VMT and Neighborhood Design: Decades of Evidence from Iskandar Malaysia

Transcription

1 VMT and Neighborhood Design: Decades of Evidence from Iskandar Malaysia Alia N. Nordin 1 M. Rafee Majid 2 and Foziah Johar 3 1 Masters Student, Department of Urban & Regional Planning, UTM, Malaysia, alianurashikin@gmail.com 2 Associate Professor, Department of Urban & Regional Planning, UTM, Malaysia, rafee@utm.my 3 Associate Professor, Department of Urban & Regional Planning, UTM, Malaysia, b-foziah@utm.my ABSTRACT: Close to sixty eight percents of the population of Malaysia is currently living in urban areas, doubling its value of thirty three percents in By the year 2020, when Malaysia is expected to be an industrialised nation, about two thirds of the population will be in urban areas living in mass-produced housing schemes first constructed in the late 1960s. These concentrated mass housing schemes, however, were not accompanied by a solid public transportation program, leaving Malaysian roads congested with private vehicles. Mileages travelled by these private vehicles are very much influenced by the design of the housing areas especially for internal travels within the areas. Nonetheless, little is known about how neighborhood designs in Malaysia influence household vehiclemile-travelled or VMT. Understanding the effects of neighborhood design towards VMT may help reduce internal travels, thus reducing transportation carbon emission. This paper presents the findings of a study carried out in Iskandar Malaysia on the influence of neighborhood design attributes density, diversity, proximity, connectivity, etc. on household VMT. Thirty residential neighborhoods representing several decades from 1970s to 2000s were selected and travel diaries of their randomly selected households were recorded. Results obtained confirm the prevalent theory on the relationship between neighborhood design attributes and VMT. On average, daily travels range from 40 km/day for neighborhoods built pre-1980s to about 100 km/day for those built in the 2000s. Preference for curvilinear design rather than the higher connectivity grid design has increased household VMT. Some of the neighborhoods are also bigger with lower density and, unfortunately, less diversity of land uses. High density neighborhoods are normally small ones exclusively zoned for housing, forcing the residents to travel beyond the neighborhoods for amenities and facilities. As Iskandar Malaysia is targeting an increase in population from the current 1.5 million to 3.0 million by 2025, serious rethinking of its neighborhood development is a must to fulfill its commitment on becoming a low carbon city. Key words: Vehicle Miles Travelled (VMT), Sustainable Neighborhood, Neighborhood Design 1. Introduction Transport is a vital link that brings people and goods together. People rely on it to get to work and for educational, social and recreational activities. Between 1985 and 1997, the modal share of public transport decreased from 34.3 percent to 19.7 percent (Kuala Lumpur Structure Plan 2020). Rapid development and urbanisation due to economic and population growth in Malaysia has been associated with increasing number of private-ownership vehicles (POV). Inadequate public transport and services worsen the situation resulting in modal split of private transportation with public transportation to a ratio of 80:20. Transportation, together with electricity and heat generation; manufacturing and construction; and residential have been identified as four main sectors that contributed to the greenhouse gas emission in Malaysia. By the year 2020, when Malaysia plans to achieve industrialised-nation status, about twothirds of the population will be in urban areas living in mass-produced housing schemes first constructed in the 1970s. Relationships between Vehicle Miles Travelled (VMT) and urban form characteristic as well as neighbourhood design has been well explored [1] [2] [3]. However, research on

2 VMT in Malaysia was barely heard off. Back in 2003 and 2007, it was used once as an indicator to measure road safety performance and as an indicator for international traffic safety comparison purposes. The parameter provides additional information about crash risk in terms of exposure for certain groups of road users. For instance, drivers who travel across long distances are more at risk to be involved in an accident than those who make shorter trips [4]. From only 8,442,208 POVs in 1998, Malaysia has recently more than doubled its POVs to 19,144,262 in 2011 (Malaysian Road and Transport Department, 2011). Among the units in 2011, Johor State with a population of 3.5 millions holds the second largest POV rate among the states in Malaysia with a total of 2,649,657 units. This POV rate is set to further increase as the population of Johor is estimated to increase to 5 millions in 2025, with the biggest share of the increment will take place in the Iskandar Malaysia region. Higher POV will increase travel trips which lead to higher VMT, thus resulting in traffic congestion and increased air pollution. Therefore, it is the main objective of this study to investigate how changing designs of neighbourhood in Iskandar Malaysia over the decades have influenced household VMT. The changing designs are described in terms of quantifiable parameters, including neighbourhood size, diversity index, proximity index (distance-location from houses to commercial areas), connectivity index and intersection density. The effects of neighbourhood design on VMT are an important aspect to consider in our effort to reduce the greenhouse-gas emissions. 2. Residential Development in Johor Bahru Urban sprawl hits many cities in Malaysia, including Johor Bahru [5]. Bordering Singapore to the south, residential development has been constantly happening in Johor Bahru since 1980 s (Table 1). Neighbourhoods of various sizes are being developed each year with each new one moves further away from the Johor Bahru city centre, encroaching into the rural area. This almost always results in higher dependency on automobile as the preferred mode of travelling. With the rapid pace of current development happening in Iskandar Malaysia, more new neighbourhood developments are expected to come. Recent trend, however, has seen more and more exclusively-zoned developments offering only residential and recreational land uses, which are not entirely a good idea if the city is trying to reduce household VMT. Table 1: Development of Neighbourhood in Johor Bahru Area over the years. Year No. of Neighbourhood Individual Size (acres) Pre 1980 s 7 From 35 to 251 acres 1980 s 99 From 4 to 2000 acres 1990 s 96 From 4 to 4800 acres 2000 s 100 From 3 to 3200 acres 3. Neighbourhood Design and VMT Cervero and Kockelman [6] on their travel research has identified built environment characteristic called three Ds that influence travel demand and behavior. They are density, diversity and design. Meanwhile, Ewing & Cervero [1] and Ewing et al. [7] next identified another two Ds which are destination accessibility and distance to transit. According to Ewing doubling urban densities results in a 25-30% reduction in VMT or a slightly smaller reduction when the effects of other variables are controlled." By other means, doubling urban density refer to compact neighbourhood. Compact neighbourhoods can degenerate vehicle trips and encourage non-motorized travel in several ways. By bringing origins and destinations closer together, the opportunity for a resident to leave their car at home and walking or cycling to a destination will increase [6]. In addition, compact neighborhoods tend to have fewer parking, better quality transit services, wider mixes of land uses, and larger shares of low-income households, all factors that reduce car usage. In Johor, high-density development can reach a maximum of 60 units per acre (for apartment). However, high demand on landed houses, which particularly comprise of 8 to 40 units per acres, makes compact development difficult to be developed. Higher density is expected to reduce the need to travel across long distances, thus lower the VMT. However, argument against this hypothesis by some scholar had made the literature on this hypothesis became ambiguous. Steiner [8], in her review of literature on the link between residential density and travel distances conclude that resident of high density areas will travel shorter compared to resident of lower density areas. Differ with Steiner, Levinson and Kumar [9], on the basis of their study across cities in the US with more than 1 million inhabitants, state the opposite. In their paper, they suggest that metropolitan residential density is principally a surrogate for city size. Large settlements offer more Page 2

3 services and facilities (high diversity), which could reduce travel distances and favour the use of slower transport modes and public transport. On the other hand, the dispersion of urban land use over a large area may lead to longer distances, which lead residents to use their cars. In addition, research conducted by Banister [10] showed that average trip distances were shorter in larger settlement than in smaller settlement. This is because, inadequate or few public transport facilities require the resident to use their own car to get to the services and facilities that were also not provided in a smaller settlement. Early simulation study by Kulash et al., [11] and McNally and Ryan [12] found that a connected or grid street design reduces VMT as well as slower average travel speeds. A connected street pattern reduces the distance of a trip [13]. In response to that, Cervero and Kockelman [6] in their study added up that good design, especially street pattern design and diversity would encourage people to commute by non-motorized modes. Their research result suggests, a neighbourhood with all rectangular or square blocks could be expected to average nine more daily personal VMT per household than one with no quadrilateral blocks." This is because rectangular or quadrilateral blocks (grid-like street pattern) will encourage walking since, the interconnectivity between destination is higher. New Urbanism and transit-oriented development are some of urban design philosophies that have gained popularity in recent years as ways of shaping travel demand and lower the VMT. These urban design philosophies cover all the three Ds discussed above and shared three common transportation objectives, which are to reduce the number of motorized trips, to increase non motorized trip (by foot or bicycle) and to reduce travel distances and increase vehicle occupancy levels (i.e. encourage shorter trips and more travel by transit, and ride-sharing) [6]. The main idea of the new urbanism concept is to place and mix residential, open spaces, shop, and commercial area not only within a neighbourhood, but also located within walking distance. Housing projects in Johor Bahru has already adopted the concept of New Urbanism in term of its mixed used development [15]. Theoretically, this is believed to help shaping travel demand and low VMT value shall be generated. However, Hussein [14] observes several aspects of New Urbanism that has been neglected by developer purposely or otherwise in developing a neighbourhood in Johor Bahru which are; i) land use not within walking distance; ii) oversized residential streets; and iii) inadequate pedestrian paths and/or cycle paths. 4. Materials and Method 4.1 Study Area The study area is Iskandar Malaysia and the samples are 30 neighbourhoods located within Iskandar Malaysia and have different land areas between 10 acres (minimum) and 3000 acres (maximum) (Fig. 1). The average land area of neighbourhoods is about 1000 acres. Spatial data, including distance and area were obtained from Iskandar Region Development Authority, IRDA. 4.2 Data Collection To understand the changing of VMT in Iskandar Malaysia, we conducted a survey of residents in thirty selected neighbourhood located at various distances from Johor Bahru City Centre. The selection of these neighbourhoods was mainly based on the time-periods during which they were developed (from 1960 till present). The survey was conducted in two stages and two different methods. Early data collection method was conducted by means of distribution of travel diary and questionnaire among school students, with the assumption that these students come from the neighbourhoods that their schools are in. Travel diary is a crucial method in order to obtain respondent vehicle miles travel data (see Cervero & Kockelman, 1997; Ewing & Cervero, 2001; Ewing et al, 2009). In this research, school students were asked to record their father or family vehicle odometer reading for one week. In addition, household data and travel data was obtained using a questionnaire that was distributed together with the travel diary. The questionnaire consists of two sections, which are Household Demographic Data in section A, while section B gathers the Household Travel Information which includes: work trip, non-work trip, provision of public transportation, neighborhood self-selection, travel frequency and travel attitude. Different with travel diary, in order to collect fair data, school students were asked to pass the questionnaire to their parent to be filled. A total of 147 respondent responses and return the questionnaires from this method. Next, additional data were gathered by conducting a Home to Home VMT household data survey. In this method, the travel diary and the questionnaire was given to the respondent for one week to be filled. The travel diary and questionnaires were distributed on Day 1 and were collected on Day 8. Respondent who return the travel diary and questionnaire were given RM5 petrol money as a gesture Page 3

4 for responding to the survey. From this method, another 229 travel diaries and questionnaires were managed to be collected. Fig.1. Residential neighbourhoods of Iskandar Malaysia by year of development 5. Analysis and Discussion Table 2 displays briefly the demographic characteristic of the respondents. From the survey, percents respondents own at least one car and percents respondent own two cars. The remaining percents respondents are those who own more than two cars. Building decade of neighbourhood Table 2: Descriptive statistic travel data of neighbourhoods by years No.of Travel Household Household neighbourhood Distance Income Member Car Ownership per House (KM) Mean Mean Mean Mean Mean Pre 1980 s s s s License Ownership per House The results of the VMT survey indicate that average daily travel per household increases against age of the neighbourhoods (Fig. 2). Recently-approved neighbourhoods, i.e. new designs, tend to generate higher VMT compared to old ones. On average, daily travels range from 40 km/day for neighbourhoods built in the Pre 1980 s to about 100 km/day for those built in the 2000 s. Page 4

5 Indexes VMT (km/day) VMT Linear (VMT) Year Layout Approved Fig. 2: Higher VMT (km/day) are generated in recently approved neighbourhoods. Findings on this study confirm the prevalent theory on the relationship between neighbourhood design attributes and VMT (Fig. 3). 3 Proximity Index Simpson Diversity Index Connectivity Index Intersection Density Linear (Proximity Index) Linear (Simpson Diversity Index) Linear (Connectivity Index) Linear (Intersection Density) VMT (km/day) Fig. 3: Vehicle Miles Travelled is influence by the neighbourhood design factors Diversity index is among the neighbourhood design attributes that have been identified to have contributed to the increase of VMTs in the study area. A study by Kassim [15] has highlighted a progressive decrease of land use diversity in neighbourhoods in Iskandar Malaysia, including the neighbourhoods in this study. Figure 3 plots the VMTs obtained from this study against the diversity index from Kassim [15] study for the neighbourhoods shared by both studies. Figure 3 indicates that VMT increases as neighbourhood land use diversity decreases. From a total of 30 neighbourhoods surveyed, 67 percents or approximately 20 neighbourhoods hold low value of diversity index range from Nine of the neighbourhoods are a smaller settlement with acreage below 500 acres while the remaining balance was a larger settlement. Residents live in small settlement with low land use diversity will need to travel far to get to the facilities or services that are not provided within their Page 5

6 neighbourhoods. As for the larger settlement, no specific requirement on where the commercial areas shall be located within the neighbourhood resulting in more than one commercial area located scattered over the neighbourhood. Poor location and dispersion of urban land uses over a large area may lead to longer distances, which lead residents to use their cars [9]. Apart from this, the location of the commercial areas within neighbourhoods also influenced VMT per household. Analysis of VMT against these variables indicates that VMT is inversely proportional to the centrality of commercial area location. Centrality of commercial area location is crudely represented by the distance between the centroid of each neighbourhood area and its commercial area. Study conducted by Aziz [16] found that the average proximity distance, both Euclidean and Manhattan distance between homes to the commercial center is increasing. From as low as 300 meters average Euclidean distance in pre 1980 s, the average distance up to almost meters in 2000s. As for the average Manhattan distance, the average values in 2000s were between meters, almost double compared to the average Euclidean distance. This shows that recent neighbourhood developments were developed with residential area located far from the commercial area. This does not only limiting the potential for respondent to walk or cycle to the neighbourhood commercial center but also encourage the usage of automobile. Another factor identified to influence VMT is neighbourhood intersection density. Theoretically, higher intersection density lowers VMT in the neighbourhood [6] and grid-like street pattern encourages walking since the inter connectivity between destination is higher. The neighbourhoods in Iskandar Malaysia have, over the years, gone through a lot of changes with the street pattern from grid to cul-desac, loops as well as curvilinear, thus resulting in lower intersection density and finally lower connectivity index [17]. VMT analysis done in this study shows that VMT increases with the decrease in connectivity index which happens to keep decreasing as more recent neighbourhoods are shying away from the grid pattern into more aesthetic curvilinear pattern or more private cul-de-sac-dominated pattern. 6. Conclusion The most important finding from the study in terms of effort to minimize greenhouse gas emission is that household VMT in Iskandar Malaysia is on the increase. More recent designs of neighbourhood seem to contribute negatively towards this, i.e. higher household VMT coming out from neighbourhoods developed the last couple of years than decades before. Malaysia in general and Iskandar Malaysia in particular are taking its mixed development concept, which has been initiated since the late 1970s, further away from the New Urbanism concept while the rest of the world are embracing New Urbanism. While many cities are talking about reduction of VMT through better and compact designs, unfortunately for Iskandar Malaysia, the demand for neighbourhood are no longer based on compactness or mixed uses either. Neighbourhoods in Iskandar Malaysia now are becoming bigger with less diversity of land uses and not to mention the layout were design with residential located far from commercial areas. Those low diversity neighbourhoods are normally small ones exclusively zoned for housing, forcing the residents to travel beyond the neighbourhoods for amenities and facilities. Preference for curvilinear design rather than the higher connectivity grid design also increases household VMT. This is cause for alarm as Iskandar Malaysia is targeting to develop a low carbon city with reduction of carbon emission up to 40 percent in Acknowledgement The author acknowledges the funding support for this research provided by the Ministry of Higher Education, Malaysia through Universiti Teknologi Malaysia under the Research University Grant Programme (GUP) - Vot no.: 00H59 8. References [1] Ewing, R., Cevero, R., Travel and the built environment: a synthesis. Transportation Research Record 1780, [2] Miller, E.J., Ibrahim, A., Urban form and vehicular travel: some empirical findings. Transportation Research Record 1617, Motortrend Magazine, < Page 6

7 [3] M. Lindsey et al The effect of residential location on vehicle miles of travel, energy consumption and greenhouse gas emissions: Chicago case studytransportation Research Part D 16 (2011) 1 9. [4] MIROS (Malaysian Institute of Road Safety Research) News & Events. Vehicle Kilometer Travelled. [online] Available at: [Accessed 1 September 2012] [5] Majid, M. R. and Hafizul Ridzwan Yahya Urban Sprawl in Malaysia: Is Compact Development The Answer?. Proceedings of The 5th South East Asian Technical University Consortium (SEATUC) Symposium, Hanoi University of Science and Technology, Hanoi, Vietnam Feb ISSN: pp [6] Cervero, R. & Kockelman, K., Travel Demand and The 3Ds: Density, Diversity, and Design. Transportation Research.-D, Vol. 2, No. 3, pp [7] Ewing, et al., Measuring the impact of urban form and transit access on mixed use site trip generation rates Portland pilot study. Washington, DC: U.S. Environmental Protection Agency. [8] Steiner, R., 1994 Residential density and travel patterns: review of the literature, Transportation Research Record, 1466, pp [9] Levinson, D. M. and Kumar, A., Density and the journey to work, Growth and Change, 28, pp [10] Banister, D Reducing the need to travel" Environment and Planning B: Planning and Design 24(3) [11] Kulash, W., Anglin, J. and Marks. D Traditional Neighborhood Development: Will the Traffi c Work? Development 21:4 21. [12] McNally, M, and Ryan. S A Comparative Assessment of Travel:Characteristics of Neotraditional Developments. Irvine, CA: Institute of Transportation Studies, University of California. [13] Calthorpe, P The Next American Metropolis: Ecology, Community and the American Dream. New York: Princeton Architectural Press. [14] Hussein, M. Z. S. M., New Urbanism and Sustainable Residential Street Design: The Case of Malaysian Neighbourhoods. In: International Seminar on Sustainable Planning and Governance (II), 2007, Korea [15] Kassim, Sh. A. I., Kepelbagaian guna tanah dalam pembangunan bercampur. Undergraduate Thesis. Universiti Teknologi Malaysia. Unpublished. [16] Aziz, N.D., Analisis kedekatan bagi kawasan perumahan ke pusat komersil di Iskandar Malaysia. Undergraduate Thesis. Universiti Teknologi Malaysia. Unpublished. [17] Rusman, M.F., Kesambungan jalan dalam taman perumahan dalam Iskandar Malaysia. Undergraduate Thesis. Universiti Teknologi Malaysia. Unpublished. Page 7