[R-sig-Geo] Spatio-temporal modelling for trip generation data

Rafael Pereira r@|@@pere|r@@br @end|ng |rom gm@||@com
Wed Sep 4 19:43:54 CEST 2019

Hi Ariel,

There are many open source options to conduct routing analysis, which are
used into accessibility modeling. One of the most used alternatives
nowadays is called OpenTripPlanner
<https://github.com/opentripplanner/OpenTripPlanner>. I've written a couple
Python scripts to call OTP e generate travel-time estimates from many
points to many points using different transport modes including public
transport (https://github.com/rafapereirabr/otp-travel-time-matrix). Marcus
Youg and others have developed a R package (opentripplanner
https://github.com/ITSLeeds/opentripplanner) that allows one to call OTP
from within R. Highly recommended! I hope this is helpful in some way.

best wishes,

Rafael H M Pereira

On Wed, Sep 4, 2019 at 2:25 PM Rich Shepard <rshepard using appl-ecosys.com>

> On Wed, 4 Sep 2019, Ariel Fuentesdi wrote:
> > I want to produce a prediction of trip generation for the next year, I'm
> > thinking could be useful a Spatio-temporal model with covariates data. I
> > have trips values in point locations since 2015 also the population in
> > census tracts and supply which are bus lines and I'm thinking to convert
> > them into an accessibility grid (from the census tracts to bus stops). I
> > have seen that Spatio-temporal models are applied to environmental
> > problems only, would you recommend me any specific approach or the use of
> > a specific model?
> Ariel,
> Your description suggests you would like to solve the classic 'traveling
> salesman' problem. I don't know that R's spatio-temporal functions are
> suitable but open street map apparently can help pick optimal routes and a
> web search for linux: traveling salesman problem found many software
> solutions.
> Travel safely,
> Rich
> _______________________________________________
> R-sig-Geo mailing list
> R-sig-Geo using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo

	[[alternative HTML version deleted]]

More information about the R-sig-Geo mailing list