AI-Based Service Desk

Project description

How might we help Vanderlande’s service desk respond to queries from the (internal) customer, using available data (e.g. manuals, known bugs and issues, technical documentation etc.) so that less issues are escalated to the 2nd and 3rd line service support?
We are required to build a solution to this problem that makes use of AI in some way, shape or form.

Context

Vanderlande, a global logistics automation leader acquired by Toyota, is shifting toward a more software-driven approach, emphasizing AI, data integration, real-time reporting, and digital twins. Their vision includes remote collaboration and autonomous operations. Within this strategy, the internal service desk—organized into four support levels—is seeking innovation. Currently, 1st-line support creates issue queries, which are passed to 2nd-line staff who must manually search documentation to resolve even simple problems. Often, these are unnecessarily escalated to 3rd and 4th-line experts, including R&D developers. This results in time-consuming workflows and reduced productivity for higher-level support teams.

Results

The expected outcome of this project is a working prototype or proof of concept that demonstrates how a tool to assist the Service Desk would function. Along with this, handover documentation like an implementation plan, business advice and/or research report should be prepared alongside the prototype. These products will provide Vanderlande's innovation department with certainty that the core concept has potential to move the company closer towards its goal of shifting towards a software-driven approach.