News

TECNALIA develops predictive models for analysing urban pedestrian mobility in conjunction with Dinycon

8 July 2021
Tecnalia participa en Bidemap para el desarrollo de modelos predictivos de la movilidad urbana peatonal

TECNALIA provides its experience to build the analytical layer of the data collected through its Smart Mobility Lab, a mobility behavioural analysis laboratory based on the Big Data paradigm.

TECNALIA is developing a tool to help municipal managers analyse aggregate pedestrian mobility behaviour in smart cities as part of the Bidemap initiative, led by Dinycon.

This study consists of developing predictive urban pedestrian mobility models - occupation of public spaces, mobility flows, origin-destination matrices and heat maps. It is intended to provide support to city managers in the context of smart cities.

In order to build these services from the mobility data collected by Dinycon, it is necessary to create data analysis solutions to detect patterns, trends, describe behaviours and predict them, and to translate them into attractive and interpretable visual formats for city managers. Digital data that is difficult to handle is converted into information that is enriched and can be interpreted by human operators.

TECNALIA provides its experience to build the analytical layer of the data collected through its Smart Mobility Lab, a mobility behavioural analysis laboratory based on the big data paradigm. Short- and long-term predictive urban pedestrian mobility models are generated from historically recorded data and data ingested in real time.

The short-term value prediction method uses a window of readings prior to the value as inputs. These data make it possible to play with different variables: depth, prediction horizon and contextual information.

The long-term predictive model is based on detecting patterns in the known data, grouped by days, other external factors and similarity. These patterns are used as the basis for new techniques for improving data quality, significantly reducing the need to monitor the collection process.

TECNALIA is currently exploring the development of virtual sensing techniques. It uses prior knowledge to generate plausible data at network locations without sensors by establishing correlations between sensors, thereby minimising the need for massive sensor deployment over time.

Further information

Dinycon has extensive experience in acquiring quality data, and TECNALIA specialises in analysing massive volumes of data in the field of mobility.

The Bidemap initiative is funded by the Hazitek R&D aid programme.