Nonparametric Regression Analysis of Cyclist Waiting Times across Three Behavioral Typologies
Abstract
:1. Introduction
2. Methodology and Model
2.1. Data
2.1.1. Cyclists’ GPS Traces
2.1.2. Maneuvers Dataset
2.1.3. Data Cleaning and Feature Selection
2.2. Model Selection
2.2.1. Random Forest Regression
2.2.2. Gaussian Kernel SVM
2.3. Nonparametric Kernel Regression
2.4. Bandwidth Selection
2.5. Bootstrapping
3. Results
4. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Acknowledgments
Conflicts of Interest
Appendix A
Maneuver Number | RHC | IIC | SIC |
---|---|---|---|
1 | 0.01 (0.15) | 1.03 (0.11) | 0.15 (0.08) |
2 | 0.07 (0.49) | 0.11 (0.52) | 0.08 (0.26) |
3 | 0.84 (0.19) | 0.00 (0.10) | 0.00 (0.06) |
4 | 0.69 (1.84) | 0.46 (2.21) | 0.00 (2.75) |
5 | 0.71 (0.33) | 0.15 (0.15) | 0.00 (0.18) |
6 | 2.43 (0.18) | 5.00 (0.13) | 3.23 (0.15) |
7 | 2.68 (0.96) | 3.76 (0.98) | 1.29 (0.81) |
8 | 0.76 (0.07) | 1.85 (0.06) | 1.02 (0.06) |
9 | 1.09 (0.36) | 1.38 (0.34) | 3.28 (0.43) |
10 | 1.71 (1.89) | 2.13 (2.30) | 2.00 (2.70) |
11 | 0.23 (0.59) | 2.07 (0.62) | 2.23 (0.48) |
12 | 0.75 (0.22) | 0.36 (0.10) | 0.09 (0.10) |
13 | 1.11 (0.28) | 2.16 (0.28) | 2.82 (0.36) |
14 | 2.28 (1.97) | 9.97 (1.93) | 6.16 (1.85) |
15 | 2.08 (0.13) | 4.80 (0.17) | 3.30 (0.20) |
16 | 6.56 (0.29) | 4.73 (0.09) | 2.75 (0.17) |
17 | 0.93 (0.35) | 2.67 (0.31) | 1.77 (0.18) |
18 | 2.13 (0.47) | 6.05 (0.48) | 1.74 (0.53) |
19 | 1.67 (0.66) | 3.99 (0.65) | 3.73 (0.75) |
20 | 0.13 (1.21) | 0.39 (0.93) | 0.13 (0.85) |
21 | 1.52 (0.03) | 2.40 (0.03) | 0.34 (0.03) |
22 | 7.90 (0.91) | 16.31 (0.89) | 8.19 (0.66) |
23 | 2.84 (1.64) | 3.62 (0.80) | 5.11 (0.96) |
24 | 8.50 (0.12) | 5.62 (0.06) | 4.00 (0.07) |
25 | 0.82 (2.61) | 2.55 (2.94) | 0.77 (2.89) |
26 | 0.01 (0.06) | 0.10 (0.07) | 0.04 (0.12) |
27 | 2.75 (0.60) | 9.94 (1.24) | 14.99 (1.20) |
28 | 3.37 (0.49) | 0.02 (0.47) | 0.28 (0.50) |
29 | 0.06 (1.67) | 0.02 (1.38) | 0.30 (1.44) |
30 | 0.16 (1.80) | 3.85 (2.28) | 10.81 (2.72) |
Maneuver Number | RHC | IIC | SIC |
---|---|---|---|
31 | 0.01 (0.50) | 0.82 (0.43) | 0.02 (0.47) |
32 | 5.81 (0.65) | 8.41 (0.24) | 3.63 (0.31) |
33 | 3.37 (1.86) | 9.55 (1.92) | 2.14 (2.03) |
34 | 0.54 (0.12) | 0.79 (0.10) | 0.91 (0.05) |
35 | 6.69 (0.61) | 12.14 (0.55) | 4.80 (0.50) |
36 | 5.20 (0.03) | 18.61 (0.06) | 4.50 (0.02) |
37 | 0.00 (0.13) | 0.81 (0.07) | 0.40 (0.03) |
38 | 1.47 (2.05) | 9.18 (2.28) | 0.15 (2.79) |
39 | 1.99 (1.18) | 2.31 (1.32) | 1.85 (2.01) |
40 | 9.04 (4.53) | 31.59 (6.18) | 14.21 (5.43) |
41 | 1.41 (0.34) | 0.17 (0.45) | 0.12 (0.49) |
42 | 0.81 (0.32) | 5.41 (0.16) | 0.51 (0.14) |
43 | 1.41 (0.38) | 0.45 (0.19) | 0.00 (0.10) |
44 | 0.60 (0.45) | 2.69 (0.15) | 2.09 (0.17) |
45 | 15.63 (2.20) | 7.84 (2.56) | 12.75 (2.69) |
46 | 3.20 (0.49) | 13.02 (0.16) | 9.23 (0.24) |
47 | 2.21 (1.70) | 2.78 (1.86) | 0.47 (1.42) |
48 | 0.00 (0.79) | 0.42 (0.69) | 0.22 (0.52) |
49 | 6.07 (0.47) | 1.82 (0.34) | 2.08 (0.38) |
50 | 0.98 (2.44) | 0.14 (1.66) | 0.54 (1.59) |
51 | 1.06 (0.56) | 11.19 (0.61) | 1.78 (0.54) |
52 | 0.21 (0.15) | 0.94 (0.06) | 0.00 (0.10) |
53 | 0.00 (1.85) | 1.93 (1.64) | 1.12 (1.68) |
54 | 0.00 (0.10) | 0.42 (0.17) | 0.00 (0.22) |
55 | 3.16 (0.25) | 3.50 (0.24) | 0.94 (0.28) |
56 | 4.30 (2.41) | 6.44 (2.17) | 9.45 (2.30) |
57 | 1.53 (2.88) | 3.18 (2.54) | 3.00 (2.45) |
58 | 4.19 (0.67) | 16.60 (0.34) | 16.40 (0.27) |
59 | 16.76 (1.07) | 13.56 (1.18) | 7.75 (0.95) |
60 | 0.83 (0.38) | 5.35 (0.17) | 1.66 (0.17) |
References
- Rupi, F.; Schweizer, J. Evaluating cyclist patterns using GPS data from smartphones. ITE Intell. Transp. Syst. 2018, 12, 279–285. [Google Scholar] [CrossRef]
- Broach, J.; Dill, J.; Gliebe, J. Where do cyclists ride? A route choice model developed with revealed preference GPS data. Transp. Res. Part A Policy Pract. 2012, 46, 1730–1740. [Google Scholar] [CrossRef]
- Ehrgott, M.; Wang, J.Y.; Raith, A.; van Houtte, C. A bi-objective cyclist route choice model. Transp. Res. Part A Policy Pract. 2012, 46, 652–663. [Google Scholar] [CrossRef]
- Dill, J. Bicycling for transportation and health: The role of infrastructure. J. Public Health Policy 2009, 30, S95–S110. [Google Scholar] [CrossRef]
- Rupi, F.; Poliziani, C.; Schweizer, J. Data-driven Bicycle Network Analysis Based on Traditional Counting Methods and GPS Traces from Smartphone. ISPRS Int. J. Geo-Inf. 2019, 8, 322. [Google Scholar] [CrossRef] [Green Version]
- Schweizer, J.; Rupi, F.; Poliziani, C. Estimation of link-cost function for cyclists based on stochastic optimisation and GPS traces. IET Intell. Transp. Syst. 2020, 14, 1810–1814. [Google Scholar] [CrossRef]
- Alonso, F.; Faus, M.; Cendales, B.; Useche, S. Citizens’ perceptions in relation to transport systems and infrastructures: Nationwide study in the Dominican Republic. Infrastructures 2021, 6, 153. [Google Scholar] [CrossRef]
- Willberg, E.; Tenkanen, H.; Poom, A.; Salonen, M.; Toivonen, T. Comparing spatial data sources for cycling studies: A review. In Transport in Human Scale Cities; Edward Elgar Publishing: Cheltenham, UK, 2021; pp. 169–187. [Google Scholar]
- Börjesson, M.; Eliasson, J. The value of time and external benefits in bicycle appraisal. Transp. Res. Part A Policy Pract. 2012, 46, 673–683. [Google Scholar] [CrossRef] [Green Version]
- Fioreze, T.; Groenewolt, B.; Koolwaaij, J.; Geurs, K. Perceived versus actual waiting time: A case study among cyclists in Enschede, The Netherlands. Transport Findings, 10 July 2019. [Google Scholar] [CrossRef]
- Rupi, F.; Poliziani, C.; Schweizer, J. Analysing the dynamic performances of a bicycle network with a temporal analysis of GPS traces. Case Stud. Transp. Policy 2020, 8, 770–777. [Google Scholar] [CrossRef]
- Strauss, J.; Miranda-Moreno, L.F. Speed, travel time and delay for intersections and road segments in the Montreal network using cyclist Smartphone GPS data. Transp. Res. Part D Transp. Environ. 2017, 57, 155–171. [Google Scholar] [CrossRef]
- Clarry, A.; Faghih Imani, A.; Miller, E.J. Where we ride faster? Examining cycling speed using smartphone GPS data. Sustain. Cities Soc. 2019, 49, 101594. [Google Scholar] [CrossRef]
- Laranjeiro, P.F.; Merchán, D.; Godoy, L.A.; Giannotti, M.; Yoshizaki, H.T.; Winkenbach, M.; Cunha, C.B. Using GPS data to explore speed patterns and temporal fluctuations in urban logistics: The case of Sao Paulo, Brazil. J. Transp. Geogr. 2019, 76, 114–129. [Google Scholar] [CrossRef]
- Cortés, C.E.; Gibson, J.; Gschwender, A.; Munizaga, M.; Zúñiga, M. Commercial bus speed diagnosis based on GPS-monitored data. Transp. Res. Part C Emerg. Technol. 2011, 19, 695–707. [Google Scholar] [CrossRef]
- Poliziani, C.; Rupi, F.; Schweizer, J. Traffic surveys and GPS traces to explore patterns in cyclist’s in-motion speeds. Transp. Res. Procedia 2021, 60, 410–417. [Google Scholar] [CrossRef]
- Poliziani, C.; Rupi, F.; Schweizer, J.; Saracco, M.; Capuano, D. Cyclist’s waiting time estimation at intersections, a case study with GPS traces from Bologna. Transp. Res. Procedia, 2021; in press. [Google Scholar]
- Poliziani, C.; Rupi, F.; Mbuga, F.; Schweizer, J.; Tortora, C. Categorizing three active cyclist typologies by exploring patterns on a multitude of GPS crowdsourced data attributes. Res. Transp. Bus. Manag. 2021, 40, 100572. [Google Scholar] [CrossRef]
- Tortora, C.; Franczak, B.C.; Browne, R.P.; McNicholas, P.D. A mixture of coalesced generalized hyperbolic distributions. J. Classif. 2019, 36, 26–57. [Google Scholar] [CrossRef] [Green Version]
- Betterpoints. Available online: https://www.betterpoints.ltd/ (accessed on 27 February 2022).
- Alonso, F.; Useche, S.; Faus, M.; Esteban, C. Does Urban Security Modulate Transportation Choices and Travel Behavior of Citizens? A National Study in the Dominican Republic. Front. Sustain. Cities 2020, 2, 42. [Google Scholar] [CrossRef]
- SUMOPy. Available online: https://sumo.dlr.de/docs/Contributed/SUMOPy.html (accessed on 27 February 2022).
- OSM. Available online: https://www.openstreetmap.org/#map=19/44.50163/11.34276 (accessed on 27 February 2022).
- Schweizer, J.; Bernardi, S.; Rupi, F. Map-matching algorithm applied to bicycle global positioning system traces in Bolognaa. ITE Intell. Transp. Syst. 2016, 10, 244–250. [Google Scholar] [CrossRef]
- Fraboni, F.; Marín Puchades, V.; De Angelis, M.; Pietrantoni, L.; Prati, G. Red-light running behavior of cyclists in Italy: An observational study. Accid. Anal. Prev. 2018, 120, 219–232. [Google Scholar] [CrossRef]
- Tengattini, S.; Bigazzi, A.; Rupi, F. Appearance and behaviour: Are cyclist physical attributes reflective of their preferences and habits? Travel Behav. Soc. 2018, 13, 36–43. [Google Scholar] [CrossRef]
- Rossi, R.; Mantuano, A.; Pascucci, F.; Rupi, F. Fitting time headway and speed distributions for bicycles on separate bicycle lanes. Transp. Res. Procedia 2017, 27, 19–26. [Google Scholar] [CrossRef]
- Highway Capacity Manual; Transportation Research Board: Washington, DC, USA, 2000.
- Schweizer, J.; Poliziani, C.; Rupi, F.; Morgano, D.; Magi, M. Building a Large-Scale Micro-Simulation Transport Scenario Using Big Data. ISPRS Int. J. Geo-Inf. 2021, 10, 165. [Google Scholar] [CrossRef]
- R Core Team. R: A Language and Environment for Statistical Computing; R Foundation for Statistical Computing: Vienna, Austria, 2020. [Google Scholar]
- Breiman, L. Random forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Liaw, A.; Wiener, M. Classification and Regression by randomForest. R News 2002, 2, 18–22. [Google Scholar]
- Boser, B.E.; Guyon, I.M.; Vapnik, V.N. A training algorithm for optimal margin classifiers. In Proceedings of the Fifth Annual ACM Workshop on Computational Learning Theory, Pittsburgh, PA, USA, 27–29 July 1992; pp. 144–152. [Google Scholar]
- Meyer, D.; Dimitriadou, E.; Hornik, K.; Weingessel, A.; Leisch, F.; Chang, C.C.; Lin, C.C.; Meyer, M.D. Package ‘e1071’. The R Journal version 1.7-3. 2019. Available online: https://rdrr.io/rforge/e1071/ (accessed on 27 February 2022).
- Nadaraya, E.A. On estimating regression. Theory Probab. Appl. 1964, 9, 141–142. [Google Scholar] [CrossRef]
- García-Portugués, E. Notes for Nonparametric Statistics, version 6.4.4; Bookdown, 2021; ISBN 978-84-09-29537-1. Available online: https://bookdown.org/egarpor/NP-UC3M/ (accessed on 27 February 2022).
- Hayfield, T.; Racine, J.S. Nonparametric Econometrics: The np Package. J. Stat. Softw. 2008, 27, 1–32. [Google Scholar] [CrossRef] [Green Version]
- Aitchison, J.; Aitken, C.G. Multivariate binary discrimination by the kernel method. Biometrika 1976, 63, 413–420. [Google Scholar] [CrossRef]
- Li, Q.; Racine, J.S. Nonparametric Econometrics: Theory and Practice; Princeton University Press: Princeton, NJ, USA, 2007. [Google Scholar]
- Li, Q.; Racine, J. Cross-validated local linear nonparametric regression. Stat. Sin. 2004, 14, 485–512. [Google Scholar]
- Hardle, W.; Marron, J.S. Bootstrap simultaneous error bars for nonparametric regression. Ann. Stat. 1991, 19, 778–796. [Google Scholar] [CrossRef]
Feature | Critical Volume | Average PCE Flow | Length |
---|---|---|---|
Critical Volume | 1.000 | 0.074 | 0.433 |
Average PCE Flow | 0.074 | 1.000 | −0.188 |
Length | 0.433 | −0.188 | 1.000 |
Feature Number | 1 | 2 | 3 | 4 | 5 | 6 | 7 |
---|---|---|---|---|---|---|---|
1 | 0.47 | 0.03 | 0.56 | 0.73 | 0.68 | 0.67 | |
2 | 0.24 | 0.38 | 0.42 | 0.31 | 0.92 | ||
3 | 0.23 | 0.25 | 0.15 | 0.63 | |||
4 | 1.33 | 1.08 | 0.75 | ||||
5 | 1.55 | 0.98 | |||||
6 | 0.84 | ||||||
7 |
Feature Name | Description |
---|---|
Maneuver typology (1) | Nominal; left turn, right turn, straight |
Lanes Edge to (2) | Ordinal; number of lanes on road which maneuver is directed to |
Traffic Light (3) | Nominal; presence of traffic light |
Number Maneuvers Crossed (4) | Ordinal; number of intermediate maneuvers crossed |
Connections Node (5) | Ordinal; total number of maneuvers at geographic intersection |
Critical Volume (6) | Continuous; amount of opposing traffic |
Average PCE Flow (7) | Continuous; amount of passenger car traffic |
Length (8) | Continuous; length of maneuver in meters |
Lower Bound | Upper Bound | n | |
---|---|---|---|
Class 1 | 0.00 | 0.75 | 20 |
Class 2 | 0.75 | 2.50 | 19 |
Class 3 | 2.50 | N/A | 21 |
Feature Number | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
---|---|---|---|---|---|---|---|---|
RHC | 0.1946 | 1.0000 | 0.5000 | 0.0006 | 0.0305 | 194.0834 | 292.8674 | 2.2087 |
IIC | 0.6667 | 0.7447 | 0.0134 | 0.0297 | 0.0322 | 0.0000 | 49.5347 | 4.0687 |
SIC | 0.3364 | 0.1337 | 0.5000 | 0.0686 | 0.0454 | 250.8760 | 45.8320 | 3.0427 |
Mean Average Deviation | |||
---|---|---|---|
RHC | 0.9944 | 0.2550 | 0.1019 |
IIC | 0.9901 | 0.5942 | 0.2408 |
SIC | 0.9955 | 0.2721 | 0.0718 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Walker, J.; Poliziani, C.; Tortora, C.; Schweizer, J.; Rupi, F. Nonparametric Regression Analysis of Cyclist Waiting Times across Three Behavioral Typologies. ISPRS Int. J. Geo-Inf. 2022, 11, 169. https://doi.org/10.3390/ijgi11030169
Walker J, Poliziani C, Tortora C, Schweizer J, Rupi F. Nonparametric Regression Analysis of Cyclist Waiting Times across Three Behavioral Typologies. ISPRS International Journal of Geo-Information. 2022; 11(3):169. https://doi.org/10.3390/ijgi11030169
Chicago/Turabian StyleWalker, Jeremy, Cristian Poliziani, Cristina Tortora, Joerg Schweizer, and Federico Rupi. 2022. "Nonparametric Regression Analysis of Cyclist Waiting Times across Three Behavioral Typologies" ISPRS International Journal of Geo-Information 11, no. 3: 169. https://doi.org/10.3390/ijgi11030169
APA StyleWalker, J., Poliziani, C., Tortora, C., Schweizer, J., & Rupi, F. (2022). Nonparametric Regression Analysis of Cyclist Waiting Times across Three Behavioral Typologies. ISPRS International Journal of Geo-Information, 11(3), 169. https://doi.org/10.3390/ijgi11030169