Forestry plays an extremely important role in Sweden's national economic and social development. Forest industry not only directly employs logging and wood processing positions, but also indirectly supports the whole value chain, such as wood processing, transportation, research and development, which provides work for many people.
Failure and downtime of heavy logging equipment can cause unquantifiable losses of time and money in a productivity-oriented industry such as logging. And the equipment repairman takes on such a responsible role for their role is to reduce customers’ losses.
However, due to the complexity and sophistication of logging equipment, there is a very high barrier to entry and communication costs to become a mechanic. And lack of a standardized solution for both training and on-the-ground repairs results in high career barrier and shortage of labour force.
In this project, we cooperated with KOMATSU and visited their factory, workshop and actual working environment, and collected some valuable information.
From a conversation with a KOMATSU technician, we learned that new technicians enter the workshop and first shadow a senior technician for about 6 months to observe and learn to mimic how they diagnose and repair machines. Thereafter, they receive more systematic training to understand each part of the machinery and begin to try to take over some repair projects on their own.
When they encounter a problem that they cannot solve, they usually call a more experienced senior technician for help.
Even though all the knowledge they need is stored on the company's platform, they rarely sit down at the computer and study what kind of problem they need to solve due to the low level of accessibility and visualization of the information and the highly hands-on nature of the mechanic's profession.
And after all those conversations and shadowing research, we conclude their current problem into 4 main topics:
1. Accessibility of platforms
2. finding the right information
3. input obstacles
4. tons of experience locked in their head
Therefore, we want to enhance mechanists' working experience in 3 scopes according to time and relevance.
Based on the above discussion, we constructed a working platform centred on information visualization. The centrepiece of the platform is a digital twin of a logging machine.
With simple clicks and swipes, a mechanic can zoom, rotate, and highlight parts on this machine model to view the part information he or she needs. We've also integrated communication and screen mirroring into the platform so that when the mechanic needs to ask a senior technician for help, he can stop struggling to verbalize exactly what's wrong with the machine. Instead, he can choose to share the screen and directly highlight the part he is trying to describe.
Each mechanic's original workflow included this: when they completed each repair task, they needed to write a short evaluation of that task, even if it was only completed or needed to be postponed.
In this session, we introduced an AI with voice recognition to simplify the technician's work. Previously, the technicians needed to type in their comments whenever they finished a task, and because of the extreme weather conditions, they tended to do this project when they returned to the workshop or when they got home from a long day of work. Now all they need is a simple click and voice input (although they will also be asked to make more specific comments), and the AI will help them through the process, while the comments recognized and analyzed by the AI will be uploaded to the details of the part in question.
This way, when a mechanic encounters a problem in the repair process, he or she can first expand the specific information of the part to see how their colleagues dealt with similar problems.As a result, I believe that this project has realized the open-source of knowledge and the improvement of efficiency.
We also conclude our design into three key values and list all the function modules which are related:
1. improvements - before and after
2. future concern
3.scalability