We are on the verge of digital society which comes with the omni-presence of information and communication technology. Modern communication technology also led to new platform economies. Platforms easily bring together different stakeholders in an integrated setting and assist in organizing business transactions. Car-sharing services are nowadays often based on a platform that keeps track of available cars and allows subscribers to make reservations and pay through traditional bank and credit card channels and modern equivalents such as PayPal, bitcoin, Apple pay, etc. Ride sharing services such as Uber are based on the same technology and business principle. The latest development in this context is the Mobility-as-a-Service concept, in which subscribers can use different transportation modes available in the bundle using discounted fares.
In order to operate the above innovative mobility options optimally, stakeholders collect data of different kinds on use and position of their transport supplies and movement patterns of citizens by gadgets installed on the vehicles as well as tracking smart cards swiping. Similarly, city authorities collect various types of data over time such as traffic flow, emission and so on. Likewise, telecommunication providers have massive amount of data on mobile internet usage and calls which can also be used to shed light on the mobility of citizens. Since these data are typically collected over a long period of time and lacks detailed information, they are sometimes referred to as 'long and thin' data.
While a tremendous amount of data is collected on the daily basis, without knowledge of how to use this data for planning, design and operating of our built environment in more efficient ways and along citizens' need, the effort in collecting those data will be in vein. Traditional travel demands forecasting models have typically been developed on travel surveys collected at the national level. Such data collection is becoming increasingly obsolete due to the high cost and low response rate especially when long term data (multiple days/ weeks) is desired. In this context it is essential to leverage big data in updating such demand forecasting models by exploiting its strength and avoiding its shortcomings (lack of details).
The use of mobility big data for planning and forecasting purposes requires dedicated knowledge on data analysis methods (data mining, deep learning, unsupervised and semi-supervised learnings) and their relations with the statistical methods used traditionally for the analysis of survey data (discrete choice analysis, conjoint analysis). Such knowledge can purposedly contribute to the development or update of decision support tools. Tools which can serve multitude of purposes, from long term to mid and short term planning. Optimum locations of newly built charging stations, shared vehicles relocation strategies, optimum dynamic pricing for electricity, demand for the first and last mile mobility options in high temporal and spatial resolution, among others, can then be devised, supported by a data driven decision tools. To be able to realize the above purposes, the candidate should have a proven background on the following areas:
- Data mining methods to analyse mobility big data
- Travel demand forecasting models, including tour based and activity based models
- Discrete choice methods and conjoint analysis
- Stated choice experiment design
In addition to research, the candidate is expected to be highly involved in the educational tasks. The candidate is expected to contribute to teaching and supervision of courses in the Bachelor of Architecture, Urbanism and Building Sciences (AUBS) and Master of Architecture, Building and Planning (APB). Courses such as 'Big data for urban analysis', 'Smart urban environments' and 'mobility and logistic' among others need to be supported by the candidate. To that end, the applicant must have a track record in teaching and supervising of Bachelor and Master students.