Amazon Transportation
Internship Project
Summer 2018 - Spring 2019
6 Months
UX Research
UX Design
Data Visualisation
Fullstack Web Development
I spent 6 months working in the Amazon Transportation EU scheduling team, a team responsible for scheduling truck deliveries between Amazon warehouses across Europe. I was placed into a team working on a system which uses machine learning to predictively schedule around ten thousand journeys a week between hundreds of sites. My role was initially focused on front-end software engineering, however after observing the use of the system and talking to users I identified the opportunity for significant improvements to the UI and took on additional UI/UX design responsibilities. The users of our system were producing week-by-week plans of truck schedules. Particularly during "Peak" season (the build up to Christmas) this was a very complex, stressful and time-pressured task. I focused on identifying the pain-points that existed within the system that were causing frustration to users and wasteing precious time. I worked closely with users to help improve the usability of the system and ensure that new features were intuitive, working closely to the Semantic UI design system. I also worked with users to help streamline other processes they carried out, often using time consuming, error-prone techniques such as cross-referencing excel spreadsheets. I worked to develop a set of dashboards that allow users to access an instant visualisation of various metrics. Previously accessing these metrics would require complex data analysis and coding. Providing these dashboards gave the users across the company a window into the data behind the process so they could focus their time on optimisation. I designed the implementation of the dashboards to be very easily scalable. I created a model of a skeleton dashboard allowing new dashboards to be configured made with minimal coding. The key design challenges I faced were presenting large amounts of data and many complex options to the user in a clean, simple user experience.
Full case study coming soon...