Enabling the smart worker

plus10 approached us with the concept of making an app for operators of a manufacturing line. The background is that every line has a theoretical target output that is realistically not reachable because of different reasons: Materials have to be refilled, machine parts constantly break or get overheated and repairing personnel is not omnipresent. This means that machines are running constantly below their potential. To optimize this plus10 created an algorithm that can analyze and predict shortages or malfunction.

Our job was to find the best possible way to get this information to the workers.

Kicking it off with the team

The plus10 team has a very diverse skillset – from engineer to machine-learning specialist. But what all of them have in common is their huge experience in working together with regular operators, maintenance staff or plant operators. Because we wanted to build up on this experience we arranged a one-day workshop with the whole team. In multiple sessions we used different Design Thinking techniques like Charetting, creating a persona or putting together a user journey map to better understand how a day in the life of a worker looks like.

Finding a way

Out of the workshops findings we created a flowchart with all features that we believed to be helpful for the user. Together with the plus10 team we narrowed this down to essential features for a MVP. In multiple iterations we distilled these features into basic wireframes, which served as the abstract base for the upcoming design.

Key Screens – Timeline

The core of the app is the timeline, where you see all events that relate to your job position and plant. They are ordered by relevance by the smart algorithm from plus10. From here you can reach all open or closed items, filter the list or access your personal settings via the header.

Key Screens – Onboarding

Most workers change their job frequently because they are hired via temporary work agencies. This makes a self-explaining app even more important than usual. Because of this we chose to implement a quick on boarding flow that asks for basic information and provides short explanation for key features.

Key Screens – Task

Every item in the timeline is a task to be resolved by the worker. A task features various informations such as when and where an incident happened, which solutions would most probably resolve this issue and previous comments by other workers with the same task.

Key Screens – Feedback

If an item consists of an unknown incident or if a worker cancels a task the system asks back for information. This is necessary to give the underlying algorithm input for matching signal data to human-understandable output.