ACCESS: An agent-based model to explore job accessibility inequalities

https://doi.org/10.1016/j.compenvurbsys.2020.101462Get rights and content

Highlights

  • This paper proposed ACCESS, an agent-based model of job accessibility inequalities.

  • ACCESS allows to simulate the residential location of different social groups.

  • Experiments reveal that transport interventions can promote gentrification.

  • Land use and transport policies should be integrated to reduce inequalities.

Abstract

This paper presents ACCESS, an agent-based model for exploring job accessibility inequalities among different social groups. ACCESS allows for investigation on the impact of public transport and land use policies on the residential location of the working population and their accessibility to job opportunities. ACCESS can be adapted to different realities, allowing to represent societies with diverse socioeconomic disparities. A utility function composed of job accessibility and neighborhood status is maximized by agents during the residential location choice process. The model outputs include Lorenz curves considering the accessibility dimension, as well as Gini metrics to support the analysis of interventions impacts on accessibility inequalities. An empirical case study is performed on the municipality of Sao Paulo, which is characterized by high levels of inequality. Five experiments were simulated considering three different socio-occupational groups. The first experiment includes (i) new public mass rail transport lines, and the other four experiments consider the new transport infrastructure from the first experiment and add (ii) new social housing location strategies; (iii) new job locations; (iv) new jobs and different social housing supplies and location strategies; and (v) provision of social housing based on a government housing program. The results show that ACCESS allowed the residential location of different social status groups to be depicted with a high correlation to the observed situation. Regarding the case study, the results indicate that only having interventions on transport system is insufficient to provide a significant change in terms of inequality. Better results that impact inequality are reached with public mass rail transport interventions associated with land use policies with different social housing and job location programs.

Introduction

In spite of the changes currently taking place in labor relations worldwide, such as the automation and modernization of activities, a substantial portion of trips in urban centers are work-motivated. The relevance of these work-motivated travels reflects the long tradition of researchers studying it, as travelling to a job is frequently reported by people as the most important travel activity (Grengs, 2015).

One convenient way of measuring how efficient a transport system is in providing access to opportunities is through accessibility measures. During the last fifty years, accessibility concepts have become central to territorial planning (Batty, 2009). Accessibility, or “the potential for reaching spatially distributed opportunities”, is considered to be the joint result of the geographical distribution of transport and activities (Paez, Scott, & Morency, 2012). Several authors have contributed to its definition, applications and overviews (Anas, 1983; Batty, 2009; El-Geneidy & Levinson, 2006; Geurs & van Wee, 2004; Hagerstrand, 1970; Handy & Niemeier, 1997; Hansen, 1959; Iacono & Levinson, 2017; Ingram, 1971; Lei & Church, 2010; Paez et al., 2012; Stewart, 1948).

Through the previous accessibility definition, it is observed that accessibility measures are capable of identifying the distributional effects of transport projects and policies across specific population groups and regions (van Wee & Geurs, 2011). Accessibility inequality is discussed by various authors in terms of its definitions and measures (Di Ciommo & Shiftan, 2017; Litman, 2002; Neutens, Schwanen, Witlox, & De Maeyer, 2010), the impacts of public transport improvements on equality (Bills & Walker, 2017; Chen, Ni, Xi, Li, & Wang, 2017; Currie et al., 2009; Monzon, Ortega, & Lopez, 2013; Pereira, Banister, Schwanen, & Wessel, 2017) and the potential benefits of multimodal transit trips (Boarnet, Giuliano, Hou, & Shin, 2017; Pritchard, Tomasiello, Giannotti, & Geurs, 2019).

To alleviate job accessibility inequalities, public policies on transport and land use can be implemented, but their outcomes are not straightforward due to the high complexity of their interaction in the urban environment. The land use and transport cycle, as discussed by Paquette, Ashford, and Wright (1972), illustrates the complex and reciprocal influence of both components. The cycle begins with the addition of transportation facilities, which results in increased accessibility and land value, leading to changes in the land use that further increase the trip generation and, thus, resulting in higher traffic needs (Paquette et al., 1972). Throughout this process, transport and land use, which are components of the gentrification process, can influence incoming residents and the burden of displaced residents, especially in cities with high transportation costs (Eckerd, Kim, & Campbell, 2019).

The work developed by Mejia-Dorantes and Lucas (2014) shows the outcomes from transport and land-use integrated policies by comparing the London Jubilee Line Extension and the Madrid Metrosur impacts on economic and land use development. The authors concluded that, besides the existing economic conditions, an economic uplift occurred around stations with land use plans to stimulate urban densities, mixed land uses, and pedestrian and cycling access.

To better understand the impacts of transport and land use policies on urban environments, it is important to develop models with the capability to simulate different transport interventions and land use policies and incorporate their dynamics to provide resulting outcomes. According to Wise, Crooks, and Batty (2017), methodologies on agent-based modeling have shown the capacity to emulate the relations between urban systems and transportation. The work developed by Torrens and Nara (2007) shows that a hybrid agent-cellular automata model is useful in representing human behavior in complex urban systems to explore gentrification process. In addition to the inherent simplification on model creation, the simulation of transport and land use policies allows insights to be obtained on what parameters are more relevant and the efficiency of different policy scenarios.

This work aims to investigate the impacts of public transport and land use policies on job accessibility inequalities through ACCESS. This model allows for outcomes on job accessibility inequality to be simulated based on different transport and land use policies while considering the working population stratified into different social groups. The model simulates the residential location choice of the working population based on the job accessibility and neighborhood status.

ACCESS is empirically evaluated using data from Sao Paulo, which is a megacity in a global south developing country that presents a remarkable unequal accessibility condition among different social groups. The municipality of Sao Paulo, Brazil is one of the five largest cities in the world, with approximately 12 million inhabitants (UN, 2018). It has a core-periphery pattern, where the central area concentrates most of the jobs, wealthy neighborhoods and public transport supply, while the periphery has resulted from an unplanned expansion and is mainly occupied by the poor and less educated population (Moreno-Monroy & Ramos, 2015). Sao Paulo metropolitan region is formed by 39 municipalities. Despite being the main economic and financial center of Brazil, responsible for 19% of the Brazilian GDP (Haddad, Hewings, Porsse, Van Leeuwen, & Vieira, 2015), it presents huge social inequalities (Slovic, Tomasiello, Giannotti, Andrade, & Nardocci, 2019). A bus network and a railway system integrate the municipalities in the metropolitan region (Moreno-Monroy, Lovelace, & Ramos, 2018).

The remainder of this article is organized as follows. In Section 2, the application of transport in residential location choice models is reviewed. In Section 3, we present ACCESS, its submodels and specifications. In Section 4 we show the results considering the calibration and the case study experiments. Finally, the discussions and conclusions are presented in Section 5.

Section snippets

Transport in agent-based residential location choice models

Agent-based models are increasingly being utilized to evaluate environmental and social systems due to their framework, which allows interactions between individual heterogeneous entities to be investigated across space and time (Millington, O'Sullivan, & Perry, 2012).

Agent-based models that use transport proxies in residential location choice were applied in several studies. In many cases, transport was incorporated by a distance proxy (Feitosa, Le, & Vlek, 2011; Guo, Buchmann, & Schwarz, 2019

The ACCESS model

The ACCESS model focuses on job accessibility and its inequalities through social groups and space. This model adopts a generative approach (Epstein, 1999), where the spatial patterns of the residential location that result in different accessibility conditions among social groups emerge from agents that are initially randomly located. In the model, three groups of agents with different relocation priorities in terms of job accessibility and neighborhood status dispute the housing market in an

Experiments

In addition to the baseline scenario described in Section 3.3, five alternative experiments were simulated. All experiments began with agents being randomly allocated in the environment and with the α parameter from utility function being set to 0.6 for the high-status group, 1 for the middle-status group and 0.2 for the low-status group.

In the first experiment, two new public mass rail transport lines, which are part of the government plan of investments in transport infrastructure, were

Discussions and conclusions

This paper aimed to simulate the potential impacts of new transport infrastructure along with land use policies related to social housing and job programs. To explore the impacts of land use policies on job accessibility inequalities, we developed ACCESS.

ACCESS was conceived to be a tool to support decision-making by providing insights into different transport and land use policy impacts on alternative realities, as well as to stimulate the debate on job accessibility inequalities and how to

Acknowledgments

The authors would like to acknowledge the financial support from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) (Finance Code 001); the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) [grant number: 310908/2017-5] and [grant number: 421423/2018-8]; the São Paulo Research Foundation (FAPESP) [grant number:15/50127-2]; and the University of São Paulo dean office. We also would like to thank Dr. Patricia S. Lavieri, the Department of

References (51)

  • A. Monzon et al.

    Efficiency and spatial equity impacts of high-speed rail extensions in urban areas

    Cities

    (2013)
  • A.I. Moreno-Monroy et al.

    Public transport and school location impacts on educational inequalities: Insights from Sao Paulo

    Journal of Transport Geography

    (2018)
  • A. Paez et al.

    Measuring accessibility: Positive and normative implementations of various accessibility indicators

    Journal of Transport Geography

    (2012)
  • J.P. Pritchard et al.

    Potential impacts of bike-and-ride on job accessibility and spatial equity in Sao Paulo, Brazil

    Transportation Research Part A: Policy and Practice

    (2019)
  • A.D. Slovic et al.

    The long road to achieving equity: Job accessibility restrictions and overlapping inequalities in the city of Sao Paulo

    Journal of Transport Geography

    (2019)
  • P.M. Torrens et al.

    Modeling gentrification dynamics: A hybrid approach

    Computers, Environment and Urban Systems

    (2007)
  • C. Zhuge et al.

    Agent-based joint model of residential location choice and real estate price for land use and transport model

    Computers, Environment and Urban Systems

    (2016)
  • W. Alonso

    Location and land use: Towards a general theory of land rent

    (1964)
  • A.S. Babakan et al.

    Impacts of transport development on residence choice of renter households: An agent-based evaluation

    Habitat International

    (2015)
  • J. Barros

    Simulating urban dynamics in Latin American cities

  • M. Batty

    Accessibility: In search of a unified theory

    Environment and Planning. B, Planning & Design

    (2009)
  • I. Benenson et al.

    Entity-based modeling of urban residential dynamics: The case of Yaffo, Tel Aviv

    Environment and Planning. B, Planning & Design

    (2002)
  • F. Di Ciommo et al.

    Transport equity analysis

    Transport Reviews

    (2017)
  • A. Eckerd et al.

    Gentrification and displacement: Modeling a complex urban process

    Housing Policy Debate

    (2019)
  • A.M. El-Geneidy et al.

    Access to destinations:Development of accessibility measures

    Transportation Research

    (2006)
  • Cited by (20)

    • Spatial and social inequalities of job accessibility in Kunshan city, China: Application of the Amap API and mobile phone signaling data

      2022, Journal of Transport Geography
      Citation Excerpt :

      The Gini index (Gini, 1912) is commonly used in economics and transportation. Using the Gini index to evaluate the equity of job accessibility has been used in both Global South and North (Grengs, 2012; Tomasiello et al., 2020). Giannotti et al. (2021) use the Gini index to compare the transport equity of job accessibility in São Paulo and London, finding that São Paulo is more unfair than London.

    • Access to urban parks: Comparing spatial accessibility measures using three GIS-based approaches

      2021, Computers, Environment and Urban Systems
      Citation Excerpt :

      First, it provides different ways for more realistic accessibility measures by comparing the importance of factors involved in the measuring process. The selection of such factors (e.g., destination choices and distance thresholds) for measuring optimization can be decided by and suitable in different research contexts and purposes (e.g., Santos, Almeida, Martins, Gonçalves, & Martins, 2019; Tahmasbi & Haghshenas, 2019; Tomasiello, Giannotti, & Feitosa, 2020). Second, the methodological framework used in this study is applicable to other study areas with different geographic contexts and data availability.

    • Simulating the urban spatial structure with spatial interaction: A case study of urban polycentricity under different scenarios

      2021, Computers, Environment and Urban Systems
      Citation Excerpt :

      Many modern urban planning theories and practices promote bringing jobs closer to home and mixed land use for smart urban growth to address the increasing challenge of urban expansion and spatial mismatch, such as increasing commuting time (Cervero & Duncan, 2006). Numerous studies have also explored the relationship between employment distribution and cities' performance (Arribas-Bel & Sanz-Gracia, 2014; Hu et al., 2020; Tomasiello et al., 2020). Under this circumstance, Singapore has already listed “Bringing jobs closer to home” as an essential agenda in its urban development policy.

    • Access to urban activities during the Covid-19 pandemic and impacts on urban mobility: The Brazilian context

      2021, Transport Policy
      Citation Excerpt :

      In other words, public policies, investment, and immediate actions based on the measurement of movement do not necessarily reflect access to essential activities and services in order to sustain the urban way of life. Mobility measures are related to speed and coverage, and usually disregard or reduce the importance of non-motorized modes, focusing on increases of vehicle traffic, which can result in reductions in walkability, livability, and access to essential urban functions (Cui et al., 2019; 2018; Deboosere et al., 2018; Farber et al., 2015; Garcia, 2016; Garcia et al., 2018; Geurs and van Wee, 2004; Lopes et al., 2020; Pritchard et al., 2019; Tomasiello et al., 2020). Generally, different population groups face different challenges and opportunities concerning the adherence to social distancing and contamination risk, which are disregarded when the data and analyses supporting decision is oriented to mobility.

    • Optimization of carsharing fleet size to maximize the number of clients served

      2021, Computers, Environment and Urban Systems
      Citation Excerpt :

      Regardless of the purpose for moving, people spend effort, time and money by commuting in their cities. Recent works aimed at improving that routine understanding job inaccessibility inequalities (Tomasiello, Giannotti, & Feitosa, 2020), by analyzing and predicting patterns of urban trips by public transportation modes (Qi, Huang, Guan, & Fan, 2018), simulating public transportation routes (Kieu, Ngoduy, Malleson, & Chung, 2019), suggesting new pick-up and drop-off points (Monteiro, Martins, & Davis Jr, 2018), reducing consumption costs of electric buses (Bartomiejczyk, 2018), and reducing travel time and costs by integrating different transportation modes (Stiglic, Agatz, Savelsbergh, & Gradisar, 2018). Although public transportation is often cheaper than private transportation, issues such as inaccessibility, inflexibility of routes, low supply at certain times, high travel duration and discomfort can demotivate people to use public transportation modes, such as buses, subways, urban train, and so on.

    View all citing articles on Scopus
    View full text