Agent-based Modelling for Green Space Allocation in Urban Areas

Factors Inﬂuencing Agent Behaviour

Marta Vallejo, Verena Rieser and David Corne

School of Mathematical & Computer Sciences, Heriot-Watt University, Edinburgh, U.K.

Keywords:

Agent-based Systems, Location-Allocation Problem, Green Space Allocation, Cellular Automata, Spatial

Optimisation, Multicriteria Analysis.

Abstract:

The task of green space allocation in urban areas consists of identifying a suitable site for allocating green

areas. In this proposition paper we discuss about a number of factors like crowdedness, design, distribution

and size that could discourage inhabitants to visit a certain green urban area. We plan to cluster our urban

residents into several population segments using an Agent-Based Model and study the system in different

predeﬁned scenarios. The overall objective of this work is to provide spatial guidance to planners, policy-

makers and other stakeholders, and shed light on potential policy conﬂicts among standard policy criteria and

user preferences. We will evaluate this potential within a targeted stakeholder workshop.

1 INTRODUCTION

Pressure on green spaces close or within the city

boundaries is signiﬁcant and likely to grow (Glick-

man, 1999). These areas have been increasingly

recognised that are important components of urban

ecosystems, providing various kind of important envi-

ronmental and social services (Haviland-Jones et al.,

2005; Takayama et al., 2014). Hence, it is vital for

planners and decision-makers that the provision of

these areas is performed maximising the impact, the

beneﬁts and the attractiveness of each of the selected

parcels (Uy and Nakagoshi, 2008) in compliance with

a sustainable urban development.

However, to fully understand the interactions of

the involved complex phenomena and be capable of

dealing with a large number of environmental and

socio-economical constraints, scientists need a bet-

ter and larger set of ecologically meaningful methods

that can be applied to spatial multi-criteria evaluation

and conservation decision making. In this regard it

can be mentioned the use of models and interactive

computer-based systems (Church, 2002). These mod-

els could both, explore and extrapolate the dynamics

of the system to infer future trends or also understand

the nature of the processes within it.

In the concrete case of green space allocation,

the existent literature covers speciﬁc range of issues

like the protection and restoration of valuable and de-

graded areas (Zucca et al., 2008), the preservation of

carbon stocks (Marinoni et al., 2009) or the deﬁni-

tion of ecological corridors (Ferretti and Pomarico,

2013) among others. In these studies, data are nor-

mally gathered by on-site surveys, a series of spatial

observations and by experts’ knowledge.

In the present proposition paper we are specially

interested in the analysis of the factors which neg-

atively inﬂuence the frequency of visit of a certain

park. In this regard, the utility of this work will be

twofold.

Firstly, as problems related to location-allocation

of resources are in nature multiobjective (Watts et al.,

2009; Nelson et al., 2009), we propose the implemen-

tation of a multiobjective planning extension of our

current urban model (Vallejo et al., 2013), focused on

these discouraging factors. The analysis of these ele-

ments in a diversiﬁed population allows us to capture

and understand the most relevant synergies and con-

ﬂicts created by the interactions with other dynamics

included in the model like demographic growth, ur-

ban extension and environmental degradation.

Secondly, we also intend to make a more active

use of population demographics and background in-

formation for grouping parks visitors according to

different interests and personal characteristics using

an Agent-Based System approach (ABS). ABS is

a technique particularly suitable for studying socio-

economic and environmental trends based on hetero-

geneous individual interactions and it complements

other equation-based techniques by means of the ex-

257

Vallejo M., Rieser V. and Corne D..

Agent-based Modelling for Green Space Allocation in Urban Areas - Factors Inﬂuencing Agent Behaviour.

DOI: 10.5220/0005284602570262

In Proceedings of the International Conference on Agents and Artiﬁcial Intelligence (ICAART-2015), pages 257-262

ISBN: 978-989-758-073-4

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

ploration of individual-levelbehaviours (Brown et al.,

2006).

In this matter, ABS has been extensively used to

study urban growth phenomena (Huang et al., 2013;

Matthews et al., 2007) where it supports the rep-

resentation of different rich individual proﬁles, nor-

mally parametrised from quantitative surveys (Robin-

son et al., 2012).

In Section 2 the description of the problem is pre-

sented. Section 3 depicts the negative factors that

could decrease the number of visits of a certain park.

Then, in Section 4, the model to be extended is brieﬂy

explained and the future methodology and modiﬁca-

tions are illustrated. Finally, Section 5 is concluded

with a short discussion.

2 DESCRIPTION OF THE

PROBLEM

Open space planning from a demand perspec-

tive (Maruani and Amit-Cohen, 2007), uses attributes

from the speciﬁc target population to ﬁnd the most ef-

ﬁcient green allocation strategy as a response to social

requirements over gardens and parks.

Distance from the park to the dwelling is com-

monly considered the major factor that inﬂuences

the visit frequency and activities undertaken in

parks (Bj¨ork et al., 2008; Woolley, 2003). The selec-

tion of this concrete feature in planning is based on

the observation that only a small percentage of users

takes any means of transport to access them (Wong,

2009). Consequently, parks at a short distance are

visited more often than large remote parks (Roovers

et al., 2002).

With the use of the spatial distribution of the green

areas and the population density among other fea-

tures, a zoning analysis can be used to create a rank-

ing of location alternatives on the basis of their over-

all attractiveness for the future new park. The process

includes the conduction of a multicriteria land suit-

ability analysis and the posterior selection of optimal

sites through an optimisation process according to a

measurable criterion.

However even if from this quantitativeperspective

the provision of green areas has been carried out fol-

lowing an efﬁcient process, this fact does not neces-

sarily imply that ﬁnal users have enough incentive to

visit them. In this matter, there is a scarce number

of studies which are focused on the factors that could

limit the use of these green areas (Bixler and Floyd,

1997; Hitchings, 2013).

Population segments divided by gender, age,

household composition and socio-economic status

differ in how they use and perceive green areas (Burke

et al., 2009; Eisler et al., 2003). This fact makes very

challenging to ﬁnd a uniﬁed policy which achieves

a complete fulﬁlment of these diversiﬁed demands.

For instance, it can be mentioned that elderly people

show lower frequency of use due to personal mobility,

health and security fears (Payne et al., 2002; Burgess

et al., 1988), meanwhile children have higher needs

of open areas for playing and social interaction when

they live in high populated dwellings (Loukaitou-

Sideris and Stieglitz, 2002; Crane et al., 2006).

Based on that premise, our main interest is the

study of some of these factors (crowdedness,size, dis-

tribution and design) in an urban growth model using

an Agent-Based System framework in order to im-

prove the understanding of the individual perceptions

that directly inﬂuence the frequency of use to these

open spaces. This knowledge can ﬁrmly contribute

to the design of more comprehensive green policies

which enhance the satisfaction of a larger number of

residents.

3 FACTORS TO STUDY

In the following section factors that could discourage

people to visit parks and their interconnections are in-

troduced.

3.1 Crowdedness

High population density in green areas is associ-

ated with various factors including population and

urban growth, a higher demand of these spaces cre-

ated by the rise of the average standard of living and

an increasingly environmental awareness on the so-

ciety (Cheshire and Sheppard, 1998; Kline, 2006;

McPherson, 2006).

Crowding perception is a subjective concept

which can be perceived when the area is highly con-

gested with heavy pedestrian and vehicular use. De-

pending on the speciﬁc user proﬁle the perception of

crowdedness may be different. For instance, fairly ex-

perienced users feel more intensively the saturation

when they compare their current visit with past expe-

riences (Ditton and Sutton, 2004; Vaske et al., 1980)

and when they use the area more frequently (Arn-

berger and Brandenburg, 2007). These local visitors

can feel the saturation as a factor which decreases

their quality of life (Lankford and Howard, 1994;

Brunt and Courtney, 1999; Williams and Lawson,

2001).

The congestion can provoke different compen-

satory measures like time and intraspatial displace-

ICAART2015-InternationalConferenceonAgentsandArtificialIntelligence

258

ment if any suitable alternative exists (Hall et al.,

2000; Shelby et al., 1988; Manning et al., 2001) and

may decrease the importance of the distance to the

park for its use (Kaczynski et al., 2008). This inter-

spatial displacement implies also extra costs in terms

of time and transportation which can be a problem for

low socioeconomic individuals who cannot afford to

move to other green areas outside the city (Heritage,

2008).

3.2 Size & Distribution

Size is an important factor to take into account when

patterns of use of green areas are analysed (Mc-

Cormack et al., 2010) considering that larger parks

are capable of supporting a wider range of activi-

ties. Consequently this factor increases their attrac-

tiveness (Broomhall, 1996). It is a common practise

to group parks into two different types: urban local

green areas selected for daily outdoor activities and

beyond build-up areas used for excursions or week-

end sports (Arnberger, 2006). Morancho (Morancho,

2003) concludes that it is better to have numerous

small green areas than a few large parks.

There are studies that connect the concept of dis-

tance with size. Pouta (Pouta and Heikkil¨a, 1998) cre-

ates a classiﬁcation relating type of green area, size

and distance. He deﬁnes the minimum size for lo-

cal parks to be between 1.5 and 3 hectare reachable

in 300 meters and outdoor recreational parks in 20-

25 hectare at 1 kilometre. The European Commission

has recommended that residential proximity to green

spaces should be limited to 300 meters with an area

of at least 5000 m

(Tarzia, 2003).

Another important aspect to consider is the topo-

logical distribution of these areas within the city. It

is rather common to ﬁnd cases where parks are non-

homogeneously distributed. Instead, they are gener-

ally concentrated over some districts which provoke

the existence of extensive areas with a lack of proper

provision.

A non-homogeneousset of green areas contributes

to the appearance of inequalities where some people

have easier access to nature areas in their local neigh-

bourhoods than others (Pickett et al., 2001). In gen-

eral underprovision and an overall lower level of veg-

etation cover are more common to ﬁnd in low income

areas (Iverson and Cook, 2000; Pham et al., 2012).

This factor is an important environmental equity is-

sue for city planners.

3.3 Design & Green Services

Design is another important element which inﬂuences

the frequency of use of green areas (Schroeder and

Daniel, 1982) and contribute to the improvement of

health and wellbeing (Floyd et al., 2008). Users travel

further distances to visit a certain green area if it has

extended characteristics and enhanced aesthetical fac-

tors (McCormack et al., 2006; Epstein et al., 2000).

Commonly size and design are also concepts linked

together due to the fact that larger parks permit the al-

location of a wider offer of services (Giles-Corti et al.,

2005).

However, as a negative factor, the level of green-

ery and physical barriers like inadequate facilities to

interact with the park (walking trails), lack of trans-

port choice, poor accessibility or unaffordable recre-

ational activities may discourage some people to use

these parks.

4 MODEL

The approach presented in this paper will be based

on an extension of a green location-allocation plan-

ning model deﬁned to work under high levels of un-

certainty (Vallejo et al., 2014). The model was suc-

cessfully applied as an exploratory tool to analyse the

potential use of Genetic Algorithm techniques in the

allocation of a certain number of green parcels in a

predeﬁned time horizon. The theoretical model fol-

lows the classical microeconomic equilibrium model

of Alonso (Alonso, 1964) within a canonical mono-

centric/policentric framework in which externalities,

in form of green areas, have been introduced.

In our framework, the landscape is represented as

a lattice of homogeneous cells (Fig. 1), each associ-

ated with a single land-cover class. The set of land

uses is divided into residential that comprises the sub-

classes available, new and old, recreational and un-

derdeveloped. Currently recreational cells are homo-

geneous in size and conﬁguration.

The model is capable of generating a city with

multiple Central Business Districts where cells tend

to be more densely populated as the number of inhab-

itants grows. These centres represent the places where

commercial and business activities are primarily con-

centrated.

In the model, urban population is represented by

agents whose interactions and endogenous economi-

cal choices are the basis for the urban infrastructure.

Agents search the maximisation of an utility func-

tion in the pursue of an economic competitive equilib-

rium for space between housing and community costs

which causes the emergence of urban patterns. Each

agent can be characterised by their age, working sta-

tus and family structure. As an additional externality

Agent-basedModellingforGreenSpaceAllocationinUrbanAreas-FactorsInfluencingAgentBehaviour

259

Figure 1: Evolution of a city with 2 Central Business Dis-

tricts.

the model includes a positive tendency to live close

to green areas. To include that feature in the model,

agents are able to pay more for this kind of dwellings.

This increases their demand and, as a consequence,

classical patterns of growth are partially perturbed.

A Genetic Algorithm approach (GA) (Holland,

1975) is a heuristic used to ﬁnd a distribution of cells

that will be transformed into parks. The selection

of green spaces is performed sequentially and it is

limited by the current conﬁguration of the system in

terms of budget and land availability. Every possible

subset of selected cells has associated a ﬁtness value

that quantiﬁes the inhabitants’ satisfaction. This satis-

faction is calculated only by the amount of green areas

located closer to them, measured in terms of distance.

4.1 Model Extension

The proposed extensions of our model can be outlined

in the following points:

• Deﬁning different agent’s proﬁles according to

their familiar structure, gender and ageing factors.

To each proﬁle it will be assigned predeﬁned mo-

tivations to visit green areas and different patterns

of use. The varied typologies will be based on

data collected from quantitative studies published

in the associated literature.

• Enriching the characterisation of green parcels to

include the aforementioned factors and their in-

terdependences. This will permit the existence of

parks larger than a single cell in size with differ-

ent designs, levels of greenery and activities to be

undertaken.

• Deﬁning the rules that allow us to decide for each

agent which parks would visit and with which fre-

quency. Taken into consideration are their loca-

tion in the grid and their patterns of use according

to their personal proﬁle. Using the resulted popu-

lation density and the size of each park, included

will be the concept of crowdedness which will de-

crease the level of satisfaction of the affected pop-

ulation.

• Creating some scenarios in which the ﬁnal spread

of parks are not homogeneouslydistributed, in or-

der to study how the population could cope with

crowded green areas or a lack of provision. The

use of scenarios is proved to be useful in the study

of the impact of multiple socio-economic dynam-

ics in an urban context (Murray-Rust et al., 2013).

In each scenario, the model will be optimised us-

ing a GA technique for a multi-objective set of

characteristics encompassing the minimisation of

economical costs and the maximisation of both

the population satisfaction and the protection of

the highest ecological valued areas.

5 CONCLUSIONS

Conﬂicting on land use could provoke a lack of provi-

sion of green services necessary for the inhabitants of

a city. The contribution of this proposition paper aims

to analyse factors which may prevent local people to

visit parks located in their vicinity. In order to do that

an extension of our current resource allocation model

will be carried out to analyse possible facilitators and

constraints that inﬂuence the frequency of use of a set

of deﬁned parks in a series of predeﬁned scenarios

within a multi-objective framework.

According to a varied set of typologies, the popu-

lation that is represented by an Agent-Based System,

will be enhanced to depict different population pro-

ﬁles according to ageing, genre, marital structure and

socio-economic factors. Each of these groups should

interact with the set of green areas in different ways.

The results and conclusions gathered will support

experts, planners and decision makers to further un-

derstand the numerous situations where people may

decrease the visit frequency of parks in liaison with

a drop in the level of service that the parks offer to

the local population. We will evaluate the level of

suitability of our approach in a targeted stakeholder

workshop.

REFERENCES

Alonso, W. (1964). Location and land use. Publications of

the Joint Center for Urban Studies.

Arnberger, A. (2006). Recreation use of urban forests: An

inter-area comparison. Urban forestry & urban green-

ing, 4(3):135–144.

ICAART2015-InternationalConferenceonAgentsandArtificialIntelligence

260

Arnberger, A. and Brandenburg, C. (2007). Past on-site ex-

perience, crowding perceptions, and use displacement

of visitor groups to a peri-urban national park. Envi-

ronmental management, 40(1):34–45.

Bixler, R. D. and Floyd, M. F. (1997). Nature is scary, dis-

gusting, and uncomfortable. Environment and behav-

ior, 29(4):443–467.

Bj¨ork, J., Albin, M., Grahn, P., Jacobsson, H., Ard¨o, J.,

Wadbro, J.,

Ostergren, P.-O., and Sk¨arb¨ack, E. (2008).

Recreational values of the natural environment in re-

lation to neighbourhood satisfaction, physical activity,

obesity and wellbeing. Journal of epidemiology and

community health, 62(4):e2–e2.

Broomhall, M. (1996). Study of the availability and envi-

ronmental quality of urban open space used for phys-

ical activity. Master of Public Health Dissertation.

Department of Public Health. University of Western

Australia, Perth.

Brown, D. G., Aspinall, R., and Bennett, D. A. (2006).

Landscape models and explanation in landscape ecol-

ogy. a space for generative landscape science? The

Professional Geographer, 58:369–382(14).

Brunt, P. and Courtney, P. (1999). Host perceptions of

sociocultural impacts. Annals of tourism Research,

26(3):493–515.

Burgess, J., Harrison, C. M., and Limb, M. (1988). People,

parks and the urban green: a study of popular mean-

ings and values for open spaces in the city. Urban

studies, 25(6):455–473.

Burke, J., O’Campo, P., Salmon, C., and Walker, R.

(2009). Pathways connecting neighborhood inﬂu-

ences and mental well-being: socioeconomic position

and gender differences. Social Science & Medicine,

68(7):1294–1304.

Cheshire, P. and Sheppard, S. (1998). Estimating the de-

mand for housing, land, and neighbourhood charac-

teristics. Oxford Bulletin of Economics and Statistics,

60(3):357–382.

Church, R. L. (2002). Geographical information systems

and location science. Computers & Operations Re-

search, 29(6):541–562.

Crane, P. R., Wyeth, S. J., Brough, M. K., and Spencer, A.

(2006). Children in inner-city suburbia: The case of

new farm, brisbane.

Ditton, R. B. and Sutton, S. G. (2004). Substitutability in

recreational ﬁshing. Human Dimensions of Wildlife,

9(2):87–102.

Eisler, A. D., Eisler, H., and Yoshida, M. (2003). Per-

ception of human ecology: cross-cultural and gender

comparisons. Journal of Environmental Psychology,

23(1):89–101.

Epstein, L. H., Paluch, R. A., Gordy, C. C., and Dorn, J.

(2000). Decreasing sedentary behaviors in treating pe-

diatric obesity. Archives of Pediatrics & Adolescent

Medicine, 154(3):220–226.

Ferretti, V. and Pomarico, S. (2013). An integrated ap-

proach for studying the land suitability for ecolog-

ical corridors through spatial multicriteria evalua-

tions. Environment, development and sustainability,

15(3):859–885.

Floyd, M. F., Spengler, J. O., Maddock, J. E., Gobster,

P. H., and Suau, L. J. (2008). Park-based physical

activity in diverse communities of two us cities: an

observational study. American Journal of Preventive

Medicine, 34(4):299–305.

Giles-Corti, B., Broomhall, M. H., Knuiman, M., Collins,

C., Douglas, K., Ng, K., Lange, A., and Donovan,

R. J. (2005). Increasing walking: How important is

distance to, attractiveness, and size of public open

space? American Journal of Preventive Medicine,

28(2, Supplement 2):169 – 176.

Glickman, D. (1999). Building cities of green. In Kollin,

C., editor, 1999 National Urban Forest of Conference.

American Forests, Washington, DC., pages 4–7.

Hall, T., Shelby, B., et al. (2000). Temporal and spatial dis-

placement: evidence from a high-use reservoir and al-

ternate sites. Journal of Leisure Research, 32(4):435–

456.

Haviland-Jones, J., Rosario, H. H., Wilson, P., and

McGuire, T. R. (2005). An environmental approach

to positive emotion: Flowers. Evolutionary Psychol-

ogy, 3:104–132.

Heritage, S. N. (2008). Health impact assessment of

greenspace a guide.

Hitchings, R. (2013). Studying the preoccupations that pre-

vent people from going into green space. Landscape

and Urban Planning, 118:98–102.

Holland, J. (1975). Adaptation in natural and artiﬁcial sys-

tems: An introductory analysis with applications to

biology, control, and artiﬁcial intelligence. The Uni-

versity of Michigan Press.

Huang, Q., Parker, D. C., Filatova, T., and Sun, S. (2013). A

review of urban residential choice models using agent-

based modeling. Environment and Planning B: Plan-

ning and Design, 40.

Iverson, L. R. and Cook, E. A. (2000). Urban forest cover of

the chicago region and its relation to household den-

sity and income. Urban Ecosystems, 4(2):105–124.

Kaczynski, A. T., Potwarka, L. R., and Saelens, B. E.

(2008). Association of park size, distance, and fea-

tures with physical activity in neighborhood parks.

American Journal of Public Health, 98(8):1451.

Kline, J. D. (2006). Public demand for preserving local

open space. Society & Natural Resources, 19(7):645–

659.

Lankford, S. V. and Howard, D. R. (1994). Developing a

tourism impact attitude scale. Annals of tourism re-

search, 21(1):121–139.

Loukaitou-Sideris, A. and Stieglitz, O. (2002). Children in

los angeles parks: a study of equity, quality and chil-

dren’s satisfaction with neighbourhood parks. Town

planning review, 73(4):467–488.

Manning, R. E., Valliere, W. A., et al. (2001). Coping in out-

door recreation: Causes and consequences of crowd-

ing and conﬂict among community residents. Journal

of Leisure Research, 33(4):410–426.

Marinoni, O., Higgins, A., Hajkowicz, S., and Collins, K.

(2009). The multiple criteria analysis tool (mcat): A

new software tool to support environmental invest-

ment decision making. Environmental Modelling &

Software, 24(2):153–164.

Agent-basedModellingforGreenSpaceAllocationinUrbanAreas-FactorsInfluencingAgentBehaviour

261

Maruani, T. and Amit-Cohen, I. (2007). Open space plan-

ning models: A review of approaches and methods.

Landscape and Urban Planning, 81(12):1 – 13.

Matthews, R. B., Gilbert, N. G., Roach, A., Polhill, J. G.,

and Gotts, N. M. (2007). Agent-based land-use mod-

els: a review of applications. Landscape Ecology,

22(10):1447–1459.

McCormack, G. R., Giles-Corti, B., Bulsara, M., and

Pikora, T. J. (2006). Correlates of distances traveled

to use recreational facilities for physical activity be-

haviors. International Journal of Behavioral Nutrition

and Physical Activity, 3(1):18.

McCormack, G. R., Rock, M., Toohey, A. M., and Hignell,

D. (2010). Characteristics of urban parks associated

with park use and physical activity: a review of quali-

tative research. Health & place, 16(4):712–726.

McPherson, E. G. (2006). Urban forestry in north america.

Renewable Resources Journal, 24(3):8–12.

Morancho, A. B. (2003). A hedonic valuation of ur-

ban green areas. Landscape and urban planning,

66(1):35–41.

Murray-Rust, D., Rieser, V., Robinson, D. T., Milii, V.,

and Rounsevell, M. (2013). Agent-based modelling of

land use dynamics and residential quality of life for fu-

ture scenarios. Environmental Modelling & Software,

46(0):75 – 89.

Nelson, E., Mendoza, G., Regetz, J., Polasky, S., Tallis, H.,

Cameron, D., Chan, K. M., Daily, G. C., Goldstein,

J., Kareiva, P. M., et al. (2009). Modeling multiple

ecosystem services, biodiversity conservation, com-

modity production, and tradeoffs at landscape scales.

Frontiers in Ecology and the Environment, 7(1):4–11.

Payne, L. L., Mowen, A. J., and Orsega-Smith, E. (2002).

An examination of park preferences and behaviors

among urban residents: the role of residential loca-

tion, race, and age. Leisure sciences, 24(2):181–198.

Pham, T.-T.-H., Apparicio, P., S´eguin, A.-M., Landry, S.,

and Gagnon, M. (2012). Spatial distribution of veg-

etation in montreal: An uneven distribution or envi-

ronmental inequity? Landscape and Urban Planning,

107(3):214–224.

Pickett, S. T., Cadenasso, M., Grove, J., Nilon, C., Pouyat,

R., Zipperer, W., and Costanza, R. (2001). Urban eco-

logical systems: linking terrestrial ecological, physi-

cal, and socioeconomic components of metropolitan

areas. Annual review of ecology and systematics,

pages 127–157.

Pouta, E. and Heikkil¨a, M. (1998). Virkistysalueiden suun-

nittelu ja hoito. Ymp¨arist¨oministeri¨o, alueidenk¨ayt¨on

osasto.

Robinson, D., Murray-Rust, D., Rieser, V., Milicic, V., and

Rounsevell, M. (2012). Modelling the impacts of

land system dynamics on human well-being: Using an

agent-based approach to cope with data limitations in

koper, slovenia. Computers, Environment and Urban

Systems. Special Issue: Geoinformatics 2010, 36(Is-

sue 2):164–176.

Roovers, P., Hermy, M., and Gulinck, H. (2002). Visitor

proﬁle, perceptions and expectations in forests from a

gradient of increasing urbanisation in central belgium.

Landscape and Urban Planning, 59(3):129–145.

Schroeder, H. W. and Daniel, T. (1982). Preferred features

of urban parks and forests. Journal of Arboriculture,

8(12):317–322.

Shelby, B., Bregenzer, N. S., Johnson, R., et al. (1988).

Displacement and product shift: empirical evidence

from oregon rivers. Journal of Leisure Research,

20(4):274–288.

Takayama, N., Korpela, K., Lee, J., Morikawa, T., Tsunet-

sugu, Y., Park, B.-J., Li, Q., Tyrv¨ainen, L., Miyazaki,

Y., and Kagawa, T. (2014). Emotional, restorative

and vitalizing effects of forest and urban environments

at four sites in japan. International Journal of Envi-

ronmental Research and Public Health, 11(7):7207–

7230.

Tarzia, V. (2003). European common indicators: towards a

local sustainability proﬁle. Ambiente Italia Research

Institute, Milano, Italy.

Uy, P. D. and Nakagoshi, N. (2008). Application of

land suitability analysis and landscape ecology to ur-

ban greenspace planning in hanoi, vietnam. Urban

Forestry & Urban Greening, 7(1):25–40.

Vallejo, M., Corne, D., and Rieser, V. (2014). Evolving op-

timal spatial allocation policies for complex and un-

certain environments. In Filipe, J. and Fred, A., ed-

itors, Agents and Artiﬁcial Intelligence, volume 449

of Communications in Computer and Information Sci-

ence, pages 351–369. Springer Berlin Heidelberg.

Vallejo, M., Corne, D. W., and Rieser, V. (2013). Evolving

urbanisation policies - using a statistical model to ac-

celerate optimisation over agent-based simulations. In

ICAART 2013 - 5th Conference on Agents and Artiﬁ-

cial Intelligence, pages 1–11.

Vaske, J. J., Donnelly, M. P., and Heberlein, T. A. (1980).

Perceptions of crowding and resource quality by early

and more recent visitors. Leisure Sciences, 3(4):367–

381.

Watts, M. E., Ball, I. R., Stewart, R. S., Klein, C. J., Wil-

son, K., Steinback, C., Lourival, R., Kircher, L., and

Possingham, H. P. (2009). Marxan with zones: soft-

ware for optimal conservation based land-and sea-

use zoning. Environmental Modelling & Software,

24(12):1513–1521.

Williams, J. and Lawson, R. (2001). Community issues and

resident opinions of tourism. Annals of Tourism Re-

search, 28(2):269–290.

Wong, K. K. (2009). Urban park visiting habits and leisure

activities of residents in hong kong, china. Managing

Leisure, 14(2):125–140.

Woolley, H. (2003). Urban open spaces. Taylor & Francis.

Zucca, A., Shariﬁ, A. M., and Fabbri, A. G. (2008). Appli-

cation of spatial multi-criteria analysis to site selec-

tion for a local park: A case study in the bergamo

province, italy. Journal of Environmental Manage-

ment, 88(4):752–769.

ICAART2015-InternationalConferenceonAgentsandArtificialIntelligence

262