Estimation of daily bicycle traffic volumes using sparse data

Calculating the annual average daily bicycle (AADB) volume at a particular cycling facility requires the availability of year-round daily volume data. Automatic counters (e.g., loop detectors) used to collect such continuous data are subject to periodic malfunctions, leading to sporadic data gaps. This problem could affect the calculated values of the AADBs and impact the estimates of the daily and monthly adjustment factors at these count stations. The impacts become even more significant if the data gaps take place frequently and/or for long periods. This research tackles the problem of missing cycling traffic volumes at count locations that potentially experience frequent sensor malfunctions during the year. The method is also applicable to any other similar research problem (e.g. missing volumes at vehicle count locations). A data-driven, yet novel, model is proposed to estimate missing volumes at some locations, using data from other nearby count locations as well as the historical volumes of the same location. The model is motivated by the spatial–temporal relationship of cycling volumes of similar nearby facilities. The proposed model is dynamic as it assumes no prior knowledge about which locations may experience sensor malfunction (i.e., missing volumes). The model is referred to as the “autoencoder neural network” and it belongs to the family of Artificial Neural Networks (ANNs). This model expresses the relationship between a vector of input variables and itself. Hence, if a daily bicycle count is available on one day at a specific location, it will be used as a model input; whereas, a missing daily volume will be treated as an output variable that needs to be determined. The model was tested using a large dataset of about 13,000 daily bicycle volumes from the City of Vancouver, Canada. The data were collected between 2009 and 2011 at 22 different count locations. The model showed a strong estimation power with an average error of about 10%. Sensitivity analyses were carried out to investigate the impact of different model parameters on the estimation accuracy. The optimum set of model parameters was consequently defined.

[1]  A. Matzarakis,et al.  Weather and cycling—a first approach to the effects of weather conditions on cycling , 2007 .

[2]  Jeffrey S. Wilson,et al.  Estimating urban trail traffic: Methods for existing and proposed trails , 2007 .

[3]  Mohamed El Esawey,et al.  Travel time estimation in urban networks using limited probes data , 2011 .

[4]  Pascal Vincent,et al.  Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion , 2010, J. Mach. Learn. Res..

[5]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[6]  Greg Lindsey,et al.  Day-of-Year Scaling Factors and Design Considerations for Nonmotorized Traffic Monitoring Programs , 2014 .

[7]  C. F. Jaarsma,et al.  Temporal variations of bicycle demand in the Netherlands: The influence of weather on cycling , 2009 .

[8]  S. Tin Tin,et al.  Temporal, seasonal and weather effects on cycle volume: an ecological study , 2012, Environmental Health.

[9]  Paul J. Werbos,et al.  Backpropagation Through Time: What It Does and How to Do It , 1990, Proc. IEEE.

[10]  Amy Lewin.  Temporal and Weather Impacts on Bicycle Volumes , 2011 .

[11]  Mohamed El Esawey,et al.  Determination and Application of Standard K Factors for Bicycle Traffic , 2015 .

[12]  Yoshua Bengio,et al.  Extracting and composing robust features with denoising autoencoders , 2008, ICML '08.

[13]  Christopher M. Monsere,et al.  Methodology to Characterize Ideal Short-Term Counting Conditions and Improve AADT Estimation Accuracy Using a Regression-Based Correcting Function , 2014 .

[14]  M. Nankervis,et al.  The effects of weather and climate on bicycle commuting , 1999 .

[15]  Greg Lindsey,et al.  Adjustment Factors for Estimating Miles Traveled by Nonmotorized Traffic , 2013 .

[16]  D. Niemeier.  Longitudinal Analysis of Bicycle Count Variability: Results and Modeling Implications , 1996 .

[17]  Yoshua Bengio,et al.  Greedy Layer-Wise Training of Deep Networks , 2006, NIPS.

[18]  T. Sayed,et al.  Development of Daily Adjustment Factors for Bicycle Traffic , 2013 .

[19]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[20]  Wesley E. Marshall,et al.  Estimating Annual Average Daily Bicyclists , 2013 .

[21]  Andres F. Clarens,et al.  Estimating Spatially and Temporally Continuous Bicycle Volumes by Using Sparse Data , 2014 .

[22]  Xize Wang,et al.  Estimating use of non-motorized infrastructure: Models of bicycle and pedestrian traffic in Minneapolis, MN , 2012 .

[23]  Robert J. Schneider,et al.  Pilot Models for Estimating Bicycle Intersection Volumes , 2011 .

[24]  Jinhua Zhao,et al.  A Seasonal Autoregressive Model Of Vancouver Bicycle Traffic Using Weather Variables , 2011 .

[25]  Luis F. Miranda-Moreno,et al.  Classification of Bicycle Traffic Patterns in Five North American Cities , 2013 .

[26]  Yoshua. Bengio,et al.  Learning Deep Architectures for AI , 2007, Found. Trends Mach. Learn..

[27]  Mohamed El Esawey,et al.  Estimation of Annual Average Daily Bicycle Traffic with Adjustment Factors , 2014 .

[28]  Luis F. Miranda-Moreno,et al.  Weather or Not to Cycle , 2011 .

[29]  Razvan Pascanu,et al.  Theano: new features and speed improvements , 2012, ArXiv.

[30]  C. Jakob,et al.  Quantifying and comparing the effects of weather on bicycle demand in Melbourne (Australia) and Portland (USA) , 2011 .

[31]  Luis F. Miranda-Moreno,et al.  Cycling and Weather: A Multi-city and Multi-facility Study in North America , 2012 .

[32]  Susan Hanson,et al.  EVALUATING THE IMPACT OF WEATHER ON BICYCLE USE , 1977 .