A Hybrid Geostatistical Method for Estimating Citywide Traffic Volumes – A Case Study of Edmonton, Canada

Source:
By:Mingjian Wu, Tae J. Kwon, Karim El-Basyouny

DOI: https://doi.org/10.30564/jgr.v5i2.4513

Abstract: Traffic volume information has long played an important role in many transportation related works, such as traffic operations, roadway design, air quality control, and policy making. However, monitoring traffic volumes over a large spatial area is not an easy task due to the significant amount of time and manpower required to collect such large-scale datasets. In this study, a hybrid geostatistical approach, named Network Regression Kriging,has been developed to estimate urban traffic volumes by incorporating auxiliary variables such as road type, speed limit, and network accessibility.Since standard kriging is based on Euclidean distances, this study implements road network distances to improve traffic volumes estimations.A case study using 10-year of traffic volume data collected within the city of Edmonton was conducted to demonstrate the robustness of the model developed herein. Results suggest that the proposed hybrid model significantly outperforms the standard kriging method in terms of accuracy by 4.0% overall, especially for a large-scale network. It was also found that the necessary stationarity assumption for kriging did not hold true for a large network whereby separate estimations for each road type performed significantly better than a general estimation for the overall network by 4.12%. References:

[1] Elvik, R., Høye, A., Vaa, T., et al., 2009. The Contribution of Research to Road Safety Policy-Making. The Handbook of Road Safety Measures, Emerald Group Publishing Limited. pp.117-141.DOI: https://doi.org/10.1108/9781848552517-006. [2] Minderhoud, M.M., Botma, H., Bovy, P.H.L., 1997.Assessment of Roadway Capacity Estimation Methods. Transportation Research Record. 1572(1), 59-67. DOI: https://doi.org/10.3141/1572-08 [3] Rakowska, A., Wong, K.C., Townsend, T., et al.,2014. Impact of traffic volume and composition on the air quality and pedestrian exposure in urban street canyon. Atmospheric Environment. 98, 260-270.DOI: https://doi.org/10.1016/j.atmosenv.2014.08.073 [4] Wu, M., Kwon, T.J., El-Basyouny, K., 2020. A Citywide Location-Allocation Framework for Driver Feedback Signs: Optimizing Safety and Coverage of Vulnerable Road Users. Sustainability. 12(24), 10415. DOI: https://doi.org/10.3390/su122410415 [5] Skszek, S.L., 2001. State-of-the-art report on non-traditional traffic counting methods. Arizona. [6] Eom, J.K., Park, M.S., Heo, T.Y., et al., 2006. Improving the Prediction of Annual Average Daily Traffic for Nonfreeway Facilities by Applying a Spatial Statistical Method. Transportation Research Record. 1968(1), 20-29. DOI: https://doi.org/10.1177/0361198106196800103 [7] Xia, Q., Zhao, F., Chen, Z., et al., 1999. Estimation of Annual Average Daily Traffic for Nonstate Roads in a Florida County. Transportation Research Record. 1660(1), 32-40. DOI: https://doi.org/10.3141/1660-05 [8] Zhao, F., Chung, S., 2001. Contributing factors of annual average daily traffic in a Florida county: exploration with geographic information system and regression models. Transportation Research Record. 1769(1), 113-122. DOI: https://doi.org/10.3141/1769-14 [9] Mohamad, D., Sinha, K.C., Kuczek, T., et al., 1998.Annual Average Daily Traffic Prediction Model for County Roads. Transportation Research Record. 1617(1), 69-77.DOI: https://doi.org/10.3141/1617-10 [10] Tang, Y.F., Lam, W.H.K., Ng, P.L.P., et al., 2003.Comparison of four modeling techniques for shortterm AADT forecasting in Hong Kong. Journal of Transportation Engineering. 129(3), 271-277. DOI: https://doi.org/10.1061/(ASCE)0733-947X(2003)129:3(271) [11] Lam, W.H.K., Tang, Y.F., Chan, K.S., et al., 2006. Short-term Hourly Traffic Forecasts using Hong Kong Annual Traffic Census. Transportation. 33(3),291-310.DOI: https://doi.org/10.1007/s11116-005-0327-8 [12] Davis, G., Yang, S., 2001. Accounting for uncertainty in estimates of total traffic volume: an empirical Bayes approach. Journal of Transportation and Statistics. 4(1), 27-38. [13] Goel, P.K., McCord, M.R., Park, C., 2005. Exploiting Correlations between Link Flows to Improve Estimation of Average Annual Daily Traffic on Coverage Count Segments: Methodology and Numerical Study.Transportation Research Record. 1917(1), 100-107. DOI: https://doi.org/10.1177/0361198105191700112 [14] Matheron, G., December 1963. Principles of geostatistics. Economic geology. 58(8), 1246-1266. DOI: https://doi.org/10.2113/gsecongeo.58.8.1246 [15] Chiles, J., Delfiner, P., 2009. Geostatistics: modeling spatial uncertainty, John Wiley & Sons. [16] Selby, B., Kockelman, K.M., 2013. Spatial prediction of traffic levels in unmeasured locations: applications of universal kriging and geographically weighted regression. Journal of Transport Geography. 29, 24-32. DOI: https://doi.org/10.1016/j.jtrangeo.2012.12.009 [17] Kwon, T.J., Fu, L., 2017. Spatiotemporal variability of road weather conditions and optimal RWIS density—an empirical investigation. Canadian Journal of Civil Engineering. 44(9), 691-699. DOI:https://doi.org/10.1139/cjce-2017-0052 [18] Wu, M., 2020. Evaluating the Safety Effects of Driver Feedback Signs and Citywide Implementation Strategies. DOI: https://doi.org/10.7939/r3-xwkc-1112 [19] Biswas, S., Wu, M., Melles, S.J., et al., 2019. Use of Topography,Weather Zones, and Semivariogram Parameters to Optimize RoadWeather Information System Station Density across Large Spatial Scales.Transportation Research Record. 2673(12), 301-311.DOI: https://doi.org/10.1177/0361198119846467 [20] Gu, L., Wu, M., Kwon, T.J., November 2019. An Enhanced Spatial Statistical Method for Continuous Monitoring of Winter Road Surface Conditions. Canadian Journal of Civil Engineering. 47(10), 1154-1165. DOI: https://doi.org/10.1139/cjce-2019-0296 [21] Zhang, D., Wang, X.C., 2014. Transit ridership estimation with network Kriging: a case study of Second Avenue Subway, NYC. Journal of Transport Geography. 41, 107-115.DOI: https://doi.org/10.1016/j.jtrangeo.2014.08.021 [22] Gneiting, T., Genton, M.G., Guttorp, P., 2006. Geostatistical space-time models, stationarity, separability, and full symmetry. Monographs On Statistics and Applied Probability. 107, 151. [23] Roll, J., 2019. Evaluating Streetlight Estimates of Annual Average Daily Traffic in Oregon. [24] Cressie, N., 1990. The origins of kriging. Mathemati cal Geology. 22(3), 239-252.DOI: https://doi.org/10.1007/BF00889887 [25] Lichtenstern, A., 2013. Kriging methods in spatial statistics. [26] Hengl, T., Heuvelink, G., Stein, A., 2003. Comparison of kriging with external drift and regression kriging. ITC Enschede, Netherlands. [27] Olea, R.A., 2006. A six-step practical approach to semivariogram modelling. Stochastoc Envrionment Research and Risk Assessment. 20(5), 307-318.DOI: https://doi.org/10.1007/s00477-005-0026-1 [28] Cambardella, C.A., Moorman, T.B., Novak, J.M., et al., 1994. Field-scale variability of soil properties in central Iowa soils. Soil Science Society of America Journal. 58(5), 1501-1511.DOI: https://doi.org/10.2136/sssaj1994.03615995005800050033x [29] Sun, X.L., Yang, Q., Wang, H.L., et al., 2019. Can regression determination, nugget-to-sill ratio and sampling spacing? Catena. 181, 104092. DOI: https://doi.org/10.1016/j.catena.2019.104092 [30] Bohling, G., 2005. Introduction to geostatistics and variogram analysis. Kansas Geological Survey. pp.1-20. [31] Solana-Gutiérrez, J., Merino-de-Miguel, S., 2011. A Variogram Model Comparison for Predicting Forest Changes. Procedia Environmental Sciences. 7, 383-388.DOI: https://doi.org/10.1016/j.proenv.2011.07.066 [32] Desktop, ESRI ArcGIS, 2011. Release 10. Redlands, CA: Environmental Systems Research Institute. 437, 438. [33] Van Rossum, G., Drake, F.L., 2009. Python 3 Reference Manual, Scotts Valley, CA: CreateSpace. [34] Mälicke, M., Schneider, H.D., 2019. Scikit-GStat 0.2.6: A scipy flavoured geostatistical analysis toolbox written in Python. (Version v0.2.6). Zenodo. [35] Oz, B., Deutsch, C.V., 2000. Cross Validation for Selection of Variogram Model and Kriging Type: Application to IP Data from West Virginia. Center for Computational Geostatistics Annual Report Papers. pp. 1-13. [36] Brooker, P.I., 1986. A parametric study of robustness of kriging variance as a function of range and relative nugget effect for a spherical semivariogram. Mathematical Geology. 18(5), 477-488.DOI: https://doi.org/10.1007/BF00897500 [37] Skøien, J.O., Merz, R., Blöschl, G., 2006. Top-kriging - geostatistics on stream networks. Hydrology and Earth System Sciences. 10(2), 277-287.DOI: https://doi.org/10.5194/hess-10-277-2006 [38] Miura, H., 2010. A study of travel time prediction using universal kriging. 18(1), 257-270.DOI: https://doi.org/10.1007/s11750-009-0103-6 [39] Thakali, L., Kwon, T.J., Fu, L., 2015. Identification of crash hotspots using kernel density estimation and kriging methods: a comparison. Journal of Modern Transportation. 23(2), 93-106.DOI: https://doi.org/10.1007/s40534-015-0068-0 [40] Bayraktar, H., Turalioglu, F.S., 2005. A Kriging-based approach for locating a sampling site in the assessment of air quality. Stochastic Environmental Research and Risk Assessment. 19(4),301-305.DOI: https://doi.org/10.1007/s00477-005-0234-8