Lessons learnt from and sustainability assessment of Indonesian urban kampong

The search of good (sustainable) model of development has been evolved from the era of City Beautiful to current era of uncertainty. Many urban planner and designers practice western modern concept and design, which are not entirely suitable for developing countries. It has been identified that local forms and characteristics in the urban context have wisdom and content that sustainable for local condition. This paper supports the argument that local characteristics have the qualities of sustainable development by evaluating sustainable development indicators and promoting method in generating aggregate indicator of sustainable development qualities of urban kampong in Indonesia. Fuzzy logic approach, which is widely used in system control design, is promoted in formulating the aggregate indicator of sustainable kampong in Indonesia. The result show that Indonesian kampongs are good in compactness, density, and access. Other indicators will also show sustainable development quality if the lacks are improved.


Introduction
Kampong (Indonesia/Melayu) is often associated with slum within urban areas, similar to favelas in Brazil or ghettos in western countries. A kampong, therefore, is perceived to carry negative aspects and treated as a slum. Kampongs became places for improvement, upgrading, and renewal or even demolition [1] [2]. Kampong Improvement Program (KIP) in Indonesia, launched in 1969, was proclaimed by UN-Habitat as the world's first slum upgrading program. Urban designers and planners have not been inspired by kampong and its forms in creating new residential areas, even for low income communities. Imported ideas from western and other 'modern' developments are more popular to them. Positive aspects of kampongs are overridden by the negative perception that kampong is slum.
There are two issues related to urban kampong in Indonesia. Firstly, social and physical characters of urban kampong convey sustainable development principles. Secondly, approaches to measure level of sustainability of urban kampong through an aggregate indicator. Descriptions of Indonesian kampong's characters can counterweigh negative perceptions of the 'Indonesian slums', several authors identify and explore positive aspects of kampong [3] [4]. These authors claimed that Indonesian kampongs are successful in managing aspects of sustainable settlement, i.e. vitality and livability. The authors focused the discussion on social characters of kampong, the so called rukun (social harmony) and gotong royong (mutual cooperation) in creating public places. Rahmi et.al., [4] underlined that Indonesian kampongs possess unique characters that cannot be described into other terms such as slum, squatter, marginal or low income settlement. She continuously argues that kampongs best describe Jane

Methods
Major Two steps analysis were utilized in this research. The first step was qualitative descriptive analyses. Qualitative descriptive analysis was used to measure quality of sustainable characters of urban kampong in Klojen District. Six variables were employed in this step: Compactness, accessibility/permeability, diversity, density, poverty, and community participation. The first four represent physical/environmental qualities, the fifth represents economic capabilities and the last represents social capital.

Environmetal or Physical aspects
Dispersion index or compactness index [17] was used to measure the efficiency in urban utility provision, the scale is ranging from 0 to 1, the closer to 1 the more compact. When the shape is simple (one patch), calculation of compactness index for 2D shapes may use the following formulas: Where Ci = compactness index Di = diameter of a circle with the same area of area i Di' = the longest distance between two points of area i Ai = the area of i [18] Or CILP = compactness index of largest patch s and p = area and perimeter of largest patch [19] 3 1234567890 3rd To measure connectivity, alpha index is a node-linkage association which measures the ratio of number of circuits relative to maximum number possible in a network [20]. The alpha index (α) is calculated as follows: = − + 1 2 − 5 where e = number of edges and v = number of nodes. The α values range from 0 to 1.
Diversity in urban landscape is often measured by the variety of land uses. Recent authors stated that mixed land use, which enabled pedestrians to walk to many destinations of diverse land uses in close proximity, is more desired than segregated land uses which are implicated in health and transportation problems [21]. Diversity of land use is often measured by entropy index (EI).
where EI is Entropy Index; K is area of land; N is number of land uses.
Building density, as the fourth indicator of physical settlement quality was simply measured by division of numbers of building by area. The density was clasified into four: slum (more than 100 buildings/Ha; high density (80-100 buildings/Ha); medium density (30-<80 buildings/Ha); low density (less than 30 buildings/Ha). The classification is based on SNI 1733-2004 (Indonesia National Standard) of urban settlement environmental planning mechanism.

Economic capabilites
Poverty and community participation measures were utilized to measure social aspects in the study area. Three basic poverty measures were employed in the research: Poverty head count index (Po), poverty gap index (P1) and poverty severity index (P2). General formula to measure poverty is= where : Pa = Poverty head count index (if a=0), or Poverty gap index (if a=1), or Poverty severity index (if a=2); N is total number of population; q is total number of poor population who are living at or below poverty line; z is poverty line; and yi is the income of the poor individual i [22] .

Social capital
Social capital comprises of trust, norms, and networks, and in Indonesian case, at the village / kampong level, it is related to gender empowerment [23]. In this research, women were prioritized in the sampling process to evaluate the affiliation network as part of social capital [24]. Social sustainabiliy of urban settlement can be measured from community participation in the community's institution. Rate of participation (RoP) is obtained from sum of diagonal value of adjacency matrix devided by number of institutions.  The second step was to formulate a composite index of sustainability of the selected kampong. In this step we inputed discrete indicators obtained from step one. Eventhough the indicators were not so comprehensive but, at least, they represented three legs of sustainable development i.e. environmental, economic, and social aspects. In formulating a composite index from a set of indicators, fuzzy logic approach, using Matlab-Fuzzy Logic Toolbox, was utilized in the research. The framework is described in figure 1. Expert judgement was used in the 'if -then' fuzzy rules. Assigned normalization for compactness, α, entropy, poverty, and affiliation network indices were classified into 0 -0.33, 0.34 -0.66, 0.67 -1 as low, medium, and high respectively. Particular for poverty, in contrast to other indicators, high means bad. Density was classified into < 30 houses/Ha, 30 -80 houses/Ha, and >80 houses/Ha as low, medium, and high respectively. Normalization for final aggregate index classified level of sustainability into five: 0.81-1 (high); 0.61-0.8 (medium high); 0.41-0.6 (medium); 0.21-4 (medium low); and 0-0.2 (low sustainability).

Results
The analysis resulted in two type of indicators: discrete indicators of urban kampong which shows each indicator of physical entities found in the three kampong and composite indicators as a result of combination of discrete indicators into single composite indicator. The evaluation of discrete indicators of these three kampongs are described in table 1.

Composite Indicators
To formulate a composite indicator, inference system based on fuzzy rule was utilized. Fuzzy rules use 'If-then' languages. Based on figure 1, three steps of inference processes were conducted. The first inference system was to formulate a composite indicator of physical aspect. The second one was for a A) composite indicator of social-economic aspect, and the last was for an aggregate sustainability indicator of urban kampong.
The inference system for physical aspects was based on 81 permutations of if-then rules (table 2) based on formula nr = 34 where n is the number of objects that can be selected and r is the amount that must be selected. The inference of 'then' from the result of combination of four indicators (independent variables) was based on Delphi analysis. Defuzzyfying of inputed data resulted in a composite indicator (crisp number) of physical aspects of the form of the three kampongs (table 4).
Composite indicators of social-economic aspects were generated with the same process. Fuzzy rules of social economic aspects consist of 9 possibilities (n r =3 2 ) (table 3).  The last process was to formulate an aggregate indicators from the two composite indicators above. The inference system or the rule base to define the level of sustainability was based from Delphi method obtained from key experts in urban development. As in social economic indicators, eight possibilities (  The determination of high, medium-high, medium, medium -low, and low is based on the set membership. The equation of levels of high sustainability is based on the increasing line of the curve, while for low sustainability is on the opposite way. and the membership of medium-high, medium, and medium-low is based on the equation Defuzzifying of two composite indicators, physical and social-economic indicators, into an aggregate indicator of sustainable kampong indicates that Arab kampong is more sustainable than the other two kampongs (table 6).

Conclusion
Most of discrete indicators of sustainability in the three kampongs in Malang were good and fair. Bad indicators were found at the entropy of Pecinan and Kebalen Kampongs, while bad indicators of poverty and participation level found in Kebalen Kampong. The evaluation of discrete indicators shows that kampongs are good examples of sustainable settlements in the form of their compactness, density, and access. The composite indicators of their environment were scored in the range of 0.44 to 0.52, while the composite indicators of social-economic aspects were in the range 0.445 to 0.740. Finally, the aggregate of all indicators were in the range from medium to medium -high. From the results, we promote that this mechanism of assessment can be improved to involve more comprehensive indicators. The good characters of Indonesian urban kampong must be explored further more to provide models for the new local post-modernism planning in the era of new urbanism. These characters of sustainable kampong can be expanded into practices at larger scale in urban context. Of course, more research must be accomplished in determining the 'acceptable zone' of compactness, density, entropy, access and other indicators of sustainable / livable place for all scale of urban development from micro, mezo, to macro settlement.