As communicators, we use abstractions on a daily basis to refer to concepts, systems, thoughts and things. Ever put on your ‘PJ’s’ before bed, or had a ‘cuppa Joe’ in the morning? If so, you’re familiar with abstractions. In the data world, we use abstractions to distill complex concepts into more manageable ideas and we often refer to these abstractions as models.
The word ‘model’ has many different meanings. It can, among other things, refer to three-dimensional representations of things, a person employed to display clothes by wearing them or even a simplified description of a system or process. Models are everywhere and they vary in complexity but they all serve to provide a better or more concise understanding of the thing that they are describing.
What a model is can often depend on who you are:
- If you are a mathematician, you may consider that rounding a number to three decimal places creates a model of the original number that had 20 decimal places.
- If you’re a statistician you may see the relationship between peoples’ height and their weight as a model – if you’re assuming that taller people are on average heavier then you’re using a model in your head to describe the relationship between height and weight.
- If you’re a town planner, you probably use Geographic Information Systems to create maps of potential developments, including their spatial extent, height and relationships to other infrastructure. This digital representation of the development (and for that matter any map) is a model that helps you to understand the system.
- If you’re a computer scientist, you may consider the simulation of an entire city as a model, enabling you to estimate the impact of changing street networks or changing bus timetables on the movement of people throughout the city.
Although these examples differ, they all demonstrate the abstraction of a more complex concept into a simpler idea. Being able to make these abstractions helps us to communicate ideas both within our domain and between domains.
Being able to generate models that use data from different domains often leads to new insight. As a result, interdisciplinary models have the potential to break down silos and add value to city experimentation and the use of data in cities. For example, air quality is not only related to the natural environment, it is also closely linked to transportation and fuel use. Understanding these relationships requires the incorporation of data from both the transport and environmental domains.
The Tombolo Digital Connector is a tool that makes it easier to combine data from different domains and generate models of different complexity depending on your needs. For example, we’ve generated an ‘Active Transport Index’ which is a model that combines data about street networks and cycle counts to summarise how ‘cycle friendly’ different parts of the UK might be. If you’re interested in building your own models, and want to have a go at using data from different domains, consider looking more closely at our Tombolo Digital Connector documentation and let us know what you think.
Try out the Digital Connector on GitHub!
If you’d like to learn more or work with us to generate models that are interesting to you please get in touch!