1

NATURAL GAS GRID FORECAST AND URBAN PLANNING - A METROPOLITAN AREA CASE STUDY IN BRAZIL

Miguel Edgar Morales Udaeta1&2, Vanessa Meloni Massara1,

Murilo Tadeu Werneck Fagá1, Geraldo Francisco Burani1 1. IEE/US, Brazil; & 2. GEPEA/EPUSP, Brazil

Corresponding Author: udaeta@iee.usp.br

Key – Words: 1. Infrastructure; 2. Natural Gas; 3. Energy Planning; 4. Urban Development; 5. city of São Paulo (Brazil). Abstract The objective of this work is to analyze the expansion of the infrastructure of natural gas distribution, identifying priorities from large metropolis using the energy planning based on urban design tools like urban dynamics and techniques like AHP. The methodology proposed uses matrices considering the relations between the concept of urban dynamics, quality of life and the possibilities of natural gas displacing other energy forms. The matrices are made up of information about social and urban development, costs of establishing the infrastructure and projections of the consumption potential in various sectors. Relating the consumption to urban development parameters and the real estate future of the areas in study, the methodology allows indicating for each district, the viability of implementing a gas network. As conclusion, the model presents the integration between the cities profile and the natural gas use, by means of a growth natural gas on districts of São Paulo City as a specific case study. 1 Introduction

In the ANP (Brazilian National Agency of Oil, Natural Gas and Biofuels) assessment for the next year’s new natural gas fields will be incoming in Brazil. For instance, the Manati field, will duplicate de actual gas production in state of Bahia solving in the short-term the problem of natural gas unsatisfied demand. In the Espírito Santo state the Peroá-Cangoa field emerges as an opportunity of complementing the gas for the Brazilian northeast and southeast regions and to reliability the Campos Basin production in the state of Rio de Janeiro. But the Mexilhão field in the São Paulo state Santos Basin, as a prelude of the context of the pre-salt reservoirs, will be anticipating the gas production as definition of de federal Brazilian administration [1].

The most important goal of above mentioned increase in the national production ( that becomes strategic) perhaps is the energy security in the scope of gas importation from South America´s countries, like Argentina and Bolivia who’s recently passed respectively economical and political crisis. In this sense in particular in the context of energy integration, Petrobras (Brazilian state-owned oil company) was discovered in another field the beginning of 2008 in Peru and emerges as another option from the Latin-American Energy Integration.

Also in January of 2008, in the pre-salt hydrocarbons offshore Brazilian reservoirs of Santos Basin (state of São Paulo) another well was discovered. The reserve that was entitled Jupiter is a large natural deposit for natural gas and light-oil, in deep waters estimated in very large gas mega-field that, despite demanding large investments, will make Brazil double its reserves [2].

With this favorable forecast of natural gas supply and considering its various uses, this work focuses on market gas demand in sectors like residential, commercial, services and industrial. The objective is to determine the priority areas for the grid expansion through an analytical methodology that integrates the understanding of the urban dynamics to the strategies of expansion in the natural gas distribution network. Moreover the procedure is to characterizing gas consumption possibilities and attractiveness ranges. Then the methodology is developed by gathering information such as family income, demographic density and construction area, percentage of land use, number of households as well as commercial, service and industrial establishments, number of real estate as well as indicative information released by the Urban Plan of the city regarding the increments in the peripheral districts. By relating the gas consumption estimated by each type of land occupation and the cost for expanding the gas distribution network, the model will indicate, for each neighborhood, the viability of implementing a gas network as well as the places with potential for growing density in the existing gas distribution system.

In this study the AHP (analytic hierarchy process) theory is used to obtain the levels of influence of the several parameters in the expansion of the natural gas distribution grid. The method makes possible that when different factors contributes to the decision making process, the relative contribution of each one can be determined, giving all the system characteristics, including all the elements so that an alteration in one of them will be reflected in all the others. The logic based on pair combination, created by Saaty in the end of the seventies, allows the verification of relationships like as demographic density and distribution of family income, residential and industrial concentration, distance between areas that can be served and others that

2

have had served by natural gas network before, defining through the attribution of a priorities scale, the key aspects in the selection of areas that would accept the natural gas expansion.

In this paper, examples of essential information that compose the methodology are presented for six districts of São Paulo city (Brazil): Ipiranga, Tatuapé, Penha, Vila Matilde Socorro and Vila Formosa, which have different socioeconomic and geographical profiles. As conclusion, the methodology provides the expansion rankings areas, verifying the coherence of the proposal based on the urban dynamics through the analysis of highly industrialized, residential and commercial outlying neighborhoods of the São Paulo Metropolitan area, identifying the importance of urban parameters in the decision making process for natural gas network expansion in Brazilian cities. The methodology tested for São Paulo city is being generalized in a computer model, that allows its uses in other cities, indicating the possibilities of the natural gas as a real energy option in the urban energy end-uses. On the other hand, the results of this work, representing fundamentals for guidelines in the Urban Plans looking for sustainable gas infrastructure incorporation in the cities.

Considering the optimistic forecast for future natural gas supply in Brazil, this work is focused on residential, commercial, service sector and industrial uses. In this sense, it is important to emphasize that the study addresses only the distribution network. The methodology proposal in this work is based on urban indicators that define areas where gas expansion service in discussion should be implemented. The method for market evaluation proposed in this article looking at natural gas distribution networks begins with the inter-relation between urban growth and issues about supply and demand for different sources of energy. Since natural gas supply chain relies on an infrastructure yet to be installed in consolidated centers, it is necessary to include the urbanization dynamics concept in the city hall urbanization plans for its sustainable incorporation. This includes commonly used criteria such as demand and pipeline expansion cost estimates to predict which areas have a potential for natural gas consumption. The premise is based on the following background: • Natural gas is evolving in the country’s energy matrix and its main link to maturation is its distribution as a network energy good [3]; • The process of civil construction has evolved with the use of a non-destructive method of cutting trenches, the trenchless technology [4];[5]. It reduces the trouble caused by the interdiction of traffic ways to rebuild the pavement, thus making its execution cheaper, mainly in already consolidated urban areas; • The existing pipeline grid is being underused as the number of builds (residentials, commercials and service rendering) that uses the gas grid instead of electricity and LPG is still small, indicating that there is a large market to be explored, even in areas already served for a few years; • Mainly in large Brazilian metropolis, the method (through the creation of an attractivity index) intends to determine the expansion of natural gas service within the cities in harmony with other networked services in a way that the development of the cities is a sustainable goal, considering life quality and improvement of production processes. 2 Objectives of the paper

The objective of this work is to analyze the expansion of the infrastructure of natural gas distribution, identifying priorities from large metropolis using the energy planning based on urban design tools like urban dynamics and techniques like AHP. The methodology proposed uses matrices considering the relations between the concept of urban dynamics, quality of life and the possibilities of natural gas displacing other energy forms. The matrices are made up of information about social and urban development, costs of establishing the infrastructure and projections of the consumption potential in various sectors. Relating the consumption to urban development parameters and the real estate future of the areas in study, the methodology allows to indicates the viability of a gas network for each district. As conclusion, the model presents the integration between the cities profile and the natural gas use, by means of a growing natural gas use in districts of São Paulo City as a specific case study, with specific advances from the proposed methodology: • Evaluation of the joint influence of social, technical and economic parameters, in the decision-making process of investments related to private enterprises’ infrastructure. • The development of a modeling tool of natural gas’ market share prospection, with simple use and based on urban dynamic concept. • Use of the modeling tool as reference in research of natural gas distribution network expansion, involving usual aspects like costs, demand and others like, urban planning and life quality, considering both private and public approaches, in the development of the supply network. • Consolidate an indicator modeling tool as a data base, also for the implementation of other urban socio-economic essentials services. • Observate the cities director plans that provides sustainable procedures to install natural gas services, in yet consolidate centers and strenghtening non-destructive methods. The process of civil construction has evolved with the use of a non-destructive method of cutting trenches, the trenchless technology [2]. It

3

reduces the trouble caused by the interdiction of traffic ways to rebuild the pavement, thus making its execution cheaper, mainly in already consolidated urban areas; 3 Development a. Methodology The methodology based on urban indicators [6] has an objective of developing procedures that permit analysis of every consumption sectors based on the urban dynamics and orientation for the expansion of physical natural gas grid within a metropolitan area.

In the decision making process for the expansion of natural gas distribution infrastructure, different factors have to be considered in order to prioritize the assistance to areas that is expected to constitute a potential market consumer for the service in question. In this work a conjunct analysis of social, technical, economical aspects, is proposed [7]; [8] associated to the following: • The occupation process of large urban centers, based on verification of the market forecast; • Most appropriate costs and techniques for the expansion of the infrastructure for distribution natural gas pipelines.

Based on the proposed study, the creation of a model that integrates the comprehension of urban dynamics and gas distribution network expansion strategies, characterizing the possibilities of consumption in an attractiveness-based scale is suggested.

The model is developed through the organization of four information systems: life quality aspects; urban planning; demand forecast (undertaken from type of soil use stratification) and civil work characteristics. Relating social data to estimated consumption and the characteristics of network branches, the methodology allows the classification of each area that composes the city by attractiveness of gas distribution network installation. To verify the methodological coherence, a specific application is selected for the Sao Paulo gas market issue.

Therefore, the attractiveness ranking is presented for the installation of natural gas grids in the city districts selected for tests, always through the model, aiming at pointing out the possibilities of natural gas substituting other energy forms in different urban energy uses.

Considering that the selection of parameters that compose the four systems (as mentioned before) must come from the concept that natural gas infrastructure commonly gets to the cities when they are already built and that service is looking for profit as the main objective. At the same time it can improve the population’s life quality and optimize the procedures of the industrial, service rendering and commercial sectors. Three groups of interest were used as a basis for modeling. • The gas utility; • The city hall represented by its departments of infrastructure and urban planning; • The consumers (domestic and those linked to economical activities) In this work, the method entitled “Urban Information Systems” join up urban parameters that characterize the cities, determining 4 databases: • System 1 - Indicators of life quality: parameters related to the existence of social facilities (schools, hospitals and leisure) and other infrastructure networks and its reflection on the population welfare. • System 2 - Urban planning indicators: parameters related to guide plans for the cities and that collaborate with the analysis indicating concentrations by soil use, areas with capability of increasing end-use, creating more demand for energy and industrial growth, areas already supplied by natural gas network, but always with capability of increase of use. • System 3 – Indicators for natural gas consumption forecast: parameters directly related to the population concentration, domestic income (or purchasing power) and to the stratification in households and economic activities. With these parameters it is possible to forecst gas consumption only from the number of units without considering size or sector, this including sampling survey for determining the conversion volumes to natural gas from other energy like electricity, oil derivatives and LPG • System 4 – Civil construction indicator; parameters not directly related to construction costs and to the problems derived from roads closure for installation of underground pipes, like the distance between the served area and areas to serve, road extensions, traffic incidence and densities constructed by type of soil use that indicate major or minor pipe branchings. The systemic modeling of the involved factors has been elaborated through the following stages: • Identification, characterization and systematization of the main factors interfering the concept of urban dynamic (like described in the introduction). • Definition of the study according to the availability of parameters information. • From the unit of study, the information is stored according to the details needed by the user of the model, working towards larger cells until the city area is reached (municipality, district, neighborhood, block and units) creating a list that must be associated to the city zones (the cardinal points). The chosen study cell will be a standard. The mixture of different geographic scales is not recommended. Once the list of cells for study has been created, this list automatically repeats itself in the next step, which corresponds to fill in the database.

4

• Systematization of the parameters: consists in filling data in the four information systems along with public organizations, building up a database. • Hierarchization of quantitative and qualitative parameters through attribution of a priority scale that unifies the dimensions of all indicators evaluated in the study, so that these can be treated mathematically.

Decision-making in a complex environment normally involves multiple criteria, imprecise information and/or incomplete, multiple agents of decision, etc. To support this process there will be a field of operational research named Multi-criteria support to decision, which was developed in the seventies. The AHP (Analytic Hierarchy Process) is one of the most used multi-criteria methods to support decision-making and resolution of negotiated conflicts, in multi-criteria problems [9]. The chosen hierarchization is the same used by the AHP [11]; [10], which allows the verification between algorithms of this method and that one of the proposal modeling (in this work), according to the association presented in table 1.

Group Semantic scale for natural gas AHP scale

1 Low attractivity to network installation 1 2 Low to medium attractivity to the installation of the network 3 3 Medium attractivity to the installation of the network 5 4 Medium to high attractivity to the installation of the network 7 5 High attractivity to the installation of the network 9

Table 1. Adaptation of AHP scale to natural gas study Source: Massara, 2007 based on Saaty, 2006.

Before the conversion to AHP scale, which goes from 1 to 9, it is necessary to put all parameters in the same unit. To do that, the model must automatically transform the parameters into percentage. Following that, the collected values are organized into 5 ranges, where this choice is based on the primary information obtained from SEADE [12], IBGE [13] and city hall [14]. That primary information generally use, for the division of tables and maps five well know groupings. This facilitates the association of the semantic scale for used for natural gas network analysis to the numerical values obtained from official sources.

The acronyms (Seade and Ibge) refer to State bureaus, whose studies contain the required information for this work available in the format of tables and maps. Data from Ibge are in a national level, whereas Seade provides urban information of cities in São Paulo State. ‘São Paulo’ refers to various departments within the São Paulo city hall and includes urban information of São Paulo’s capital. The calculation of the 5 numerical intervals for each parameter is elaborated based on the numerical scale. It is worth mentioning that each factor demands a single calculation to determine its maximum and minimum values (in order to compose the five ranges), such as: • Application of the algorithm for the attractiveness index calculation: The mathematical formula for determining the index comprises the mean of the simple sum of all weights attributed to each parameter by study unit that is based on the algorithm of official indexes of the São Paulo city hall [14]; [15]. • Attribution of the ordinal scale to study units according to the value calculated from the indexes, in decreasing order, in other words, the greatest mean indicates first place and so on; Elaboration of the attractiveness ranking (general or using information systems): corresponds to a classificatory ranking of the units of study as function of the obtained value in the calculation of the attractiveness index. • Validation of the decision making support model for planning of the natural gas grid expansion through case studies in São Paulo state municipals and Brazilian (in this case São Paulo’s capital) and comparison between the obtained results by the model and mapping of the already installed network. • Validation of the results through comparison between the obtained ranking by the model and by the Analytic Hierarchy Process [11]. For the construction of this model the basic premises were verified: • Verification of extreme values that are out of range with the collected values (remember that the data when not specimen is already tested statistically by the competent bureaus) Division of each parameter in n categories, where n is equal to 5, since these parameters are commonly offered by the official bodies (above mentioned) in 5 classes; • Linearization of all parameters to the same unit (percentage), permitting conversion to the AHP scale. • When necessary the information collection on the parameters by sampling, use the sampling concept by convenience, selecting the research units by burden while energy consuming and dispensing parametric tests and non parametric. b. Analysis of Natural gas forecast Modelling and the Urban Dynamics

The concept of urban dynamics is a starting point to execute the model. This concept is based on the continuous “movement” and “transformation” of urban space, according to [16], whose studies addresses various scopes including spatial distribution of productive areas, population concentrations, infrastructure and income conditions, as well as distances to the most developed centers [15]. The concept of urban

5

dynamics described by [16] is wide and can be adapted to various studies. The approach of urban dynamics in this work uses the study of factors related to the concentration of activities and the constitution of an urban center as a function of the economical advantages of agglomeration of the activities inside the only build, constituting a greenfield area to attract ameliorations and new activities [14]. According to the possibilities of attraction (and repulsion) to the natural gas network, generated by the competitive advantages inserted in the urban dynamic concept [6]; [14], the factors that interfere with the installation of urban infrastructures are many, from the possibilities of consumption from family income and sophistications in soil usage to the costs incurred by the distance to serve, as well as the concentration of buildings, industries, hotels and future intensification projects on occupation of areas that were previously represented by horizontal residential use.

According to [17], intervenient factors in urban infrastructure implementation are many, since the consumption possibilities by family income and sophistication of land use, until the costs of transport, building concentration, industries, hotels and projects that may intensify the occupation, previously represented by horizontal residential use.

Another determining point is the sequence of services deployment, from those known as essential, like sanitation, lightening and electrical energy, that play important role in the attraction or repulsion, to other services in a network, which are usually implanted when the others are already widespread.

The location of districts [18] is an important agent of change in land uses [19] and plays important role in the attraction of urban services network. The urban infrastructure includes a set of crucial elements for the structuring and development of cities. As closer to the served center, greater is the likelihood demand for infrastructure networks and sophisticated equipments, generally determining the social status [20], translated in this study as social exclusion index [21] and human development index [22]; [15], which as well as essential networks, plays a fundamental role in "social capability" to receive and consume natural gas. The availability of these networks is an essential factor to measure urban population life quality related to living and working conditions. The presence or absence of this services infrastructure has direct impacts on the processes of social inclusion or exclusion [14].

The distribution of wealth rather than population density, is an attraction factor for the deployment of sophisticated networks and applications, particularly for the consumption of energy [23].

Similarly to the technical facilities and establishment costs, represented by the distances to serve both in the extensions of distribution as in the ramifications within the districts. As [4], beginning from the core, there is an implicit idea that the installation of networks becomes simpler and cheaper in already served areas, what justifies the faster way the neighborhoods boarding the core have received the services.

An infrastructure author [25] achieves a complete study of the network deployment, including the distances to serve different types of land uses and, consequently the built densities as key points in the adoption of areas to receive the service. Also evaluates as prevailing factors in the decision-making, those related to investments in civil construction and the volume that can be transported in tubes, based on consumption estimative in areas with industrial predominance, (hence the introduction of the zoning concept), showing that the population has little effect on the decision-making, because there is a minimum limit of diameters and such amendment is not allowed, even just to pass small quantities of gas to serve little communities. The author highlights in the same work that like all other goods and services for community needs, the effective demand for services is linked to the effective wealth of the population.

Projects aiming the urban development of boarding areas [15], common to all Brazilian cities, with the zoning can be considered as an intensifying tool for the low density areas occupation, where there are voids or low built density. The zoning with its current function, and urban planning, with their future role, may be actors to increase staff in the natural gas use.

The concentration of economic activities as a projection method of consumption described by [26], considers in addition to detailed volume, the simple projection based on the number of outlets and type of activity, and was adapted to the model proposed in this thesis, according to the National Register of Economic Activities (Ibge) through stratification by land use and its productive sectors. The example of detailing hospitals types in [27] also provides examples of stratification of the services sector used in other areas, aiming to determine on-site the possibilities of natural gas consumption.

Since the intensification of areas vertically built and connected to gas natural network, will be factor for appreciation (with gains for construction) and while increase consumption of GN (with gains to the utility). The buildings with internal tubing for natural gas are a new trend in the construction sector. According to constructors and architects, the demand for these buildings grows every year, and prior planning of the tubing has become, more than one facility, a market need. In addition, according to professionals, the planning of natural gas tubing even in the plant, decreases construction spending, avoids waste of material, and provides beauty and security to work [28].

In localities where there is a law ensuring the gas installation for hot water, such as the city of São Paulo, is a clear economic advantage of using gas as energy for heating water [29]. Based on these statements, as a influencing component in the natural gas expansion model is the parameter of number of construction enterprises.

6

Although the MND is a method that has as one of the goals, to reduce the inconvenience of banning road traffic, the distribution still grows. According to Comgas [30], the MND is applied in 95% of cases of polyethylene ducts and 85% when steel ducts are employed. GASMIG applies approximately 20% of its constructions, and intends to expand the use of the method for 80% of total. Although, the method is not completely widespread, it is considered to be a component that reflects the area mapping (and the consequent ban on the traffic) the number of major traffic routes.

As the last component included in the modeling is the urbanization rate [22], which explains itself the very concept of dynamic urban areas, emphasizing areas with percentage of urban concentration higher than the rural side. Below, the components (parameters) of each of the four information systems that compose the proposed model are described.

c. Information system 1 – Quality of life indicators (QLI)

The economic and social city life of is closely linked to the achievement of better standards of urban services. According to [31], with the assessment of the urban infrastructure networks can be defined the criteria of real life conditions of living people in urban areas.

The extent, availability and quality of basic urban services such as water, sanitation, electricity, waste collection and rainwater collection, indicate the habitability of living in the city or metropolis. Other items of urban infrastructure, such as pavement, public transport, telephone, gas network, and so on, configure important and necessary additions to the development of urban pattern.

This item parameters are related in association with life quality of cities that indirectly interfere in the attraction of the gas system deployment, such as deficits in basic infrastructure - water supply, sewage collection, public lighting - and factors that determine longevity, better social situation and schooling. As it seems, in areas with concentrations of residential uses, commercial and service, its use will be not attractive if there are still deficits in services, equipment and basic networks. This information system (table 2) is represented by numeric values, distributed in calculated intervals according to its dimension, expressed by the three factors described below.

Parameters Acronym Unit Characteristic Function

Human development index HDI dimensionless Finite between 0 e 1,0

Good indices attract the natural gas network (not a priority)

Index of social exclusion ISE dimensionless Finite between -1,0 e +1,0

Good indices attract the natural gas network (not a priority)

Attending by water supply network

AWS % Finite between 0 e 100%

Proceed to the natural gas network

Attending by sewage collection network ASC % Finite between

0 e 100% Proceed to the natural gas network

Attending by public illumination network API % Finite between

0 e 100% Proceed to the natural gas network

Table 2. Information system 1 – Quality of Life Indicators (QLI) Source: Massara, 2007.

Human Development Index (HDI): It is an adaptation of the index created by the UN (United Nations) and aims at comparing the level of human development among the districts. It considers three components; longevity, literacy (adults) and life pattern (income production adjusted to local life cost). It includes variables like level of education, basic health conditions and social facilities per district [32]. These indicators are transformed in indices that, when summed up, compose the HDI, in an interval that varies from 0 (zone 1, the worst developmental condition) to 1 (the best condition).

In Brazil, the institution responsible for calculating the HDI for all Federation units including more than 5,500 counties is the IPEA - Instituto de Pesquisa Econômica Aplicada (Institute for Applied Economic Research) which is linked to the Planning, Budget and Management Ministry. However, there is no systematic calculation of this index for smaller units than the counties, a practice that would be very useful when working with big towns and all the diverse mosaic of economic and social situations found in them [12]. In 2002 was confected the calculation of HDI for the districts of the city of São Paulo.

Index of social exclusion (ISE): The aim of this index is to identifying the degree of social development of the districts, considering variables associated to the life quality [12]. The ultimate index attributed to each one of the districts was calculated from the sum of the indices in each field, creating a scale that has an interval between -1.00 (minus one) reflecting the worst situation of exclusion, range 1, and +1.00 (plus one) reflecting the best situation of exclusion, range 5. Unlike the HDI has two other acquisitions [21]: • First, compares rates from the same cultural context. So, there is no comparing between countries where some situations may be different due to cultural habits and customs alimentation, clothing, housing etc.. Thus, we can say that the differences refer to the same context; • Second, install a measurement form in decimal notes that are more intelligible to the common sense skillful with decimal use and to attribute good grades to good situations and bad grades in bad situations. Thus, the

7

map gives the condition of negative grades to refer to exclusion and positive to relate the inclusion. This condition adds a symbolic dimension of what is needed or not in habitability. Percentage of attending by priority networks (AWS, ASC, API): “Priority” includes water supply networks, sewage collection and public illumination. This parameter corresponds to the idea that a given district will not be attractive to natural gas network if it still does not possess such infrastructure. According to the IBGE definitions: • Percentage of households with water network: percentage of households or plots served with water from the general supply network; • Percentage of households with sewage network: percentage of households in which the pipes for serverd water and waste from the bathroom or toilet, are connected to a collect system that leads to a general spillway area, even if the system does not have waste water treatment; • Percentage of public lighting system installed: a set of equipment that allows full operation of the lighting of the city, such as transformers, arms, lights, auxiliary equipment, wiring and bulbs and therefore the number of lightened points in a region.

The index is calculate from the average of the costumer service conditions of the three networks above mentioned, expressed in five range percentages with intervals of 20% [12]; [32].

The EMPLASA (S.A. São Paulo Company of Metropolitan Planning), is a company to provide information about the metropolitan region of São Paulo, which consists, the capital as well as 38 other smaller cities and the region called “ABC” (municipalities of Santo André, São Bernardo do Campo e São Caetano do Sul). The “ABC” is composed by the three most important industrial cities in São Paulo State, with the greatest concentration of Brazilian automobile industries.

d. Information System 2 –Urban plan indicators (UPI)

The parameters listed in this group are from the basic contents that all Urban Plans should have. The “Director Plan” is a Municipal Law which organizes the city organization and growth. It determines the goals to be achieved in each area of the city, and identifies urban tools and strategic actions to be implemented. Guides investment priorities, i.e. indicates structural constructions to be build [33]. Defined as a central technical and legal management of urban space, which defines the major urban guidelines. Traditionally, these guidelines include standards for density, territorial expansion, definition of areas of land use and network infrastructure. It is also a global and strategic policy development, critical for all public and private actors operating in the city. To assess the expansion potential of natural gas whereas the "urban dynamics" were extracted from [14] are defined in (table 3).

These indicators are developed in a database composed of qualitative and quantitative parameters. For the analysis of non numeric values, the stratification into zones was based on the mapping and classification of the managing urban plan for São Paulo city, verifying the current kind-of-uses and its perspectives of expansion, with the intention of elaborating a profile of the neighborhoods with a great tendency of industrial agglomeration (large attractiveness to the use of natural gas), followed by the other uses, according to the projection scale for consumption, described below.

Parameters Acronym Unit Characteristic Function

Land Use (Residential, commercial, services rendered, industrial)

Lures LUcom LUserv LUind

% Finite between 0 e 100%

Characterize areas with the greatest potential of natural gas consumption

Zoning Z dimensionless Qualitative Characterize areas with expansion of the potential of long-term natural gas consumption (emphasis on soil use)

Urban development UD dimensionless Qualitative Characterize areas with expansion of the potential of long-term natural gas consumption (emphasis on infrastructure)

Real estate launching (residential and services)

REres REserv number Not finite

Characterize areas with expansion of the potential of long-term natural gas consumption (emphasis on soil use)

Urbanization rate UR % Finite between 0 e 100%

Differentiate areas where the services can be or cannot be networked.

Table 3. Information system 2 – Urban Planning Indicators (UBI) Source: Massara, 2007.

Land use (LUres, LUcom, LUserv, LUind): Nature of occupation in the district, by the soil use map [14], considering the largest percentage of blocks with each kind of use. In general, the maps of land use are developed through a methodology that considers percentage with a predominant type of occupation by 60% or more of the blocks that compose the area under study. These maps function is to demonstrate city areas with intense use, so where there is greater density of built area per square meter, its type (horizontal and

8

vertical) of occupation, the standard (low, medium and high ) of buildings and type of use (residential, industrial, trade and services etc.), that allow different approaches and relationships to support analysis and diagnosis in global or regional policies development or. The blocks where is not possible the identification of a particular predominant type of use, applies the mixed use category, which represents the combination of two or more predominant uses, whereas in the model the percentage of blocks divided and inserted in such number as are in use.

Although the city hall provides various categories of occupation, for the effect of projection of attractiveness to natural gas consumption a grouping of 5 main uses was elaborated such that:

Range 1: predominance in horizontal residential occupation; Range 2: predominance in mixed use (commercial and horizontal residential); Range 3: predominance in vertical residential occupation; Range 4: predominance in mixed use (commercial, services rendered and vertical residential); Range 5: predominance in mixed use (residential and industrial).

Urban development (UD): according to the guiding plan of the municipality of São Paulo [15], the importance zones were attributed with reference to the projection of attractivity to the use of natural gas in relation to 5 descriptions of “macro areas” that correspond to the specifications regarding current urban development and the future of São Paulo districts. Such five descriptions are:

Range 1: protection of the environment – limits of public areas and preservation areas; Range 2: urbanization and urban qualification – areas occupied predominantly by low income population, with large concentration of irregular land divisions and slums; Range 3: restructuring and requalification – areas with good infrastructure, but experience processes of population exhaustion and buildings evacuation; Range 4: urbanization in consolidation – areas with conditions to attract private real-estate investments in residential and commercial and service establishments; Range 5: consolidated urbanization – areas formed by consolidated neighborhoods inhabited by medium and high income population and with good conditions for urbanization.

Zoning: According to [15], the zoning is a concept of managing urban space based on the idea of electing the possible uses for certain areas in the city. Therefore, the aim is to avoid unpleasant coexistence between uses. “From the expression in portuguese “Zoneamento”, is a rule imposed by the city hall that limits the use of soil, that is, the destination of the various “parts” of the city for specific use (commercial, services, housing, industries). These categories are different, usually referring to blocks density (rules for maximum percentage of land area that can be built, number of building floors or built area).

Through consultation of the city hall maps for São Paulo [14], the predominance of these rules on soil occupation has been verified, using the greatest percentage of blocks with each type of zoning. Five zones were created for attractivity to the implantation of the (natural gas) network, grouped as shown below:

Range 1: exclusive residential zone of low density / environmental protection zone; Range 2: exclusive residential zone of medium density / mixed low density zone; Range 3: exclusive residential zone of high density / mixed medium density zone / special uses (airports, campus); Range 4: mixed high density zone; Range 5: large industrial occupation zone. In this item, the definition of “mixed zone” corresponds to the combination of residential, commercial

and service rendering uses.

Real-estate Launchings in residential and in the service sector (REres, REserv): related to the exigency of the construction code in São Paulo for to foresee gas facilities in new constructions, what increases the natural gas consumption. For the attribution to the gas grid attractivity weight, the grouping was elaborated in five zones, having in mind the greatest and the least number of launchings per district, and also considering the survey elaborated by the EMBRAESP (Brazilian Company of Patrimonial Studies) [34]. Embraesp is an acronym that refers to a private company that provides information about development of the real estate market in the same geographic area of Emplasa, in other words, the São Paulo metropolitan region, that will be presented in figure 1.

Urbanization rate (UR): percentage of the district’s total area occupied for urban use (residential used for commerce and industries), in relation to the total population [14], represented in 20% intervals. e. Information system 3 – Natural gas use potential indicators (NGPI) In general, the energy demand projection is always considered a prospecting model of markets, aiming to deploy the service network. Here also, the proposed model to estimate the consumption of NG uses the definition of conversion other energy into natural gas, but also uses urban dynamics as an aid to a faster projection without

9

considering "packets", defining the possible consumption according to households number, shops, services and industrial facilities, what will be classified in productive sectors according to the National Register of Economic Activities [13]. Thus, first of all, economic activity is aggregated into total of establishments. Then it explores the sector divisions of economy, allowing consumption estimative according to a particular concentration of different productive areas in the region of study, without considering size and consequently each energy consumption. The options for detailing the land uses and estimating the NG demand are- method 1: the volume, which requires sampling on site or method 2: by units number based on the division degree of the productive sector (depending on the precision level required and time that for the collecting information).

Associated with the characterization of "local units", a Ibge concept [13] defined as the address of the company's performance which is, generally, a continuous area, where are developed one or more economic activities, identified by serial number (suffix) the entry in the National Register of Legal Entities, the Ministry of Finance, are aggregated in this item parameters like demographic density and family income, based on relationships: consumption versus wealth and consumption versus population. The first considers the relative social status (purchasing power or income) as a determinant of energy consumption, the second relationship, considers the agglomeration as demand determinant, this time not as a possibility than the social position offers but as a necessity, although without the discussing the political nature between outskirts and income distribution.

In this system there are stored data from Ibge database [13]. That data is about the occupation of the neighborhoods combined to the gas utility information (table 4). The parameters are all numeric, making it possible to create the five zones through the simple division of the obtained values, listed below.

Parameters Acronym Unit Characteristic Function

Demographic Density DD habitants/km2 Not finite Define NG demand (emphasis on the population concentration)

Family Income (or power to buy) FI R$ or minimum salary Not finite

Define NG demand (emphasis on the financial possibility of consumption)

Stratification (residential, commercial, services, industrial)

Sres, Scom, Sserv, Sind Dimensionless Not finite Define NG demand (emphasis of

consumption in the sector) Table 4 - Information system 3 – Indicators for the potential of natural gas (NG) use

Source: Massara, 2007.

Stratification by consumption sectors (Sres, Scom, Sserv, Sind): Consists in dividing the neighborhood in dwellings and in diverse economic activities according to the national registration [13], grouped in consumption sectors based on the counting of the number of establishments more likely to utilize natural gas, like hotels, hospitals, companies, supermarkets, ceramic, chemical and metallurgic industries, among others, in the unit being studied. Figure 1 shows the division in five detailing zones based on the stratification of the economic activities defined by [13].

Figure 1. Stratification for the natural gas use forecast.

Own elaboration: based on IBGE, 2003.

10

This method combines information on energy consumption and urban planning. In the last levels, 4 and 5 (see figure 1) the models requires a sample characterization of industries, commercial, services and residential edifications, for the detailed gathering of physical data about the buildings; including the use of other energy sources and the number of equipment that can be changed to natural gas end-use. However, the model expects indirect consumption based on the characteristics of the soil use and occupancy expressed by the hypothesis that, the larger the number of dwellings, commercial/service units and industries/facilities located in the unit being studied, the larger the attractivity of implantation of natural gas network. In this case, it is suggested that the user gives an external weight that is multiplied by the value already existent in function of the number of units, to differentiate the branches of activity that are more likely to utilize natural gas or that for other reasons are more attractive to the utility. It is proposed that the utilization of the model always initiate at level 5 that utilizes in loco verification of each unit, according to the samples. In this case, the inclusion of these characteristics makes a “looping” within the model, in other words, it repeats itself as much times as there are units put in, always relating to the level of anterior stratification (level 3 for residential and level 4 for other types of use). In case there are no sufficient information the initial levels also determine the attractiveness index but, with less precision, as it does not work directly with estimation of energy consumption.

Therefore, the user of the model must always start by the level of field survey, working with information on energy consumption, equipment and characteristics of the sector and verify if it has the data in question. In case of a negative response, the user must proceed to the next level where fewer details on the data are required and so forth until level one is reached where informing the area of study is sufficient, like: • The number of industrial installations; • The number of commercial establishments and service rendering; • The number of dwellings.

Demographic density (DD): corresponds to the quotient between the number of people residing in the area and the district area [12], having in mind that a large population concentration generates demand for energy. Family income (FI): corresponds to the mean of the minimum salaries in the district’s dwellings [35], considering the influence of the relation, “income versus energy consumption”.

In the same way as for the other parameters that have been described, all the weights referent to each are summed up indicating the “Attractiveness Index” of natural gas within the area of study, according to land use.

f. Information system 4 – Civil construction indicators (gas grid implantation) (CWI) Where as civil construction technological development and the government role in infrastructure construction in already established urban areas [14], which is facilitating the networks deployment in consolidated cities, inserted in this item there are parameters involving the distance to be covered within the study area and between it and the last served point, and their importance in connecting other districts and municipalities, requiring greater agility in the construction process. Numerical values to the distances were not involved, whereas coarse errors in the construction detailing can be committed and that the mere representation of distances represents the costs of the construction. Another important point of the proposed model in this work is the inclusion of the parameter "built density by land use" that indicates the house-holds concentration and manufacturing establishments, considering its size as built area. Then it’s possible an interpretation of vertical areas or areas with predominance of large facilities with economic activity, which indirectly reflects the need for smaller or larger tube branching, based on the premise that the greater the built concentration of residential area, the larger is the network branching needed to connect buildings. And on the contrary, the greater the industry built area, the lighter is the need for network branching in general, resulting on a faster and cheaper implementation. The function of the information system that contains the indicators for civil work is to represent the value associated with the cost of the civil construction, like the extension of the gas pipeline and its diameter, as well as indicating the guidelines on the interdiction of important roads in the district (table 5).

11

Parameters Acronym Unit Characteristics Function Distance from point to serve to the last point already served D km Not finite Determinant of the network’s

physical expansion Total extension of the roads to serve in the area being analyzed

E km Not finite Determinant of the network’s physical expansion

Incidence of busy roads T dimensionless Not finite Determinant of the network’s physical expansion

Constructed density (residential, commercial, services, industrial)

DCres DCcom DCserv DCind

m2/km2 Not finite Determinant of the network’s physical expansion

Table 5. Information System 4 – Civil construction indicators (gas grid implantation) (CWI) Source: Massara, 2007.

Like in the previous system, the five zones are elaborated by a simple division of the obtained

values, expressed by four factors like below are showed.

Extension of the distribution area (D): distance between the district to serve and the covered area by the local utility. This concept is based on simple analysis of an already served area and another isolated one, distant from the core in, as if it were only a "trunk-line" without any intermediate branch between the two regions.

Extension of branches (E): sum of the internal ways in the study area to be served. Represents the extent of the ways within the district under study, with a smaller diameter pipelines, or distribution network, considering the contour curves and topographical differences. To obtain this information is no longer needed the utility map, that can be extracted from any scale map or city entries maps with greater accuracy.

Streets with important traffict (T): even with the evolution of the construction process, and being that the interdiction of the same block limited to two days, this parameter indicates the importance of the district as a link between neighborhoods and other municipalities and the special attention that must be given to the interdiction plan. The number of ways is obtained through simple consultation of the street mappings, taking into account the avenues and streets with great extensions in the district.

It is worth mentioning that in these three parameters, the zones are considered in an inverted manner, in other words, the smaller the distance, the larger the weight. Therefore, the districts located in the partially served area by piping natural gas receive index 9, expressing the best attractivity condition.

Density constructed (DCres, DCcom, DCserv, DCind): it is obtained by the quotient between the constructed area by type of soil use (residential, commercial, services and industrial) and the area of the district, provided through the Registration of Real Estates of the City Hall [14]. In districts with large industrial constructed density, smaller investment with branches is presumed (capillarity of the distribution network).

This parameter considers that the concentration of large industrial establishments minimizes the branching expense, therefore, the higher the density built in a industrial district, the lower are branching expenses.

According to the description in the methodology, all the values gathered are unified to the same unit (percentage) and divided in five zones of attractiveness that are converted in the hierarchisation scale of 1 to 9. After this, the weights are attributed they are submitted to the calculation of “Attractiveness index”, whose algorithm is based on the simple sum of the weights for each area of study as summarized in table 6.

Attractivity Index by Information Systems

QLI= (HDI+ISE+AWS+ASC)/n

UPI= (LUres+LUcom+LUserv+LUind+Z+UD+REres+REserv+UR)/n

NGPI= (DD+FI+Sres+Scom+Sserv+Sind)/n

CWI= (D+E+T+DCres+DCcom+DCserv+DCind)/n

General Attractivity Index

IG= (QLI+UPI+NGPI+CWI)/n

Table 6. Summary of the algorithm for the Attractiveness Index Source: Massara, 2007.

Note: “n” represents the number of parameters effectively used.

12

4 Results Considering that the infrastructure for the natural gas reaches the cities in general when those are

already settled and highly occupied. The use of a method that generates a minor possible disturbs in terms of time of physical construction, interdiction area and noise pollution, has a more relevant role on the city dynamics. This sustainability concept of network implementation, that introduces the non-destructive technology instead of excavation, considers tree main groups of interest. • The distribution service utility of the city, that aims to install the network with the best technique; • The City Council, represented by its construction department, infrastructure and urban planning , that manages the insertion of the network around the urban area; • The population that will use the installed service with the least problems during the construction.

The city urban development policy represented by its Director Plan also can influence of how the service choice can be amplified. The option to break the pavement in settled urban areas, besides its costly price, requires an interdiction planning on the road network. Such a choice can impact the local traffic, generating a problem for the local community that should be avoided. By the other side, if the option of the network expansion is underground, the use of “tatuzinhos” to excavate the ground minimizes the interdiction problems and re-pavement of the road. However, the sharing of the underground with other infrastructures and the lack of detailed information of its exactly location, the excavation must be monitored to avoid intersection which those already implemented infrastructures.

The Brazilian of Nodig Technology Association [36], affiliated with the International Society for Trenchless Technology [4] is developing the divulgation of methods trough out the seminar “A city without holes”, that gathers the experience of different underground city services (water, waste, telephone, natural gas, electricity) and cities councils and the results of optimization of implementation in urban areas with the use of Nodig construction.

The process of civil construction has evolved with the use of a non-destructive method of cutting trenches, the trenchless technology [4]. It reduces the trouble caused by the interdiction of traffic ways to rebuild the pavement, thus making its execution cheaper, mainly in already consolidated urban areas;

The so-called non-destructive methods (trenchless technology) uses special machines that perforate the subsoil horizontally, between two access wells, through which the piping are passed. In this way, it is not necessary to tear away the entire floor extension under which the installation will be passed. This method is extremely useful when crossing big traffic areas, since the transit of vehicles is not influenced by the construction. In Brazil, its relevance is destined to the execution of services in polyethylene tubulations and steel for work until 2 meters deep that can be: distribution of oil derivatives and gas; crossing of avenues, highways, rivers and railways; underground drainage systems; industrial installations; pipe substitution, etc. The direct cost, in many cases, is already equivalent to the method with continuous cutting of trenches, but the advantages are larger: Accuracy in work execution; deadline reductions; no traffic interruption in the work area and large reductions of social costs.

At the metropolitan area of São Paulo, the directional method (HDD) is done by the use of a main hole at each 160m on the average with “cachimbo” (an square opening of 2x2 meters) for the union of the tubes. The company installs the plumbing network based on the main network, where it is installed the “city gates”, where its done the transference of the Petrobas gas network to the Comgas gas network. From that point, its added smell to the gas (for security purpose) and is introduced to the primary network, that has a higher pressure so it can attend the industrial district, and the secondary network, to supply the residential demand [30].

The open void process is used when it does not have insurance of where it is the other pre-installed infrastructure. Comgas has started the use of NDM since 1996, when it was acquired the necessary machinery and it uses its own personnel. In 1999, Comgas has aquired a 12t machinery and became a national reference.

Since the implementation and expansion of new areas are done at the same time, it was necessary the renovation of the steel networked, implemented between the 20’s and the 50’s decades. Those networks were used to supply the gas-based illumination system and nowadays supply natural gas to residential and commercial use of the central urban area of Sao Paulo (Pari, Brás, Jardins and Campos Elíseos districts). The implementation of this service was done by the use of two non-destructive method used by Comgas. One of then was the insertion of smaller diameter polyethylene tube inside the old tube. The second one was used when it was not possible to insert the new tube, so it is contructed a new line, parallel to the old one, using the NDM [36].

As the network renewal projects are located in urbanized areas and it is necessary to restore the supply - temporarily stopped - within 12 hours, it runs short stretches (100 m) with equipment capacity varying between 3 and 12 t. The work speed is more closely linked to the number of consumers in the area, than the length of the sentence to be renewed and the average is 4 to 7 km / month [30].

It´s important to remember that the MND has restricted use: as in underground unknown interference, in diameters above 14" and in hard ground and rock construction, the same as the PEAD which is still not widespread for pressures above 4 bar. The underground tubing shares the cities subsoil with various networks of water, sewer, electricity, telephone, natural gas, cable TV, Internet and others sectors

13

that were allocated in different periods of cities development in Brazil. As the growing construction, constant reviews of already installed infrastructure are necessary, and also, the expansion of services that is not yet fully widespread.

Considering the cities dynamics, the use of non-destructive method acts as a diffuser and optimizer of natural gas network expansion, by facilitating its introduction in everyday sustainable urban uses.

As suggested in the recommendations about natural gas policy [37], the choice of São Paulo (capital of São Paulo State) as an all-gas city, is to demonstrate the complete utilization of natural gas in all its applications. In the context of this work, the choice has background in: • The possibility of increasing the density of use of the current grids: the increment of the factor of utilization of the infrastructure implanted in the expanded center, besides the extending the consumption market can leverage the financing of the network expansion. • As mentioned by [38, p. 162]: “In the São Paulo municipality, along the roads at the margins of rivers Tietê and Pinheiros, the greatest commercial areas in the country are concentrated here, with various shopping centers and large office buildings. All of them are located less than 2 km from the high pressure Comgas (gas utility) pipe-ring, but rarely use it”. This situation will be the same in the next years, according to the map [39], with indication of the free zone areas expansion in the use of natural gas in the mentioned regions, however, with great possibility of increasing its use.

The map of the gas utility [39] shows the natural gas grid distribution within the city of São Paulo, in other words, which districts are already served by the network, permitting the selection of areas not served for testing the model proposed in this work. The referred gas utility had the concession service of piping natural gas in three important metropolitan regions in the state of São Paulo: Metropolitan region of São Paulo, São José dos Campos and Baixada Santista, what makes up the most developed concession urban area in the state.

That possibility of natural gas expansion coming from the complexity of expansion of the natural gas grid infrastructure into the districts in the periphery of already consolidated and developed urban areas, it is widely known that the amplification of the pipeline mesh is onerous and probably not justifiable in the extreme periphery of the city, however in the nearest districts of the area already being served, it seems, in terms of superficial analysis, to be executable, considering both the distance of extension of tubulations and the aspects of demand and the population’s income. The point of greatest conflict is in the digging of trenches in busy and wholly urbanized ways, as well as, the lack of underground mapping indicating the position of other grid infrastructures like electricity, water and sewage, fibre-optic and telephone, as confirmed by [40, p. 98]: ” The lack of infrastructure to make natural gas available to the clients is one of the drawbacks to the great growth of natural gas participation in the energy matrix in São Paulo (the state). Within the city, the situation is a bit more complicated, as it involves CONVIAS (São Paulo Utility of Highways) and the Regionals Administrations, among other entities”.

Figure 2 shows the city of São Paulo within the metropolitan region of São Paulo. And in this case it is worth to mention that the whole area is a cement forest.

Figure 2. City of São Paulo (ash area) inserted in the Metropolitan Region.

Source: SEADE, 2006. (note: no scale figure)

14

There is no doubt that the service sector is the most important in the capital city but that is still not able to express like any participation in cities in developed countries, which are responsible for two thirds of the economical activities [35].

According to the reports, about Urban Dynamics of the Municipal [14], the industrial establishments present a strong concentration in the central region of the city, mainly due to the presence of a large number of textile industry establishments. Another important aspect is the greatest density of establishments along large road axes of the municipal. The distribution of commercial establishments is more diffuse. This explains itself by the relation between the presence of this sector and the population’s concentration. Once again, downtown presents the largest number of establishments in the city. In the eastern part of the city, the presence of the establishments is also strong and is more intense immediately besides the central region and scarce in the extremes.

Regarding the consumer service, outside downtown and its peripheries, there is low density. Considering the wide and extremely heterogeneous activities, both in what is referred to as economic functions and to the size of the companies. Regarding the constructed density, especially in the industrial use, there was a decrease in the southeastern neighborhoods (Marginal Pinheiros road axes), where old industrial warehouses were used for other uses or substituted by residential buildings or provide services buildings. This reduction of the constructed industrial area reflects the real estate valorization process that makes the sophistication of soil usage more profitable.

On the other hand, outside the expanded downtown, mainly in the periphery in the process of urbanization of rural zones, present in the north, south and west, there is difficulty in the gathering of information. According to São Paulo [35] in the same report, TCPL (Territorial and Building Registration of Conservation and Cleaning) there is partial covering in districts of extreme regions, which are those districts that present the biggest deficits in various factors of urban dynamics. It is clear that the expanded downtown concentrates more income, more sophistication on soil usage and also, an already consolidated natural gas network. As a form of limitation of space within the city of São Paulo, a division into 96 districts, according to the information provided by the Municipality Secretary of Urban Planning [15], will be used. Figure 3 shows the 6 districts selected for the case-study in different zones of São Paulo (capital city) with its various characteristics. Some of them already have a natural gas network in a scarce manner (For example Tatuapé and Ipiranga). However, all of them can be considered to be in the expansion stage since none of them is completely served.

Figure 3. The 6 elected districts in São Paulo municipality (capital city).

Source: SEADE, 2006. (note: no scale figure)

Legend: N1 - District of Ipiranga ; N2 - District of Tatuapé; N3 - District of Penha; N4 - District of Vila Matilde; N5 - District of Socorro; N6 - District of Vila Formosa

In order to demonstrate the use of the model developed in this work, for instance, one application is presented across the table 7.

Therefore, in this example, the following simplifications were considered:

15

• Calculation of the attractiveness index, using the “General” grouping (without the division in information systems) and its respective ranking; • Study area: districts; • Natural gas consumption forecast (information system 3): using the stratification level 1 for all the uses (residential, commercial, services and industrial); • Extra weight to emphasize the segments more likely to use natural gas is = 1.0 for all the uses. • Algorithm application according to the sums (table 6).

Selected Districts of São Paulo Municipality System Parameters

Ipiranga Tatuapé Penha Vila Matilde Socorro Vila Formosa

ISE 7 9 7 7 9 9 HDI 3 5 5 5 7 7

AWS 9 9 7 9 9 9 ASC 9 9 9 9 1 9

1

API 9 9 9 5 9 7 LU res 3 5 5 7 3 9 LU com 5 7 7 5 3 3 LU serv 3 9 9 5 1 1 LU ind 5 1 1 1 7 1

Z 9 7 7 5 5 7 UD 9 5 5 5 9 9 UR 9 9 9 9 9 9

RE res 5 9 5 1 1 3

2

RE serv 3 9 9 1 5 1 DD 3 3 5 5 1 5 FI 3 5 1 1 3 3

S res 5 3 5 3 1 3 S com 9 7 9 3 9 5 S serv 5 7 9 5 5 5

3

S ind 9 5 5 1 9 3 D 9 9 3 5 9 3 E 5 7 1 5 9 5 T 1 1 1 3 1 3

DC res 5 9 7 7 3 9 DC com 9 3 7 5 3 9 DC serv 9 3 7 5 3 9

4

DC ind 1 5 3 1 9 1 Table 7. Attribution of weights according to the four information systems

Source Massara 2007 based on Seade, 2006; São Paulo, 2006.

The evaluation option is the general ranking aiming to simplify the demonstration of this case study. Table 8 presents the attractivity ranking for the expansion of the natural gas grid obtained with de application of the prototype and using the AHP methodology [11]; [9] both of them developed in [6]. For methodological basis validation, the result was compared to the current condition of the natural gas network as well as considering the expansion plan of the gas utility, which that comparison has obtained a satisfactory outcome. The result of the application of the method to the 6 districts is demonstrated on table 8.

District General index AHP index (%)

General Ranking

AHP ranking

Ipiranga 6,1 18,9% 2º 2º Tatuapé 6,3 22,2% 1º 1º Penha 5,9 16,8% 3º 3º Vila Matilde 4,7 10% 6º 6º Socorro 5,4 15,6% 5º 5º Vila Formosa 5,6 16,5% 4º 4º

Table 8. Results for the case-study – The general attractiveness index and its respective ranking Source: Massara, 2007 based on Saaty, 2006.

The rounding off to the first decimal place in some districts can be evaluated better, taking into

consideration “best location” through the identification of the cardinal points of each study area, according to the direction to be followed in terms of the network’s expansion (gas utility’s decision), or by the analysis of the partial rankings that the model generate divided into the four information systems.

The main objective of the model is to work as a decision making auxiliary tool for gas tubing network implementation in a city. The model offers different attractive rankings, which one with different focus, so that

16

the user (according to table 6), either a company or the city council, can chose its priority (i.e. high consuming industrial areas, areas with high residential demand). However, since that the same cell can appear in different positions in each rank, the model will say the official rank of the cell trough the organization of the information that has more influence into the investment return. • 1º. Consider the district highest consume demand district. (general sum SI3) • 2º. Consider the district with the highest civil construction level. (general sum SI4) • 3º. Consider the district with the highest urban planning level (general sum SI2) • 4º. Consider the district with the highest life quality level (general sum SI1) • 5º. Consider the most proximity city zones trough the cardinal points (see the predominance zones on the first places) • 6º. Consider the higher scale weight considered to Area Use – Industrial Predominance (within of SI2 select the study cells that has scale weight scale 5) • 7º. Consider the higher scale weight considered to Urban Zoning (within of SI2 select the study cell that has scale weight scale 5) • 8º. Consider the highest scale weight to Real state Launching (within of SI2 select the study cell that has scale weight scale 5) • 9º. Consider the highest scale weight to Demographic Density (within of SI3 select the study cell that has scale weight scale 5) • 10º Consider the highest scale weight to Familiar Incoming (within of SI3 select the study cell that has scale weight scale 5)

The analysis under urban dynamic focus shows that the parameters choice connected to City profile can be adapted to the Market Research of the Natural Gas Plumbing Network, translated the possibilities of consumption in an Urban Dynamic based scenario. This shows that, the better is the characteristics of the primary urban infrastructure, zoning and area uses, more attractive the study cells (in this case, the district) the better the zone for natural gas plumbing network investment.

The first units in the ranking of the model, are districts with the largest development in the real-estate market, largest leeway of zoning. These districts are nearest to the areas already served and, of course, have the largest forecast of natural gas consumption, mainly, in the vertical residential use (with highlight for the following districts: Ipiranga, Tatuapé and Penha). The rest of the districts concentrate the least income, horizontal residential use and a small commercial, services and industrial concentration (Vila Formosa and Socorro). Apart from the characteristics of the intermediate group, the last classification (Vila Matilde) is in an area with environmental limitations to expansion and, therefore, of small urban development. The result is also validated when compared with the real expansion of the service [39], which indicates the Tatuapé as a district with expanding network, followed by the Districts of Ipiranga and Penha and less emphasis, in the districts of Vila Formosa and Socorro. Then, the final placement the Vila Matilde, with punctual assistance, demonstrates the ranking order obtained by using the model.

5 Conclusions and Perspectives

First of all it is possible to conclude that the methodology makes it possible to detect from internal differences to “global” tendencies of the city through the synthesis of variable parameters. The possibility approach of the variables in different territorial units (also promotes diverse studies of the city, the larger the aggregation, the larger homogenization occurs, hiding the “local” realities). In this way, the territorial unit of analysis used must be in accordance with the objectives of the research, in other words, on the level of the details desired and that is possible to achieve. Therefore, it is interesting to use smaller study areas whenever it is possible and to increase the precision of the rankings. The comparison between the ranking using AHP [10], and the model ranking [6] has satisfactory results (presented in table 8).

The verification of each study area (in this case the district), must always de related to analogous data collected about it, based on the Euclidian distance of the already-served area. This observation is of great importance to the districts of the periphery that can show in the numeric study inadequacy to the network’s expansion, but which in the joint analysis of the neighboring districts, with better attractivity index, can come to offer new consumption approaches at short-term.

The model based on urban indicators has presented coherent results when tested in cities of the state of São Paulo, proving to be a calculation tool of good level of precision, easy to comprehend and to use as adjunct in the process of decision making about the expansion and densification of natural gas grids in the Brazilian cities.

The creation of a residential and commercial market based on the gradual introduction of natural gas in the habitual uses can end up inducing more consumers for this less pollutant energy source. Them with a proven incremental natural gas uses that can permit the introduction of a more sophisticate grid infrastructure for natural gas distribution in large Brazilian cities.

However, in order to have greater potential of expansion of the natural gas grid service, it is necessary that there exists strengthening of the relation between the utilities, network engineers of natural gas pipelines facilities and all the civil construction related to the urban plans. In these case it is good looking

17

for the development of equipment for the diverse purposes that can utilize natural gas, including the formal assisting from the public entities in the implementation of infrastructure and in advertising of natural gas use as an agent for urban development as well.

In this way, the model intends to perform an auxiliary role in this processes of increasing de gas market, identifying the possibilities of using piping natural gas both in neighborhoods near large urban centers of significant real estate development and largest income concentration and in periphery areas with the same intensity and promptness that electricity is currently used in the near future, thereby generating new jobs linked to civil construction both in the ambit of insertion of underground pipelines and in adapting the internal grid facilities of the buildings, in the industrial maturity and national development & research and, finally, better quality of life for the population, regardless of its social condition. Acknowledgment

To the ANP (National Agency of Oil, Natural Gas and Biofuels), FINEP (Studies and Research Financier) and of the MCT (Ministry of Science and Technology) through the PRH-ANP/04 (Program of Human Resources of ANP number 4) and to FAPESP (The State of São Paulo Research Foundation) through the research project Process 03/06441-7 (“New Instruments for Regional Energy Planning seeking Sustainable Development”) that permitted the utilization of the Decision Lens Program. References [1] Banco Nacional de Desenvolvimento Econômico e Social. Setor de Petróleo e Gás: Perfil dos

Investimentos. Brasília, BNDES. 2006. [2] Gas Brasil, 2008. Newsletter Semanal [line]. Disponible: http://www.gasbrasil.com.br, 2008. [3] Udaeta, M. E. M., et al. Brazil natural gas market as a factor for energy integration in Latin América. In:

International World Energy System Conference - WESC: proceedings. Tokyo, 2002. p. 244-249. [4] International society for trenchless technology. Trenchless techniques [on line]. Disponible

http://www.istt.com/, 2007. [5] Najafi, M. Trenchless technology: pipeline and utility design, construction, and renewal. McGraw-Hill, New

York, 2005. [6] Massara, V.M., A Dinâmica Urbana na Otimização da Infra-Estrutura para o Gás Natural. São Paulo.

Tese (Doutorado em Energia). Programa Interunidades de Pós Graduação em Energia da Universidade de São Paulo, 2007.

[7] Udaeta, M.E.M., Planejamento Integrado de Recursos Energéticos - PIR para o Setor Elétrico (Pensando o Desenvolvimento Sustentável). São Paulo. Tese (Doutorado em Engenharia). Escola Politécnica da Universidade de São Paulo, 1997.

[8] Udaeta, M.E.M; Grimoni, J.A.B.; Galvão, L.C.R Iniciação a Conceitos de Sistemas Energéticos para o Desenvolvimento Limpo. São Paulo, EDUSP, 2004.

[9] Cicone Jr, D.; Udaeta, M.E.M.;Grimoni, J.A.B., Galvão, L.C.R. Functionality of the approach of hierarchical analysis in the full cost accounting in the IRP of a metropolitan airport. Energy Policy. , v.36, 2008. p.991 – 998.

[10] Saaty, T. Decision Lens. Londres. Programa de Computador, 2006. [11] Saaty T. The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation. Londres,

McGraw-Hill, 1980. [12] Fundação Sistema Estadual de Análise de Dados. Pesquisa Municipal Unificada. São Paulo, SEADE,

2006. [13] Instituto Brasileiro de Geografia e Estatística.Cadastro Nacional das Atividades Econômicas (CNAE).

Rio de Janeiro, IBGE Cd-rom, 2003. [14] São Paulo (Cidade) – Secretaria Municipal de Planejamento Urbano. Dinâmica Urbana. São Paulo,

2006. [15] São Paulo (Cidade) – Secretaria Municipal de Planejamento Urbano. Plano Diretor estratégico do

Município de São Paulo 2002-2012. São Paulo. Cd-rom, 2002. [16] Forrester, J. W. Urban Dynamics. Cambridge, M.I.T. Press., 1969. [17]Massara, V.M. O perfil da infra-estrutura no Município de São Paulo e sua relação com as

transformações de uso do solo: o centro expandido e a região de São Miguel Paulista. São Paulo, Dissertação (Mestrado), Escola Politécnica da Universidade de São Paulo, 2002.

[18] Alonso, W. Location and land use. Cambridge, Harvard University Press, 1964. [19] Villaça, F. J.M. Espaço intra-urbano no Brasil. São Paulo, Studio Nobel, 1998. [20] Santos, M. O Espaço do Cidadão. São Paulo, Nobel, 1987. [21] Sposati, A. Mapa da Exclusão/Inclusão Social da cidade de São Paulo: dinâmica social dos anos 90.

São Paulo, CDRom, 2000. [22] Instituto Brasileiro de Geografia e Estatística. Censo Demográfico 2000. Características da população

e dos domicílios – Resultados do Universo – Notas Metodológicas. Rio de Janeiro, IBGE, 2001.

18

[23] Vetter, D. M., Massena, R.M. R., Rodrigues, E.F. Espaço, valor da terra e equidade dos investimentos em infra-estrutura urbana: uma análise do Município do Rio de Janeiro, 1982.

[24] Rolnik, R.. A Cidade e a Lei: legislação, política urbana e territórios na Cidade de São Paulo. São Paulo, Studio Nobel Fapesp, 1997.

[25] Mascaro, J. L.. Custos de infra-estrutura: um ponto de partida para o desenho econômico urbano. São Paulo, Tese (Livre-Docência), Faculdade de Arquitetura e Urbanismo da Universidade de São Paulo, 1979.

[26] Williams, N. The structure of urban zoning and its dynamics in urban planning and development. New York: Buttenheim Pub. Corp., 1966.

[27] Tomasquim, M. et al. Mercado de gás natural na indústria química e no setor hospitalar do Brasil. Rio de Janeiro, Edições COPPE / UFRJ, 2003.

[28] Gas Natural São Paulo Sul. Ajinomoto economiza mais de 10% substituindo óleo combustível por GN. In Revista Gas Net, junho, 2006.

[29] Fossa, A.J. Pierrobon Neto, E.; Chaguri, J.J. Substituição da energia elétrica pelo gás natural nas maiores demandas de uso residencial e comercial. In: II Encontro Brasileiro dos Profissionais do Mercado do Gás, 2000.p.1-7.

[30] Meier. E.P. Memória do ciclo de infra-estrutura urbana na área central de São Paulo. São Paulo, Associação Viva o Centro, 1997.

[31] Ministério das Cidades. Plano Diretor Participativo – Guia para a Elaboração pelos Municípios e Cidadãos. Brasília, 2006.

[32] Empresa Metropolitana de Planejamento. O perfil institucional da Região Metropolitana de São Paulo. São Paulo, EMPLASA, 2004.

[33] Ministério das Cidades. Plano Diretor Participativo – Guia para a Elaboração pelos Municípios e Cidadãos. Brasília, 2006.

[34] Empresa Brasileira de Estudos de Patrimônio. Relatório anual sobre número de lançamentos imobiliários por região. São Paulo, EMBRAESP, 2006.

[35] São Paulo (cidade) – Secretaria Municipal de Planejamento Urbano. O uso do solo segundo o cadastro territorial e predial. São Paulo, SEMPLA, 2002.

[36] Associação Brasileira de Tecnologia Não Destrutiva. Tecnologia. Endereço de Acesso: http://www.abratt.com.br, 2007.

[37] Instituto de Engenharia de São Paulo. Recomendações sobre a política para o gás natural. São Paulo, IE, 2004.

[38] Moutinho dos Santos E.. Fagá, M.T.W., Villanueva, L.D., Zamalloa, G. Gás natural: estratégias para uma nova energia no Brasil. Editora Anamblume, FAPESP, PETROBRAS, 2002.

[39] Companhia de Gas de São Paulo, Mapa da área Coberta pela Rede. São Paulo, COMGÁS, 2007. [40] Moraes, E. G. O mercado de gás natural no Estado de São Paulo: histórico, cenário, perspectivas e

identificação de barreiras. Dissertação (Mestrado em Energia). Programa Interunidades de Pós Graduação em Energia da Universidade de São Paulo, 2003.