July 8, 2019
Smart Technician is a project that originated from the Digital Impact Program, which a training program organized by the Digital Innovation Lab. In this training, one of the teams worked on a case for Pon Equipment and Pon Power, called ‘Remote Assistance’. This case revolved around a number of challenges in the service departments of PEPP, such as a declining workforce, low first-time fix and high parts return.
One of the key findings within this case, was that each technician has their own learning curve, leading to large differences in skill levels between junior technicians and experienced master technicians. The technical knowledge that is in the heads of technicians is often not shared sufficiently, and is lost completely when technicians retire.
The aim of Smart Technician is to capture this knowledge and accelerate the learning curve of technicians using Artificial Intelligence (AI). The idea behind Smart Technician is that if all repair jobs are stored and consolidated into one single knowledge database, we can use AI to learn which type of problems usually require which type of solutions. Technicians could then use this tool to get advice about the cause, the solution and the required parts based on their problem description.
To investigate the viability of this idea, the Datalab started a Proof of Concept together with Ben van der Werff and Raymond Elders from Pon Equipment. We gathered all repair history of Pon Equipment and built a simple algorithm that compares new problems to all problems in history, and gives the most common solutions and parts. The next step is to also perform this exercise for Pon Power, and then determine if we want to build an operational MVP.
Kevin HaverLead Data Scientist
Kasper MeijjerSenior Data Scientist