Because these IoT-based sensors gather different types of data, often in different formats and sometimes vendor-specific, the successful integration of these diverse datasets and data feeds can be quite challenging. In addition, their coalescence within decision support systems and other analytics platforms is paramount to the usefulness of the data for data products development that support improved decision-making.Other challenges include: 

  • There are few unified standards that address the many possible aspects of a smart city program.
  • Smart cities are built on pervasive, open networking and cheap, disposable hardware (i.e., system of systems), which is the opposite of how cities and states currently develop or procure their information systems (i.e., traditional system design). This is new territory for most transportation agencies and can lead to differing, and sometimes competing, ideas on data architecture solutions, and methodologies.
  • It is difficult to compare the costs of on-premise options with cloud-based architecture. Because cloud-based technology fees are based on use, this cost-benefit comparison is nearly impossible to make without first knowing how much data will be collected and how frequently it will be queried. Furthermore, no data exists to baseline the cost of the new system, as there is no equivalent on-premise system using that much data.
  • Necessary funding levels for projects can be difficult to define if information about data volumes and access frequency are not yet known.
  • The lack of an “umbrella agreement” available creates difficulties in sharing the data. This can result in cities having to create new data sharing agreements for any new data request. This approach is increasingly frustrating and inefficient, both for data requesters and city staff.

Referencing Page: