Why Real-Time Data Are Not Used to Improve Urban Systems?

Posted: March 12th, 2008 | No Comments »

In a discussion at the MIT Responsive Ctiy meeting with Prof. Nigel H.M. Wilson who presented current research in the analysis of real-time transportation data for understanding users behaviors, the question was raised to know why these data where not massively used nowadays (related to what Christopher Zegras discussed last week). Many transportation agencies still rely on expensive surveys with a limited statistical significance and overlook large automatically collected data their system provide. Nigel answered that first, these agencies want to make sure these data bring at least the same amount of knowledge than their “manual” data (e.g. surveys). Second, they do not necessarly have the skills and knowledge to analyze and grasp the significance of these data (which is my feeling when it comes to Bicing or Velib usage data). We seem to be in a chicken and egg situation here. Finally, there is a people concern that these data will be used to their detriment. For instance, it could reveal that the number of routes in certain areas should be decreased, or that user could wait longer for connexions. In other words, these data do not always lead to positive and favorable decisions for certain users.

Relation to my thesis: A major issue in leveraging people generated content to reveal new aspects of the city is about validating and proving the significance of these data with already existing analysis techniques.