Ding | Devpost

Our Inspiration:

Last year 4,593 VAWG offences were recorded across the transport network, only around 3% resulted in a charge or summons, while around 58% of cases identified no suspect. [Source: Written evidence submitted to the Police and Crime Committee by British Transport Police.] Unfortunately, this is also only the recorded number of offences. From our own personal experiences as women, safety when travelling is something we always have to take into consideration. Whether we are travelling back from a night out or going home after a late lecture, it's not ideal when Google Maps identifies the best or quickest route and it includes walking through a dark alleyway. As students we also frequently have to weigh up the cost, time and safety every single time you travel which takes up time.

What it does:

Our project is an app that calculates the safest route from location A to location B where our target audience is women. There will be 2 main categories: safety and travel time. The user can enter in safe areas which will be stored on Vultr and as local storage. Although we are not able to include every single one, we have considered several factors to determine the safest route. By weighing these factors, sourcing the public data and combining these factors we get a safety ranking of different possible routes. The user can then choose which route to take, avoiding any pubs or clubs indicated by a tiny flashing light on the map. As of right now, we are focusing on London only and walking routes for the hackathon.

How we built it:

We developed Ding, a walking safety navigation app by first designing an intuitive user interface and implementing user authentication. We integrated mapping and routing APIs, including Google Maps, Google Routes and OpenRouteService to generate walking routes between locations. We then built a database to store safety-related features, their weightings and route information. Using this data, we developed a scoring algorithm to evaluate routes based on safety factors such as crime, street lighting and nearby open businesses. Our rerouting algorithm identifies sections that fall below a minimum safety threshold and generates alternative routes to avoid them. Finally, we implemented route comparison, colour-coded safety ratings, live navigation and rerouting functionality to help users choose routes that balance safety and journey time.

The challenges we ran into:

One of our biggest challenges was developing a way to quantify walking safety as it's subjective and depends on several different factors. We also had to integrate multiple APIs, manage incomplete or inconsistent location data and balance route safety against journey time. [Designing an algorithm that could identify lower-scoring sections and generate alternative routes without creating excessively long detours was particularly challenging.]

Accomplishments that we're proud of:

Ding addresses an extremely common problem that women face, particularly when walking alone. A key achievement was developing a transparent, data-driven safety scoring system that considers crime, street lighting and nearby open businesses. We're also proud of our approach to rerouting which prioritises avoiding potentially concerning sections.

What we learned

This project taught us how to integrate mapping APIs, work with geospatial data and develop algorithms that balance multiple competing factors. [We gained experience with databases, user authentication and route optimisation while learning how important it is to account for missing data and the limitations of safety predictions.]

What's next for Ding!

Our next step is to develop Ding into a fully functional mobile app that also takes into account public transport and expand our coverage beyond London. We also hope to improve our safety scoring model using more comprehensive data and user feedback, introduce real-time safety updates and further develop personalised routing based on individual preferences. Ultimately, we want to make travelling feel safer and easy to access.