Smart Cities and Mobility Systems: How Fair are Data Models and Algorithms

Like in many industries, contemporary transportation practices often combine intelligent technology systems and algorithms, to optimize and scale operations, and build functional efficiencies. Smart cities and mobility systems are built on the combining multiple technology systems and business intelligence and analytics applications to optimize the delivery of transportation services.

In transportation operations, the availability and use of technology and data is an added-value to service delivery. But, technology on its own, and by itself is only a tool. As Cathy (2016) pointed out in her book, models are opaque, may be encoded with human prejudice, misunderstanding, and bias that are built into the software systems.[1] So, because of the fallibility of data and technology systems, it is imperative for systems engineers to designed to have an iterative feedback loop, with the ability to learn, actively calibrated and recalibrate (hone and fine-tune) data algorithms.

Overall, descriptive and predictive models have their inherent limitations, including the potential to make bias and adverse abstract representations of data. In practice, we must recognize that technology systems are built in the context of our human experience, taking into account our conditional or unconscious social, political and economic biases. In a “just city” context, quality or good data is not equal to facts. The obscured connection between economic innovation and diversity may just be a “staged authenticity”.[2] As such, it is worthy to consider data biases and potential injustices that data and related technology systems may perpetuate.

Work Cited

[1] O’Neil, C. (2017). Weapons of math destruction: How big data increases inequality and threatens democracy. Broadway Books.

[2] Fainstein, S. S. (2005). Cities and diversity: should we want it? Can we plan for it?. Urban affairs review, 41(1), 3-19.