GIS-Based Analytical Tools for Transport Planning:
Spatial Regression Models for Transportation Demand Forecast
imone Becker Lopes 1, Nair Cristina Margarido Brondino 2 and Antônio Nélson Rodrigues da Silva 1,*
1 Department of Transportation Engineering, São Carlos School of Engineering, University of São Paulo, Av. Trabalhador São-carlense 400, 13566-590 São Carlos, Brazil; E-Mail: lopes.simoneb@gmail.com
2 Department of Mathematics, Faculty of Science, São Paulo State University, Av. Luis Edmundo Carrijo Coube 14-01, 17033-360 Bauru, Brazil; E-mail: brondino@fc.unesp.br
* Author to whom correspondence should be addressed; E-Mail: anelson@sc.usp.br; Tel.: +55-16-3373-9595; Fax: +55-16-3373-9602.
ISPRS Int. J. Geo-Inf. 2014, 3, 565-583; doi:10.3390/ijgi3020565
Abstract:
Considering the importance of spatial issues in transport planning, the main objective of this study was to analyze the results obtained from different approaches of spatial regression models. In the case of spatial autocorrelation, spatial dependence patterns should be incorporated in the models, since that dependence may affect the predictive power of these models. The results obtained with the spatial regression models were also compared with the results of a multiple linear regression model that is typically used in trips generation estimations. The findings support the hypothesis that the inclusion of spatial effects in regression models is important, since the best results were obtained with alternative models (spatial regression models or the ones with spatial variables included).
This was observed in a case study carried out in the city of Porto Alegre, in the state of Rio Grande do Sul, Brazil, in the stages of specification and calibration of the models, with two distinct datasets.
Keywords: transport planning; transport demand; spatial dependence; spatial regression
1 Introduction
Spatial relationships play an important role in transport. Even though, there are not so many studies focusing on the explicit introduction of spatial issues in transport planning modeling. Thus, as a contribution to the field, the objective of this study is to analyze the results obtained from different approaches of spatial regression models. Next, the outcomes of these spatial models are also compared with the results of a multiple linear regression model that is typically used in trips generation estimations. The key research question of this study was thus whether or not the inclusion of spatial variables improves transport demand models. Many researchers have already discussed the importance of considering spatial effects in urban and transportation analyses. Páez and Scott [1], for instance, have made a review of techniques and examples of applications illustrating how spatial statistics can be used in urban transportation and land use planning. The objective of that study was to discuss some of the major technical issues in spatial analysis (i.e., spatial association, heterogeneity and the modifiable areal unit problem) and the authors indicated a promising trend for the application of increasingly sophisticated spatial statistical methods in urban analyses. These topics are still timely, as recently discussed by Wang et al. [2].
Spatial dependence and its effects on transportation demand models, which are the focus of this study, are undoubtedly among the issues concerning spatial analysis that have not been fully explored in transport planning yet. This can be seen in Table 1, in which a review of studies conducted in the past three decades about spatial effects on transportation and urban analysis was summarized. The table is organized in such a way that the references are shown in the central column, the spatial analytical issues explored are listed on the left side of the table and the fields of application are listed on the right side of the table. Regarding the spatial analytical issues, most of the selected studies focused on issues of spatial association (i.e., spatial dependence or spatial autocorrelation). Regarding the applications, only a few of them dealt with transportation demand analyses. It is worth mentioning that almost all studies have reached a common conclusion: the inclusion of spatial effects improved the analyses results. This is not really a surprise, but it calls the attention to the fact that many studies that are not listed in Table 1 still do not explicitly include spatial analysis elements in their analyses.
Regression models, for example, are commonly used in the trip generation phase of transport planning.
They are statistical tools that explore the existing relationships among two or more variables, so that one of them can be explained (and therefore its value can be estimated) by the other(s). However, in the presence of a significant spatial autocorrelation, model estimations have to consider and to incorporate the spatial structure of data. Spatial regressions, or regression analyses incorporating the existing spatial dependence of data, are likely to improve the predictive power of the regression models.
Bolduc et al. [3–5], Haider and Miller [6], Wang [7], Czado and Prokopenko [8], Kawamura and Mahajan [9], Vichiensan et al. [10], Zhou and Kockelman [11], Ribeiro and Antunes [12], Chalermpong [13], Hackney et al. [14,15], and Novak et al. [16] provide examples of applications of spatial regression, some of them in urban and transportation planning. In general, the spatial models tested had a better fit to the actual data than the respective non-spatial models.
References
1. Páez, A.; Scott, D.M. Spatial statistics for urban analysis: A review of techniques with examples. GeoJournal 2004, 61, 53–67.
2. Wang, C.; Quddus, M.A.; Ryley, T.; Enoch, M.; Davison, L. Spatial Models in Transport: A Review and Assessment of Methodological Issues. In Proceedings of the 91st Annual Meeting of the Transportation Research Board, Washington, DC, USA, 22–26 January 2012.
3. Bolduc, D.; Dagenais, M.G.; Gaudry, M.J.I. Spatially autocorrelated errors in origin-destination models: A new specification applied to aggregate mode choice. Transp. Res. Part B 1989, 23, 361–372.
4. Bolduc, D.; Laferriere, R.; Santarossa, G. Spatial autoregressive error-components in travel flow models. Reg. Sci. Urban Econ. 1992, 22, 371–385.
5. Bolduc, D.; Laferriere, R.; Santarossa, G. Spatial Autoregressive Error Components in Travel Flow Models: An Application to Aggregate Mode Choice. In New Directions in Spatial Econometrics; Anselin, L., Florax, R.J.G.M., Eds.; Springer-Verlag: Berlin, Germany, 1995; pp. 96–108.
6. Haider, M.; Miller, E.J. Effects of transportation infrastructure and location on residential real estate values: Application of spatial autoregressive techniques. Transp. Res. Rec. 2000, 1722, 1–8.
7. Wang, F.H. Explaining intraurban variations of commuting by job proximity and workers’ characteristics. Environ. Plann. B 2001, 28, 169–182.
8. Czado, C.; Prokopenko, S. Modelling transport mode decisions using hierarchical logistic regression models with spatial and cluster effects. Stat. Model. 2008, 8, 315–345.
9. Kawamura, K; Mahajan, S. Hedonic analysis of impacts of traffic volumes on property values. Transp. Res. Rec. 2005, 1924, 69–75.
10. Vichiensan, V.; Páez, A.; Kawai, K.; Miyamoto, K. Non stationary spatial interpolation method for urban model development. Transp. Res. Rec. 2006, 1977, 103–111.
11. Zhou, B.; Kockelman, K.M. Microsimulation of Residential Land Development and Household Location Choices: Bidding for Land in Austin, Texas. In Proceedings of the 87th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 13–17 January 2008.
12. Ribeiro, A.; Antunes, A.P. Accessibility and the Development in Portuguese Off-Cost Regions: A Spatial Regression Analysis. In Proceedings of the European Transport Conference, Noordwijkerhout, The Netherlands, 5–7 October 2009.
13. Chalermpong, S. Rail Transit and Residential Land Use in Developing Countries: Hedonic Study of Residential Property Prices in Bangkok, Thailand. In Proceedings of the 86th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 21–25 January 2007.
14. Hackney, J.; Bernard, M.; Bindra, S.; Axhausen, K.W. Explaining Road Speeds with Spatial Lag and Spatial Error Regression Models. In Proceedings of 86th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 21–25 January 2007.
15. Hackney, J.; Bernard, M.; Bindra, S.; Axhausen, K.W. Predicting road system speeds using spatial structure variables and network characteristics. J. Geogr. Syst. 2007, 9, 397–417.
16. Novak, D.C.; Hodgdon, C.; Guo, F.; Aultman-Hall, L. Nationwide freight generation models: A spatial regression approach. Netw. Spat. Econ. 2011, 11, 23–41.
17. Bender, B.; Hwang, H. Hedonic housing price indices and secondary employment centers. J. Urban Econ. 1985, 17, 90–107.
18. Eom, J.K.; Park, M.S.; Heo, T.; Huntsinger, L.F. Improving the prediction of annual average daily traffic for nonfreeway facilities by applying a spatial statistical method. Transp. Res. Rec. 2006, 1968, 20–29.
19. Wang, X.; Kockelman, K.M. Forecasting Network Data: Spatial Interpolation of Traffic Counts Using Texas Data. In Proceedings of the 88th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 11–15 January 2009.
20. Bhat, C.; Zhao, H.M. The spatial analysis of activity stop generation. Transp. Res. Part B 2002, 36, 557–575.
21. Kwan, M.P. Interactive geovisualization of activity-travel patterns using three-dimensional geographical information systems: A methodological exploration with a large dataset. Transp. Res. Part C 2000, 8, 185–203.
22. Steenberghen, T.; Dufays, T.; Thomas, I.; Flahaut, B. Intra-urban location and clustering of road accidents using GIS: A Belgian example. Int. J. Geogr. Inf. Sci. 2004, 18, 169–181.
23. Li, L.; Zhang, Y. A GIS-Based Bayesian Approach for Identifying Hazardous Roadway Segments for Traffic Crashes. In Proceedings of the 86th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 21–25 January 2007.
24. Gundogdu, I.B.; Sari, F.; Esen, O. A New Approach for Geographical Information System-Supported Mapping of Traffic Accident Data. In Proceedings of the FIG Working Week 2008, Stockholm, Sweden, 14–19 June 2008.
25. Khan, G.; Santiago-Chaparro, K.R.; Qin, X.; Noyce, D.A. Application and Integration of Lattice Data Analysis, Network k-Functions, and GIS to Study Ice-Related Crashes. In Proceedings of the 88th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 11–15 January 2009.
26. Guo, F.; Wang, X.; Abdel-Aty, M.A. Corridor Level Signalized Intersection Safety Analysis Using Bayesian Spatial Models. In Proceedings of the 88th Annual Meeting of the Transportation Research Board, Washington, DC, USA, 11–15 January 2009.
27. Páez, A.; López, F.A.; Ruiz, M.; Morency, C. Development of an indicator to assess the spatial fit of discrete choice models. Transp. Res. Part B. 2013, 56, 217–233.
28. Victoria, I.C.; Prozzi, J.; Walton, C.M.; Prozzi, J. Environmental justice concentration zones for assessing transportation project impacts. Transp. Res. Rec. 2006, 1983, 75–80.
29. Ibeas, A.; Cordera, R.; dell’Olio, L.; Coppola, P.; Dominguez, A. Modelling transport and real-estate values interactions in urban systems. J. Transp. Geogr. 2012, 24, 370–382.
30. Efthymiou, D.; Antoniou, C. How do transport infrastructure and policies affect house prices and rents? Evidence from Athens, Greece. Transp. Res. Part A 2013, 52, 1–22;
31. Horner, M.W.; Murray, A.T. Excess commuting and the modifiable areal unit problem. Urban Stud. 2002, 39, 131–139.
32. Lopes, S.B.; Rodrigues da Silva, A.N. An Assessment Study of the Spatial Dependence in Transportation Demand Models. In Proceedings of the 13th Pan American Conference of Traffic and Transportation Engineering, Albany, NY, USA, 27–29 September 2004.
33. Anselin, L. Under the hood: Issues in the specification and interpretation of spatial regression models. Agricult. Econ. 2002, 17, 247–267.
34. Fotheringham, A.S.; Brunsdon, C.; Charlton, M. Quantitative Geography: Perspectives on Spatial Data Analysis; Sage: London, UK, 2000.
35. Anselin, L. The Moran Scatterplot as an ESDA Tool to Assess Local Instability in Spatial Association. In Spatial Analytical Perspectives on GIS; Fischer, M., Scholten, H., Unwin, D., Eds.; Taylor & Francis: London, UK, 1996: pp. 111–125.
36. Anselin, L. GeoDa™ 0.9 User’s Guide. Available online: https://geodacenter.org/downloads/ pdfs/geoda093.pdf (accessed on 31 December 2013).
37. Getis, A.; Griffith, D.A. Comparative spatial filtering in regression analysis. Geogr. Anal. 2002, 34, 130–140.
38. Greene, W.H. Econometric Analysis; Prentice Hall: Upper Saddle River, NJ, USA, 2003.
ليست هناك تعليقات:
إرسال تعليق