Skip to content

The model that has been uploaded to this repository aspires to describe routing behavior of micro-mobility modes, e.g., e-bikes and e-scoters, in relationship with traditional modes, e.g., private car and walking.

License

Notifications You must be signed in to change notification settings

lotentua/Perceived_safety_choices

Repository files navigation

The PERCEIVED SAFETY CHOICES model

While safety seems to be a significant factor when choosing to use these new modes, this model utilizes the notion of perceived safety to model travel behavior in inner urban areas. Therefore, the developed model is built on the hypothesis that perceived safety affects travel behavior of micro-mobility services users and is related to road environment. It combines ordinal logistic regression model, which predict perceived safety in different road environments using a 7-point Likert Scale, with discrete choice or simulation models which simulate the mode/route choices. The input variable is the road network which consists of links and nodes. The conceptual model of the Perceived_safety_choices is presented below:

The different functions of the model are parametric to take into account variations in "tastes" among individuals by proposing advance modeling techniques. All the parametric can be calibrated by collecting data related to safety perceptions considering various road environments with mixed traffic conditions and first/last mile mode/route choices in each urban area that are used. The repository contains example datasets and default models that can be used.

Based on this concept, the Perceived_safety_choices proposes some tools in order to investigate the overall impact of perceived safety on travel behavior, transport equity and transport system sustainability. There is a continuous development of these tools by the NTUA research team and external partners who still commit.

Lastly, Perceived_safety_choices creates a path to combine agent-based transport modeling (e.g., MATSim) with spatial analysis and GIS tools. The new scoring function is estimated based on the Value-of-Safety (VoS) which refers to how many kilometers of less travelling a road users is willing to exchange in order to experience a better safety level.

The Perceived_safety_choices repository contains:

  • Psafechoice: contains tools to import a shapefile with the links and nodes, estimate traffic parameters and perceived safety per link, export csv and xml files for further analysis and a routing model based on Djikstra algorithm which defines the shortest, fastest and safest path per transport mode; to do so, the Dijkstra package is utilized; yet the weights change.
  • empirical: contains advance modeling techniques based on PandasBiogeme and Rchoice package to compute first/last mile route/mode choice models.
  • scenario_athens: is an example scenario where Psafechoice functions are utilized to investigate the impact of perceived safety on route choices in the road network of the city center of Athens. Four transport modes are considered: car, e-bike, e-scooter and walk.
  • EcarGobikes: contains a multi-depot optimization model to design (flexible) first/last mile logistics service using e-bikes and inexperienced crowdshippers. The design takes into account perceived safety ratings per link.

+++ A contribution to MATSim called "Psafe Module" is under development. In essence, it is an updated version of Bicycle Module following a more universal approach fully based on perceived safety parameter and covering all first/last mile modes. A randomized marginal utility for the perceived safety parameter is applied in this model. The model is firstly tested using the experimental scenario in Athens.

To install the PsafeChoices package (ONLY) please type:

pip install git+https://github.com/lotentua/Perceived_safety_choices

Unfortunately, requirements list in the current version of the package had not fully defined! Some of them are contained in the: requirement.txt

The tools contained in this repository were developed within various research project of LoTE.

About

The model that has been uploaded to this repository aspires to describe routing behavior of micro-mobility modes, e.g., e-bikes and e-scoters, in relationship with traditional modes, e.g., private car and walking.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published