, resource-efficient environment. The importance of this research is in the translation to multiple buildings, developments, towns and city scales. For example the following questions might be tackled. (a) How do flows aggregate across scales? (b) How might an individual’s behaviour and technology adoption relate to that of a community? Is it the sum of individual buildings or something far more complex? (c) Is it possible to change the nature of urban infrastructures? These
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
questions could have global significance when the findings are aggregated and applied through existing city analysis models.
The future performance framework (FPF) described in this paper (Section 2) provides a methodology that supports the coupled human–natural–engineering view of the world with a focus on coupled systems that exist in cities and that the ESE framework advocates. An urban futures (UF) tool developed to facilitate quantification (with respect to urban flows) is described in detail (in Section 3) and implications for the advancement of theory and practice of this new methodological approach are discussed prior to conclusions being drawn in the final section.
In ESE a key engineering concept to examining environmental problems is through a systems analysis approach (the wellestablished principles of which are described by Dandy et al. (2008) and Kossiakov et al. (2011)). Adopting a systems analysis approach to interrogate complex city centre landscapes and explore the potential for sustainable provision of urban infrastructure assets would appear to be wholly appropriate, and its potential application in this respect is investigated further as part of this paper. According to Gibson (1991) a systems analysis approach should consist of six steps & determining the goals of system & establishing the criteria for ranking alternative candidates & developing alternative solutions & ranking alternative candidates & iterating & action.
Allied to the above is the development of the following three key stages (a)
(b)
(c)
descriptive scenario – a baseline that tells us where we are, how things got this way and what is good and bad about it; that is, what should be kept and what might be changed normative scenarios – future scenarios that tell us where we want to be or where a system should operate under ideal conditions transitive scenarios – implementation processes/strategies; that is, how to get from the descriptive scenario to the normative scenario; in other words, it provides a clear roadmap from where we are to where we want to be (Van der Heijden 1997).
All three scenario stages require complex analysis, using numerous well-defined variables, indicators, benchmarks and datasets, and therefore if system changes are to be implemented, their impacts must be fully understood and clearly communicated to a range of stakeholders (Elias et al. 2000; Mitchell et al. 1997). This requires a robust possibility space to be used since many possible iterations can exist in stages (b) and (c). A tool that can facilitate this process by allowing for a system to be pushed and pulled in real time, while providing instantaneous graphical interpretations of impacts, will result in the implementation of a more innovative and sustainable solution while allowing for trade-offs to be managed to best effect. Lombardi et al. (2011a) show that elucidating tensions and trade-offs early on (i.e. at the visioning stage) increases the likelihood of more sustainable solutions being adopted. These are key elements in the advancement of ESE.
2.
The FPF: moving from qualitative to quantitative analysis
Explorative scenarios permit the inclusion of seemingly unlikely scenarios (i.e. they explore future possibilities that, unlike normative scenarios, may not take us to where we aspire to be). These could have major local and global implications. While these scenarios are not adopted directly within a systems analysis approach, they form an underlying feature of a UF analysis within an FPF. Using a combined systems analysis and UF analysis within an ESE framework might reveal significant benefits. Explorative scenarios allow the users to step inside worlds that are possible, plausible, relevant, and internally consistent (Kahn and Weiner, 1967; Schwartz, 1991). Moreover, if the future worlds are chosen to be at the extreme of likely developments in the far future and cover the full range of possible directions in which the world might travel (Hunt et al., 2012a; Rogers et al., 2012b), today’s urban solutions can be tested for their likely efficacy and today’s investments judged for their likely long-term value. The process for deriving explorative scenarios requires firstly consideration of the driving forces of change that operate in, for example, social, technological, economic, environmental and political (Steep) or Pester, where regulation is also included (Foresight 2005; Stout, 2002). These should be sorted according to aspects of greatest importance and uncertainty. When considering future demand profiles (e.g. supply and disposal streams (flows) for water, energy and waste within any development and at any scale) the two influencing factors with greatest importance might be considered to be social and technological. For example, an electrical demand exists only because society has chosen to adopt and use a range of technologies in the home (e.g. the TV, washing machines, dishwashers, the microwave). Demand is highly dependent on the existence of both of these drivers; damage to the environment occurs as a consequence of these choices, and policy can only legislate and provide direction for the choices made (e.g. by restricting the adoption of high water-flow power showers, enforcing metering and hose pipes in droughts). Economics influences individual choice through a trade-off between running costs and the cost of investment in the technology (e.g. I may choose to use my TV less because fuel 259
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
prices have increased or I may wish to purchase a more energyefficient TV).
this as a world where consumers have to amend their behaviour because resources are scarce and modern technologies have yet to become widespread (e.g. in less developed areas of India). Quadrant 3 represents a world in which user behaviour is worse but steps to improve technological efficiency have been employed (e.g. a low-flow shower). One might envisage this as a world where technology fixes are widely adopted and enforced through a general strengthening of policy, while changes in behaviour are left to individuals’ personal choice (e.g. new-build homes in the UK). Finally, quadrant 4 represents a world in which the best technological efficiencies are accompanied by step changes in user behaviour (e.g. a # 4 min low-flow shower). One might envisage this world as an exemplar of sustainability where socio-technological changes accompany behavioural change driven by personal resolve rather than through a strengthening of policy, whether people’s values are influenced by resource scarcity (e.g. Australia) or not (e.g. eco-villages in Sweden). Figure 1 defines a very useful immersive futures space that facilitates interdisciplinary communication.
There is uncertainty surrounding future changes in social and technological fields (Figure 1). The possibilities range from a significant improvement to a significant worsening (both compared to today). The probability of these uncertainties is deliberately not assigned – to do so automatically assumes that we know what will happen in the future, something that explorative scenarios aim to move us away from. By plotting these drivers of change on two opposing axes, commonly referred to as an axis of uncertainty approach (Hunt et al., 2012a), four quadrants can be formed. Each would traditionally be referred to as an explorative scenario describing a particular socio-technological world view. In qualitative terms, quadrant 1 represents a world in which user behaviour (e.g. the duration of shower) and technology (e.g. the flow rate of the shower) have changed for the worse (e.g. . 8 min power showers are the norm). One might envisage this as a world in which consumers use considerably more than one planet’s worth of resources (e.g. the sociotechnological trends prevalent in the USA). Quadrant 2 represents a world in which technological efficiency is worse, but user behaviour has improved significantly (e.g. a # 4 min shower with no water-saving technology). One might envisage
Sustainability might traditionally be associated with quadrant 4 alone. Here technological efficiency improves and is accompanied by similar (willingly adopted) step changes in user behaviour; that is, both combine to deliver benefits (Lombardi et al., 2011b). In contrast, resilience requires much
Technological efficiency significantly worse
QUADRANT 1
QUADRANT 2
User behaviour significantly worse
User behaviour significantly improved
QUADRANT 3
Technological efficiency significantly improved
Figure 1. Qualitative analysis of demands using an ‘axis of uncertainty’ approach
260
QUADRANT 4
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
broader considerations; that is, a deeper understanding of the likely impacts that could occur if society moves into any of the four quadrants (Rogers et al., 2012a, 2012b).
& Fortress world – those who live in privileged enclaves and
operate in a similar manner to those under market forces (FWH). The have-nots (FWHN) live outside the enclaves. They have limited resources and capital and must preserve scarce resources through a step change in behaviour.
The UF long-term explorative scenarios for 2050 have previously been adopted to test the resilience of engineering solutions adopted today in the name of sustainability (Rogers et al., 2012b). Adapted from global scenario group scenarios and tailored for UK analysis (Boyko et al., 2012; Hunt et al., 2012a; Lombardi et al., 2012), these four scenarios define a clear set of archetypal features that are evident in more than 280 scenarios (Hunt et al. 2012a). The scenarios are particularly useful because they provide clarity on how the key drivers of change (Steep) can push or pull society towards or away from each future. Figure 2 shows their approximate location when assigned to the axes proposed in Figure 1.
The map of the world in Figure 2 refers to a descriptive scenario and shows the situation today. These scenarios have been adopted with respect to analysing water demands at city scale in Lancaster, UK, by Farmani et al. (2012). Engineers seek to quantify what they see, and faced with two (nominally scaled) axes they are likely to question the exact location of any world within each of these quadrants rather than leaving their placement down to a qualitative judgement, as in Figure 2. Hunt et al. (2012a) suggest that exact location of a scenario in a quadrant diagram allows for greater specification and therefore that a numerically based tool of analysis is required. These authors considered the impact on site water demands when implementing changes to technological efficiency with no net change in behaviour in Lancaster (Figure 2). With respect to technological efficiency, quantification is fairly straightforward and not readily open to criticism. However, quantifying user behaviour begins to throw up questions of whether we can (or should) try to quantify the intimate, private, personal behaviour of individuals. Should these remain unquantified or, as is typical practice, be assumed
& A new sustainability paradigm (NSP) – individuals and
society change their behaviour and adopt efficient technologies because it is considered the right thing to do. Policy is not required as a driving force for change. & Policy reform (PR) – improvements to technological efficiency are enforced through tightening of policy, but unchecked user behaviour exists. & Market forces (MF) – free market economics allows unchecked user behaviour, and technological efficiency to deteriorate.
Technological efficiency
Farmani et al. (2012)
significantly worse
Hunt et al. (2012b)
QUADRANT 2
QUADRANT 1
MF
FWH FWHN
User behaviour significantly worse
User behaviour significantly improved
PR
NSP
QUADRANT 4
QUADRANT 3
Technological efficiency significantly improved
Figure 2. Possibility space and explorative scenarios used for UF water demand analyses
261
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
to be static? (The analysis by Hunt et al. (2012b) shown in Figure 2 adopted this approach.)
user behaviour axis. A marker showing our current situation (i.e. our descriptive scenario) can be located where the axes cross. A marker representing a 50% reduction in demand due to changes in user behaviour (with no change in technology) can be placed mid-location on the right limb of the horizontal axis. Likewise a marker representing a 50% reduction in demand due to changes in technological efficiency (with no change in user behaviour) can be placed on the bottom limb of the vertical axis. A fourth marker representing a combination of both options can then be placed as shown in Figure 3(a). (The 25 marker indicates the demand is now 25% of the current situation.) Likewise additional markers can be placed anywhere within this grid to represent increases in demand, as shown in quadrant 1.
Consequently this paper proposes a method for defining more clearly a possibility space, in this case specific to current and future demands, and in so doing begins to examine the problems for improved quantification that may exist. The first step is to adopt an exact scale of change for both axes (Figure 3(a)). For this paper water usage is considered and a scale running from a 100% increase in demand to a 100% decrease in demand is adopted. With a current norm of 300 L/d for a dwelling type (dual occupancy), moving along the technology axis would yield extremes of 0–600 L/d. The same is true for moving along the % Change in demand due to user behaviour % Change in demand due to technological efficiency
100
80
60
40
20
0
–20
–40
–60
–80 –100
100 80 60
150
225
40 20 150
0 –20
50
Descriptive scenario
–40
25
50 –60 –80 –100 (a)
% Change in demand due to user behaviour % Change in demand due to technological efficiency
100
80
60
40
20
0
–20
–40
–60
–80 –100
100 80 60
Demands increase
1
With this defined and by looking more closely at Figure 3(a) it is intuitive that there must be a range of contours of equal demand throughout the possibility space. The markers can be plotted as in Figure 3(a) and the positions of equal magnitude joined to define the contours or, alternatively, an equation for plotting generic contours with any pairing of axes can be defined.
40 Descriptive scenario
20 0 –20 –40
As more markers are placed it becomes apparent that contours must exist. The first of these, denoted by a thick black line in Figure 3(b), separates demand increases (shaded region above line) from demand decreases (non-shaded region below line), that is, a contour of no change in user demand. On this contour scenario 1 represents a world where user behaviour has improved by 40% (perhaps due to people’s recognition of the need to change for the good of the planet or through policy measures requiring metering which makes individuals aware of the cost of using resources). However technological efficiency has decreased by 60% (perhaps caused by ageing technologies and system leakage, in the case of water), meaning that no overall change in demand has occurred. In scenario 2 these percentage changes are reversed on each axis, with the same overall effect; that is, demand remains unchanged. This is an important line to define as it shows that any efforts that are made to reduce demands in one direction can be directly counteracted if they are not kept in check in the other direction, and accounts for the unintended outcome of more efficient technologies being used for longer in the belief that ‘technology has solved the problem so I can use it (whatever resource) without restriction’.
2
–60
Demands decrease
–80 –100 (b)
In Figure 4 the per cent change along the vertical axis, the per cent change along the horizontal axis and the per cent change in contour (A) are generically defined. As such the following equation can be derived (Equation 1).
Figure 3. (a) A possibility space for analysing % changes in demand; (b) moving from qualitative to quantitative analysis for demands – defining the possibility space (contour showing no change in demand)
262
1a.
1z
Y 100
X A 1z ~ 1z 100 100
(N.B. The values of X and Y can be positive and negative along the scales.) This can be checked by using the values shown in
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
% Change in X Decrease
Increase 100 100
80
60
40
20
0
–20
–40
–60
–80
–100
A = 400%
80 60 40
A = 200%
% Change in Y
A = 100%
20 0 –20
A = 100%
–40 –60
A = 20%
–80 –100
A = 0%
Decrease
Figure 4. A fully quantified generic possibility space (contours at 20% spacing)
Figure 3(a) (if X 5 250 and Y 5 250, then A 5 25, and if X 5 50 and Y 5 50, then A 5 225). To facilitate plotting on the grids Equation 1a can be rearranged as follows.
00
1 1 A BB C C 1b. Y ~@@ 100 A{1A|100 X 1z 100 Figure 4 shows also the resulting contours when using values of A ranging from 400 to 0%. When A 5 100 (i.e. the 100% contour) no change in household (or neighbourhood, or city) demands occur; when A 5 400 (i.e. the 400% contour) demands quadruple; when A 5 0 (i.e. the 0% contour) demand has been removed completely, initiated through changes in technology and/or user behaviour. If this were used for water resources the user can plug in different assumptions about user behaviour/technology through a water resources model (e.g. the UF tool, see Section 3) to derive the changes in demand. This generic plot can then be used for analysing explorative, normative and transitive scenarios. A long-term normative scenario might be the achievement of 80% reduction in resource flows (or emissions) in cities (Figure 5). This proves extremely useful because it shows explicitly that the exact same level of performance can be achieved in numerous ways along the 80% reduction contour. What is most important is that it shows also that city well-being or, perhaps, city liveability is
not just a matter of achieving a specified level of performance (e.g. 30 L/(person?d)) in the future. The way in which it is achieved requires a deeper understanding of how levels of performance are being (or could be) achieved (see www. liveablecities.org.uk). For example, it might be argued that C is sustainable due to the dual role of technology and behaviour, but could this be justified for A, B, D and E? Figure 6 shows the steps (i.e. long-term pathway or roadmap) that might be adopted in order to reach scenario D in Figure 5. In this demand possibility space the descriptive scenario (the current position) is represented by D-i, the normative scenario (the desired position) is represented by D-v and the transitive scenarios (the steps taken to get from D-i to D-v) are represented by D-ii, D-iii and D-iv. Taking water demands as an example, D-i would represent a water demand of 150 L/ (person?d) and D-v would represent a water demand of 30 L/ (person?d) (Table 1). The steps in between are required to enable the transition to the final scenario. In this case the transition is achieved through changes in technology alone – this very much mirrors the roadmap in current UK policy and some of the UK benchmarking systems, like the Code for sustainable homes (CSH). Each transitive scenario reflects an improvement in technological efficiency in order to achieve the specified reduction in water demands. The demand reduction could be adoption of a more water-efficient washing machine, a dual-flush toilet, a low-flow shower, a greywater (GW) system or many other technological options. Examples of how 263
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
% Change in demand due to technological efficiency
% Change in demand due to user behaviour 100 100
80
60
40
20
0
–20
–40
–60
–80 –100
80 60 A 40 Descriptive scenario
20
B
0 –20
No demand change contour
–40 C –60 80 % reduction contour
D
–80
E
–100
Figure 5. Examples of 80% demand reduction scenarios (A, B, C, D, E)
different transitive scenario benchmarks could be allocated at different time stages for different flows (i.e. energy, water and waste) are shown also in Table 1. This particular choice of descriptive, transitive and normative scenarios would achieve 80% reduction demands but simply ignores the influence (and role) user behaviour could play (see Hunt et al. (2012b)). In other words D-i to D-iv could just as easily be transposed onto the horizontal axis, in which case all reductions in demand would be achieved through behavioural changes alone. Alternatively, it could be a combination of both. The advantage of this % Change in demand due to user behaviour % Change in demand due to technological efficiency
100
80
60
40
20
0
–20
–40
–60
–80 –100
100 80 60
multilayered possibility space approach is that it can be presented in a range of forms (Figures 1–6) that are internally consistent and translatable, engaging a range of audiences (i.e. from engineers, social scientists, planners and developers to the general public) that are happier with different approaches. The refining of the possibility space allows for detailed quantitative scenarios analysis at the small scale, for example, the impact of individual or multiple technologies adopted in the home and office. The difficulty lies in translating the findings of this methodological framework using a bottom-up approach to a complex working society. It was recognised that the complexities of numerical analysis involved would require an integrated platform that allowed the manipulation and testing (in isolation and combination) of the impacts of user behaviour and technological efficiency. The next section discusses the development and application of this platform.
3.
The UF tool: quantitative analysis within a future performance framework
The process of qualitatively defining, then quantitatively deriving, a future possibility space for scenario-based analysis (illustrated in Figures 1–6) requires a robust and rigorous methodology to analyse the future efficacy of the engineering solutions that are currently being proposed, many in the name of sustainability or resilience – the UF methodology (Lombardi et al., 2012). The freely available Excel-based open-source UF tool complements the methodology by being able to provide detailed quantitative analyses of the technological options available to designers and planners and options relating to end user behaviour. The current version of the UF tool includes a range of advanced Excel features, such as macros and Visual Basic. Its main deliverable was a quantitative evidence-base for testing the resilience of a range of engineering solutions through explorative future scenarios (e.g. Monte Carlo) analysis. The Excel platform was chosen due to its widespread adoption and ease of use, while capable of being modified and upgraded as new technologies emerge and as datasets age.
40 20 0
D-i
–20
D-ii
–40
D-iii
–60
D-iv
–80
D-v
–100
Figure 6. Descriptive (D-i), transitive (D-ii, D-iii, D-iv) and normative (D-v) scenarios within demand possibility space
264
When developing the tool it was firstly deemed necessary to map (at a fundamental level) the interdependencies that exist between what is built above ground and how this affects demands, flows in and flows out. The flow model shown in Figure 7 provides the basis upon which the UF tool was founded. It illustrates a range of complex interactions, from which the UF tool can test capacity and functionality requirements (or performance) of engineered solutions above and below ground. To facilitate this process the system was necessarily broken down into micro nodes (i.e. an individual building) and links (i.e. infrastructure connections for flows between micro nodes), as illustrated in Figure 7. For this example the micro node is a domestic dwelling. The UF tool allows users to quantify (and better understand) the diversity and impact of (technological, human
Engineering Sustainability Volume 166 Issue ES5
Scenario name (type) D-i (Descriptive) D-ii (Transitive) D-iii (Transitive) D-iv (Transitive) D-v (Normative)
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
Nominal longterm timeline
Specified reduction (%)
Internal water use (L/(person?d))
2012 2015 2020 2030 2050
0 20 40 60 80
150 120 90 60 30
Energy use for Waste production heating (kWh/(m2?yr)) (kg/person) 229 183?2 137?4 91?6 45?8
350 280 210 140 70
Table 1. Plausible demand quantification for descriptive, transitive and normative (80%) scenarios
and natural) operating conditions that may exist at a micro node in current and future world situations. These are key threads to ESE, as noted in the introduction. The simple operation procedures of the UF tool, that have been repeatedly refined, allow for the long-term performance and capacity requirements of nodes and (ultimately) links to be assessed. The UF tool allows the user to interrogate the impact of drivers (e.g. (a) and (b) below) on demands (e.g. (c) and (d)); subsequently the influence of other factors and the effectiveness of adaption options (listed below as (e) to (p)) can then be assessed in isolation or combination (a) user technologies (e.g. cookers, fridges, dishwashers, showers, taps, WCs) (b) user behaviour (e.g. WC flushes per day, TV watching per day) (c) energy demands (thermal and non-thermal) (d) water demands (potable versus non-potable) (e) benchmarking (e.g. CSH)
(f) alternative climates (i.e. different locations) (g) climate changes (low, medium and high variants) (h) scale (individual to development) (i) building type (i.e. houses and offices) (j) floor area, roof area and type (i.e. sloped and flat) (k) occupancy rates (i.e. single and multiple occupancy) (l) fuel supply options (i.e. electricity versus gas) (m) energy supply technologies (mains, solar thermal, photovoltaics, wind, ground source heat pumps, combined heat and power (CHP)) (n) water supply and re-use/recycling options (mains, rainwater harvesting (RWH) and greywater recycling (GW)) (o) RWH tanks (sizing and storage volumes) (p) garden watering requirements (including effect of crop type). The added value of the UF tool is its ability to easily assess a range of inflows (e.g. electricity, gas or water), user demand profiles (e.g. high to low) and outflows (e.g. stormwater, a
Natural environment
b c
Water Energy Food Consumer goods
Building Technologies Human behaviour
Micro node
d
Emissions Stormwater Wastewater Solid waste
Links
Figure 7. A micro node with inflows, demands and outflows (simplified in top right)
265
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
wastewater or emissions to the air). This allows the impact of interdependent system changes to be assessed in isolation or combination; an important factor for ESE and other engineering disciplines (Chou et al., 2009). Previous research has shown that users find it difficult to deal with more than four explorative scenarios, excluding the baseline, at a time (Foresight 2009; Stout 2002). Therefore the UF tool has been set up to compare simultaneously a maximum of five micro node scenarios for supply (mains and/or local supplies), demand (internal and external, regulated and unregulated) and disposal (including renewable, recycling and re-use options) at a micro node scale (Figure 8(a)) (‘a’ represents the environment acting upon the house, ‘b’ represents demand, ‘c’ represents inflow and ‘d’ represents outflow; the numbers relate to different micro nodes).
lowest to second highest. This represents an increase in demand of 156%. The various locations of points b1 to b5 using the FPF previously described in Section 2 are shown in Figure 10. The impact of adopting a range of garden sizes (a behavioural change) is also highlighted. An increase in garden size is represented by a gradual shift to the left. This gives just a flavour of what the FPF coupled with the UF tool can do. The user may wish to identify through sensitivity analysis the impact of any of options (e) to (p) on each micro node, for example, a change in occupancy rates, a change of location, the adoption of more efficient garden watering technologies and so on.
Figure 9(a) shows how the UF tool can be used to generate water demand at five different micro nodes, each a domestic dwelling. The water demand in b2 (147?1 L/(person?d)) is equivalent to water demands in an average UK dwelling located in Birmingham. The water demand has been changed in each micro node through changes to technological efficiency alone. Figure 9(b) shows how the UF tool can then be used to identify the effect on water demands if homeowners in b5 decided to grow home vegetables on a 50 m2 plot. The water demands in July (150?5 L/(person?d)) would increase from
The UF tool can equally be applied at the development scale (i.e. a macro node), as shown in Figure 8(b). This is a vital requirement if the results need to be aggregated upwards towards a city scale. (A represents the environment at the larger scale, B represents aggregated demand, C represents aggregated inflows, D represents the aggregated outflows.) Considering the previous examples and now imagining a macro node (i.e. at development scale) that consists of 90 micro nodes (10 6 b1, 10 6 b2, 10 6 b3, 10 6 b4 and 50 6 b5), the UF tool can directly output the total water demands (Figure 11(a)). It can be seen that, due to garden watering, the total water demands at this macro node scale are increased by a factor of 1?6 and the non-potable demands (i.e. those that
a A
b1 c1
d1
B
d2
15 x
b2
500 x
c2 b3 c3
d3
C
b4 d4
c4
D
5x 50 x
b5 d5
c5
Micro node (a)
Figure 8. (a) A micro node tool (mode A); (b) a macro node tool (mode B)
266
Macro node (b)
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
250
150
26.5
WC
Washing machine
Dishwasher
Sink
Showers
Baths
Basins
200
27.3 10.4
3.6 26.5 21.0 10.4
100 104.9
19.9
3.6 21.0 10.4
52.4
50
3.6
12.8 10.4 26.2 10.7
35.0 25.3 0
L/(person·d)
L/(person·d)
200
250
Garden watering
1.6
b1
25.3 b2
12.8
1.6
1.6
b3 Micro nodes (a)
11.5 2.4
1.6
b4
12.8 10.4 24.0
150
100
11.5 0.0
50
1.6
0
91.8
12.8 10.4 24.0
11.5 0.0 1.6
b5 (withgarden) garden) (with (b)
b5
Figure 9. (a) Water demands at five micro nodes; (b) additional garden watering demands at micro node 5 in July
be used to calculate the cumulative effect on supply requirements (i.e. water inflow) of adopting RWH in all 50 b5 micro nodes (Figure 11(b)). The impact of summer gardening would increase water supply demands by 33% (i.e. to 22?1 L/(m3?d)) as compared to 55% in total demand B.
do not require mains quality water) are increased by a factor of 2?6 (i.e. demand A 5 17?6 m3/d, assuming no garden watering, and demand B 5 25?8 m3/d, assuming garden watering in all 50 b5 micro nodes). Conventionally all these demands would be supplied through mains water. The UF tool can subsequently
% Change in demand due to user behaviour Decrease
% Change in demand due to technological efficiency
Increase 100 80 100 400%
60
40
20
0
–20
–40
–60
–80
–100
100%
80 60 40 b1 20 0
b2
–20
b3
–40
b4
–60
b5
b5
b5
b5
–80 0%
–100 Decrease 10 m garden
20 m garden
30 m garden
Figure 10. UF possibility space mapped out for micro nodes b1 to b5
267
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
30
30
25
25
Non-potable demands
3.6
Potable demands
Potable mains 20
15.1 m3/d
m3/d
20
15 5.9
15
10
10
5
RWH
10.7
10.7
Total demand A (no gardens)
Total demand B (gardens in micro node b5)
22.1
5
0
0
(a)
Total supply B (gardens in micro node b5) (b)
Figure 11. (a) Water demands at macro node in July; (b) total mains and RWH supplies at macro node in July
This is just one example of the questions that can be asked and of what the UF tool can calculate. In brief, any number of domestic dwellings (five user-defined types at a time) and any number of offices (five user-defined types at a time) can be considered in one single analysis, although the user can choose to run several analyses, each with different input parameters. In the macro-scale mode associated harmful emissions (e.g. carbon dioxide, nitrogen oxide, particulate matter PM10 and PM2.5) under a range of natural environment conditions can also be assessed. This increase in scale from one micro node to many macro nodes is a move towards a city-scale type of analysis of flows and is particularly important for ESE. There are six interactive worksheets in the UF tool covering various kinds of input/output information for micro nodes and macro nodes. To avoid information overload numerous worksheets are hidden (but accessible) and include interconnected calculations and datasets. Each of the interactive worksheets consists of a user interface panel for numerical inputs and a visual display panel for instantaneously viewing graphical outputs.
4.
Discussion
This paper paired a technological driver (efficiency) with a social driver (user behaviour) on axes of uncertainty in a contoured space to show how quantifiable improvements can be brought about. This facilitates the exploration of future demands within the home (or office) and the pressure testing of modifications to these demands. Other Steep drivers, which can greatly influence demands in homes and offices, can be 268
considered as overlying layers or lenses of information. For example, through the environmental lens different climatic conditions can be applied in order to investigate how they influence behaviour (e.g. more frequent showers in summer months) and technological performance (e.g. more water collected by a RWH system when rainfall is higher in winter months). In addition it is possible to consider how user behaviour and technologies impact upon the environment (e.g. associated carbon dioxide emissions). In the same way the economic lens can be overlain to investigate investment requirements for different water strategy options or to help to establish different water pricing strategies to influence user behaviour. The political lens can be overlain to understand better the role of current or future policies in incentivising, influencing, or even forcing changes in demand through changes in technological efficiency and/or user behaviour. The robust framework presented here therefore allows this layering of information to take place in a systematic way. It might be argued that a third axis could be included (perhaps to integrate the time element), but this would add a degree of complexity that confuses rather than enhances the key messages. The framework allows users such as planners, developers, engineers and academics to easily differentiate between how to define something that is sustainable, how to test for resilience and how to achieve the same levels of performance in a variety of ways, some of which may be neither sustainable nor resilient, and in this way it is an extremely powerful form of communication.
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
The FPF and UF tool in combination should be considered an important part of any ESE analysis and an essential undertaking prior to the implementation of mitigation and/or adaptation strategies now and in the future. Such insight is of importance to decision-makers, planners and scientists, particularly for infrastructure engineers who have the responsibility for ensuring that flows (inward and outward) can be met into the far future, ideally without disruption, within our urban environments. Collectively the FPF and UF tool allow for an abundance of explorative ‘what if’ questions related to water and energy to be answered through meaningful quantitative scenarios analysis. For example, a home or office owner (or potential purchaser) may want to know the answer to the following questions.
Future development of the tool should seek to include a wider selection of development types (e.g. hotels, commercial and schools) and locations (UK, Europe and worldwide) and allow for the geospatial location of buildings in order that the impact of nodal changes on link performance can be better assessed and understood. Perhaps even dietary choice, which will ultimately affect inflows and outflows, might be included.
& What if I have invested in a CSH level 6 house today, but
decide in the future to grow vegetables at home? While my internal demands and therefore benchmarks may remain broadly static year round (80 L/(person?d)), how will my total mains water consumption be affected at different times in the year? What if I invest in a RWH and/or GW system(s)? How might this further influence these demands? Therefore, are current benchmarking methods appropriate? If my behaviour changes also (for better or worse) what impact will this have? & What if I invest in a certain size of RWH tank today and in the future my water demands change due to technology (for better or worse)? How will this affect the ability of my system to meet demands? How will this impact on my stormwater outflow (year round) and potential for pluvial flood risk prevention (see Hunt et al. (2012b))? What if the climate changes? What if I oversize (or undersize) the RWH tank (to save money or to seek a more resilient long-term solution)? How will this influence the results (Hunt et al., 2012c)? & What if I invest in CHP infrastructure (gas or biomass) in the future? How will my energy demands be affected and how will this influence localised carbon and other air emissions (e.g. nitrogen oxide, PM10)? How might other mixes of renewable energy influence the supply/demand mix at different times in the year? What influence might changes to user behaviour and technology have in this respect? This list is not exhaustive and is intended only to give a flavour of the capabilities of the UF tool. For example the FPF and UF tool could be used to plan developments both now and in the future by specifying the right mix of dwellings to achieve a specified level of performance (i.e. benchmark) or to provide a mix of dwellings that will meet existing and future supply capabilities with minimum infrastructure investment. It might be argued that sustainable infrastructure is eminently more likely to be achieved because a smaller footprint can be planned for long before detailed design or construction takes place.
5.
Conclusions
The FPF and UF tool described in this paper support the coupled human–natural–engineering view of the world. With a focus on coupled systems that exist within cities and that the ESE framework advocates, this paper has described how users can move successfully from qualitative to quantitative scenarios-based approaches. By quantitatively contouring a generic possibility space created by pairing axes of uncertainty (for the two primary Steep drivers for change of technological efficiency and user behaviour), an improved definition and understanding of explorative, descriptive, normative and transitive scenarios can be achieved. Such a framework provides a useful contribution to ESE because the process for describing a range of possible futures, as opposed to a single possible eventuality, in a structured and informed way helps decision-makers identify critical issues of concern. When combined with a decision-making process that allows for all disciplines to be engaged at the concept stage with an equal voice, pathways to truly resilient and sustainable futures can be sought. While technological efficiency and user behaviour are the dominant influencers of resource requirements and their impacts (waste and greenhouse gas emissions), the other primary Steep drivers of economy, environment and policy in turn influence, or can be used to influence, the technological and social drivers of sustainability performance. These are conceptualised as lenses that overlie the axes of uncertainty and cause a series of movements across the contoured space – a series of discrete vectors causing incremental movements towards or away from a more sustainable state. Due to the interdependencies and complexities of what flows in and what flows out of a building or an area (a development site, community or neighbourhood), a UF tool was developed in parallel to the FPF. This tool allows users to change demands in homes and offices (through user behaviour and technological efficiency, in isolation or combination) and undertake more detailed quantitative real-time analysis/pressure testing. Users can very easily push and pull a descriptive scenario to produce explorative, transitive and normative scenarios. This is vital in order to understand the relative long-term impacts and interdependencies that exist when considering real-life interdisciplinary perspectives and issues related to human, natural and technological systems, in short ESE. The chosen platform 269
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
(Excel) is deemed wholly appropriate because it allows for widespread adoption, and moreover it allows for transparency in calculations and ease of modification and updating.
sustainability in cities: the role of indicators and future scenarios. Global Environmental Change 22(1): 245–254. Bugliarello G (2001) Rethinking urbanization. Bridge 31(1): 5– 10. Chou CC, Tseng SM and Ho TW (2009) Data collection and analysis of critical infrastructure interdependency relationships. In Computing in Civil Engineering (Caldas CH and O’Brien WJ (eds)). American Society of Civil Engineers, Reston, VA, USA, pp. 280–289, http://dx.doi. org/10.1061/41052(346)28). Dandy G, Walker D, Daniell T and Warner R (2008) Planning and Design of Engineering Systems. Taylor and Francis, London, UK and New York, USA. Do¨llner J, Kolbe TH, Liecke F, Sgouros T and Teichmann K (2006) The virtual 3D city model of Berlin – managing, integrating, and communicating complex urban information. In Proceedings of the 25th Urban Data Management Symposium UDMS 2006 in Aalborg, Denmark, 15–17 May. Elias AA, Cavana RY and Jackson LS (2000) Linking stakeholder literature and system dynamics: opportunities for research. In Proceedings of the International Conference on Systems Thinking in Management, Geelong, Australia. Deakin University, Geelong, Victoria, Australia, pp. 174–179. Farmani R, Butler D, Hunt DVL et al. (2012) Scenario based sustainable water management for urban regeneration. Proceedings of the Institution of Civil Engineers – Engineering Sustainability 165(1): 89–98, http://dx.doi. org:10.1680/ensu.2012.165.1.89. Foresight (2005) Intelligent Infrastructure Futures Scenarios toward 2055 – Perspective and Process. Office of Science and Technology, London, UK. Foresight (2009) Scenario Planning. Guidance Notes. Government Office for Science, London, UK. Gibson JE (1991) How to do a systems analysis. In The System Analysts Decalog (Scherer WT (ed)). Department of Systems and Information Engineering, University of Virginia, Charlottesville, VA, USA. Hall JW and O’Connell PE (2007) Earth systems engineering: turning vision into action. Proceedings of the Institution of Civil Engineers – Civil Engineering 160(3): 114–122, http:// dx.doi.org/10.1680/cien.2007.160.3.114. Hunt DVL, Lombardi DR, Atkinson S et al. (2012a) Scenario archetypes: converging rather than diverging themes. Sustainability 4(4): 740–772, http://dx.doi.org:10.3390/ su4040740. Hunt DVL, Lombardi DR, Farmani R et al. (2012b) Urban futures and the code for sustainable homes. Proceedings of the Institution of Civil Engineers – Engineering Sustainability 165(1): 37–58, http://dx.doi.org/10.1680/ensu.2012.165.1.37. Hunt DVL, Jefferson I and Rogers CDF (2012c) Testing the resilience of underground infrastructure solutions through an urban futures methodology. Proceedings of REAL CORP 2012, 14–16 May, Vienna, Austria, pp. 825–834.
An important outcome of the FPF and the UF tool is their potential use for planning urban regeneration or new development schemes. An appropriate mix of developments with technologies can be planned and designed to match likely future resource supply and accounting for climate change. Moreover, a radical approach to sustainable and resilient infrastructure provision could be to plan for expansion without increasing the supply capacity, using a combination of planned new and regeneration developments alongside technology retrofitted to existing developments. In the case of water supply, for example, this would obviate the need for upsizing any element of the distribution system. Instead, the maintenance of existing systems by staged refurbishment or replacement could be adopted and weaknesses or bottlenecks strategically addressed by adding redundancy (hence resilience) to the system. Confidence that this novel approach would work into the far future could be gained by applying the UF methodology to the proposed scheme in its particular context.
Acknowledgements The authors wish to thank the Engineering and Physical Sciences Research Council for their support under the current liveable cities (EP/J017 698) programme grant and previous urban futures (EP/F007 426) sustainable urban environments grant. REFERENCES
Allenby BR (1999) Earth systems engineering: the role of
industrial ecology in an engineered world. Journal of Industrial Ecology 2(3): 73–93, http://dx.doi.org/10.1162/ jiec.1998.2.3.73. Allenby BR (2001) Earth systems engineering. Editorial. The Bridge 31(1): 3–4. Allenby BR (2002) Earth systems engineering and management: the biotechnology discourse. In Engineering and Environmental Challenges: Technical Symposium on Earth Systems Engineering. National Academy Press, Washington, DC, USA, pp. 51–56. Allenby BR (2005) Reconstructing Earth: Technology and Environment in the Age of Humans. Island Press, Washington, DC, USA. Becker T, Nagel C and Kolbe TH (2013) Semantic 3D modeling of multi-utility networks in cities for analysis and 3D visualization. Progress and New Trends in 3D Geoinformation Sciences. Lecture Notes in Geoinformation and Cartography. Springer, Berlin, Germany, pp. 41–62, http://dx.doi.org/10.1007/978-3-642-29793-9. Boyko CT, Gaterell MR, Barber AGR et al. (2012) Benchmarking 270
Engineering Sustainability Volume 166 Issue ES5
Scenarios analysis through a futures performance framework Hunt, Rogers and Jefferson
Kahn H and Weiner A (1967) The Year 2000: A Framework for
Rogers CDF, Bouch CJ, Williams S et al. (2012a) Resistance and
Speculation on the Next Thirty-Three Years. Macmillan, New York, USA. Kossiakov A, Sweet WN, Seymour SJ and Biemer SM (2011) Systems Engineering Principles and Practice. John Wiley, New York, USA. Lombardi DR, Caserio M, Donovan R et al. (2011a) Elucidating sustainability sequencing tensions and trade-offs in development decision-making. Environmental Planning B 38(6): 1105–1121. Lombardi DR, Porter EJ, Barber ARG and Rogers CDF (2011b) Conceptualising sustainability in UK urban regeneration: a discursive formation. Urban Studies 48(2): 273–296, http:// dx.doi.org/10.1177/0042098009360690. Lombardi DR, Leach JM, Roger CDF et al. (2012) Designing Resilient Cities: A Guide to Good Practice. IHS BRE, Bracknell, UK. Mitchell R, Agle B and Wood D (1997) Towards a theory of stakeholder identification and salience: defining the principle of who and what really counts. Academy of Management Review 22(4): 853–886.
resilience – paradigms for critical local infrastructure. Proceedings of the Institution of Civil Engineers – Municipal Engineer 165(2): 73–84. Rogers CDF, Lombardi DR, Leach JM and Cooper RFD (2012b) The urban futures methodology applied to urban regeneration. Proceedings of the Institution of Civil Engineers – Engineering Sustainability 165(1): 5–20, http:// dx.doi.org/10.1680/ensu.2012.165.1.5. Schneider SH (2001) Earth systems engineering and management. Nature 409, 417–421, http://dx.doi.org/10. 1038/35053203. Schwartz P (1991) The Art of the Long View: Planning for the Future in an Uncertain World. Doubleday, New York, USA. Stout D (2002) The Use of Scenarios in Foresight 1994–1999. An Information Document Prepared for the OST. Office of Science and Technology, London, UK. Van der Heijden K (1997) Scenarios: The Art of Strategic Conversation. John Wiley, Chichester, UK.
WHAT DO YOU THINK?
To discuss this paper, please email up to 500 words to the editor at
[email protected]. Your contribution will be forwarded to the author(s) for a reply and, if considered appropriate by the editorial panel, will be published as discussion in a future issue of the journal. Proceedings journals rely entirely on contributions sent in by civil engineering professionals, academics and students. Papers should be 2000–5000 words long (briefing papers should be 1000–2000 words long), with adequate illustrations and references. You can submit your paper online via www.icevirtuallibrary.com/content/journals, where you will also find detailed author guidelines. 271
